Criminal Law and Philosophy

, Volume 7, Issue 2, pp 231–253 | Cite as

The Revolution and the Criminal Law

Original Paper


Egyptians had many reasons to overthrow the government of Hosni Mubarak, and to challenge the legitimacy of the interim military government. Strikingly, among the leading reasons for the uprising and for continued protest are reasons grounded in criminal justice. Reflection on this dimension of the Egyptian uprising invites a broader examination of the relationship between criminal justice and political legitimacy. While criminal justice is neither necessary nor sufficient for political legitimacy, criminal injustice substantially undermines political legitimacy and can provide independent reasons for revolution. A state may compromise its legitimacy by committing criminal acts, by perverting or subverting the criminal process, and by failing to discharge its duty to punish serious wrongdoing—a duty that then falls to individuals to discharge either directly (through vigilantism) or indirectly (through revolution). Contrary to the views of many leading criminal law theorists, the duty to punish serious wrongdoing applies to individuals and not only to states. The relevance of political legitimacy to criminal justice is more complicated. Individuals are morally obligated to follow the morally justified laws of an illegitimate state, but are not morally obligated to follow the morally unjustified laws of a legitimate state. Nor may any state punish in the absence of moral wrongdoing and moral fault. However, illegitimate states may be incapable of justly holding individuals accountable to the state, to the community, or to victims through criminal trials. This incapacity provides an additional reason to overthrow illegitimate states and replace them with legitimate states capable of justly administering a just criminal law.


Justice Legitimacy Authority Obligation Revolution Vigilantism Democracy Detention Retributivism Punishment 

On June 10, 2011, I walked through Cairo’s Midan Tahrir (Independence Square) for the first time. Less than 4 months earlier, millions of Egyptians gathered there and in city centers across the country to bring down the authoritarian government of then-President Hosni Mubarak. As I made my own way across the square, a few families and groups of friends lounged on the central rotunda; street vendors sold t-shirts celebrating the uprising; and a few dozen activists stood around chatting, waiting for a small protest (about conditions in Cairo’s many unplanned housing developments) to begin. There were few police officers in or around the square, and indeed across the city the police seemed limited to directing traffic (to the extent possible), protecting the homes of the wealthy, and guarding tourist attractions. I crossed the square many times in the following weeks, on my way to meet with professors and students, artists and doctors, eager to hear their thoughts on the pressing issues of the day: the drafting of a new constitution; the timing of parliamentary and presidential elections; the formation of new political parties; and the role of religion in a democratic Egypt. Around their dinner table, a taxi driver and his children argued that the billions of dollars Mubarak and others had embezzled should be redistributed to the people on a per capita basis or, failing that, invested in the country’s schools. Politics dominated virtually every conversation, and I often found myself translating what I heard into the language of political theory. Constitutional essentials. The basic structure of society. Overlapping consensus. Public reason. Primary goods. Fair equality of opportunity.

There was, however, another recurring topic of conversation. Approximately eight hundred protesters had been killed by state security forces during the original uprising. The interim government led by the Supreme Council of the Armed Forces (SCAF) promised to criminally prosecute the officials responsible for the killings. But indictments were few and far between, and scheduled trials were repeatedly postponed. Friends and family members of those killed, many political activists themselves, grew impatient. At a theater in downtown Cairo, a new play, ‘No Time for Art’, began by asking each audience member to rise and read aloud from a flyer. Each flyer called on the interim government to investigate the killing of one of the fallen. Every member of the audience read at least one name. Some added the names of their own loved ones. Some cried.

Before I left Egypt my new friends told me to watch for a major, nation-wide protest on July 8. Sure enough, that day saw the largest protests since the original uprising. Around seventy thousand Egyptians took to the streets. But their demands were not constitutional or socio-economic in nature. They did not demand that a new constitution precede new elections, or that a newly elected parliament draft a new constitution. They did not call for elections to be postponed so that new political parties would have more time to organize, or to hold elections as scheduled so that civilian rule would be swiftly established. Nor did they call for any specific package of social or economic reforms. On that particular day, with all their numbers, they demanded only that the long-promised criminal prosecutions finally take place. Organizers called the day the ‘Friday of Accountability’ (Associated Press 2011). In Cairo, protesters held aloft a white sheet on which they had written ‘Retribution [for] the killers of martyrs’. In Alexandria, a banner overlooking the Mediterranean read simply ‘Retribution’.

Justice and Legitimacy

Egyptians had many reasons to overthrow the Mubarak regime and to challenge the legitimacy of the interim government. It is nevertheless striking that both the original uprising and many later protests were motivated in part by reasons grounded in criminal justice. Reflection on this dimension of the Egyptian experience invites a broader examination of the relationship between criminal justice and political legitimacy.

I take criminal justice to involve the moral norms governing criminal law, criminal procedure, and criminal punishment. I take political legitimacy to involve the morally justified legal power of a political authority to create legal rights and impose legal duties within a given society. A political authority is morally justified in exercising its legal power if it is morally better for that authority to exercise its legal power than for no one to exercise that power (anarchy) or for that power to pass to some other individual or group (through abdication, succession, or revolution). It seems to follow that an unjust authority may retain political legitimacy if the dissolution or transfer of power would result in greater injustice still. Indeed, it seems to follow that political authorities may maintain political legitimacy through unjust acts. This may have been the strategy of the Mubarak regime: by killing, imprisoning, and exiling potential rivals the regime sought to ensure that no credible opposition arose; by violently repressing any organized opposition it sought to make revolution impossible or morally worse than life under the regime; and by sponsoring criminals or armed loyalists it sought to ensure that the fall of the regime would bring chaos in its wake. However, it is more accurate to say that unjust authorities have a categorical obligation to cease their unjust acts, allow the formation of morally preferable alternatives, and then surrender their legal power; it is only on the condition that unjust authorities violate this categorical obligation that they are morally justified in exercising their legal power. I will refer to unjust authorities, including those that seek to preserve their legitimacy through unjust acts, as at best conditionally legitimate.1

Can a politically illegitimate or conditionally legitimate authority justly enact criminal legislation, defining offences and defences; search homes, businesses, and places of worship; arrest, detain, interrogate, jail, and bring to trial those suspected of crime; and condemn and punish those convicted? Conversely, can the failure to enact just laws, employ just procedures, and impose just punishments deprive an authority of its political legitimacy? These will be my primary topics, though my discussion will touch on several others, particularly the extent to which criminal justice is part of political morality or ordinary morality.

I shall argue that, while criminal justice is neither necessary nor sufficient for political legitimacy, criminal injustice substantially undermines political legitimacy and can provide independent reasons for revolution. A state may compromise its legitimacy by committing criminal acts, by perverting or subverting the criminal process, and by failing to discharge its duty to punish serious wrongdoing—a duty that then falls to individuals to discharge either directly (through vigilantism) or indirectly (through revolution). Contrary to the views of many leading criminal law scholars—including Antony Duff, Malcolm Thorburn, and Guyora Binder—the duty to punish serious wrongdoing applies to individuals and not only to states.

The relevance of political legitimacy to criminal justice is more complicated. Individuals are morally obligated to follow the morally justified laws of an illegitimate state, but are not morally obligated to follow the morally unjustified laws of a legitimate state. Nor may any state punish in the absence of moral wrongdoing and moral fault. However, illegitimate states may be incapable of justly holding individuals accountable to the state, to the community, or to victims through criminal trials. This incapacity provides an additional reason to overthrow illegitimate states and replace them with legitimate states capable of justly administering a just criminal law. Or so I shall argue.

Crimes of State

Criminal justice and political legitimacy might relate to one another in one or more ways. Political legitimacy might make a necessary, sufficient, or otherwise significant contribution to the justification of criminal law, criminal procedure, or criminal punishment. Or an otherwise justified system of criminal law, criminal procedure, or criminal punishment might make a necessary, sufficient, or otherwise significant contribution to political legitimacy.

In principle, criminal justice is neither necessary nor sufficient for political legitimacy. If there is no credible opposition, if the regime uses violence to maintain power, or if the fall of the regime would release violent forces that an opposition government could not contain, then even a seriously unjust regime could retain at least conditional legitimacy. Conversely, if an orderly transition of political power is possible, and if a more benevolent, equitable, humane, or representative opposition is prepared to receive it, then a regime with an otherwise just system of criminal laws, procedures, and punishments would still lose its legitimacy.

In practice, however, criminal injustice severely undermines political legitimacy because such injustice makes it easier for opposition forces to offer greater justice than the existing regime and weighs heavily against the morally relevant costs of political transition, including revolution. Indeed, the Egyptian uprising was in substantial part a response to years of murders, tortures, kidnappings, and assaults perpetrated by the police and security forces against ordinary citizens. Each case involved multiple violations of criminal justice: the most serious crimes, committed by those charged with enforcing the criminal law, often in the course of criminal investigations, went almost always unprosecuted and unpunished.2 Despite its disunity and unpredictability, the Egyptian opposition still seemed preferable to the certain repetition of regime criminality. Moreover, when a population is threatened with routine criminal violence at the hands of the state, the risks and costs of revolution become dramatically easier to justify. These risks and costs were grave: eight hundred Egyptians were killed in the uprising, and violent crime increased (albeit from a very low baseline) following the retreat of police and security forces from the streets. But faced with a regime that, to a substantial extent, governed through crime, one can understand why many Egyptians thought these sacrifices worth making and these consequences worth bearing.

It should come as no surprise that the enforcement of criminal law, and the enforcers of criminal law, can make or break an authority’s political legitimacy. For it is generally the police, rather than the armed forces, that are authorized to use force against ordinary citizens. In the best cases, this authority is used to invade the privacy and restrict the liberty of criminal suspects. In the worst cases, this authority is abused to blackmail, extort, and torture not only criminal suspects but also personal and political enemies. The proximity of police power to political legitimacy is one reason why, during the original uprising, Egypt’s police and state security forces were viewed as enemies of the people while the army and the people were declared to form ‘one hand’. As of this writing the armed forces police the streets, and the army’s increasingly brutal treatment of protesters has prompted mass demonstrations in a growing number of cities (Kirkpatrick 2011). When the police power is abused, or used in the service of unjust laws, the state can quickly become a greater threat than ordinary crime to ordinary citizens.

State criminality may be committed both through the abuse of the criminal process and through the abrogation of the criminal process in favor of more expedient measures of social control. For example, Egypt’s 1959 Emergency Law permits the president, upon declaring a state of emergency, to ‘[r]estrict people’s freedom of assembly, movement, residence, or passage in specific times and places; arrest suspects or [persons who are] dangerous to public security and order [and] detain them; allow searches of persons and places without being restricted by the provisions of the Criminal Procedure Code; and assign anyone to perform any of these tasks’ (Law 162/1958, Qanun bi Sha’n Halah al-Tawari’ [Law Concerning the State of Emergency], as amended, art. 3). The abrogation of the Criminal Procedure Code, which forbids warrantless searches and seizures and limits pre-trial detention, left Egyptians even more vulnerable to mass arrests, indefinite detention without trial, and torture. In addition, a system of State Security Courts was created to hear violations of emergency orders as well as criminal cases concerning terrorism, public incitement (including by the press), and public demonstrations and gatherings. The State Security Courts provided the Mubarak regime with a legal weapon with which to silence its critics, suppress opposition groups including the Muslim Brotherhood, and persecute sexual minorities (Reza 2007).

After the uprising, the interim government controlled by the SCAF detained, prosecuted, or imprisoned between seven and ten thousand Egyptians on charges ‘including “thuggery,” assault, and threatening the security of the Egyptian state—a catch-all phrase once employed by Mubarak’s despised ancien regime’ (Hammer 2011).

Those accused include pro-democracy demonstrators, bloggers, and other prominent activists swept up in the chaos that preceded and followed Mubarak’s fall, as well as common criminals and bystanders. Thousands have been convicted and sentenced to terms of between several months and five years in prison. The procedures tend to be swift and are conducted before single judges in military uniform who are not known for scrupulous attention to the evidence. (Hammer 2011)

The original uprising could not have succeeded without the acquiescence of the armed forces. But the crimes of the interim government, the impunity enjoyed by its agents, and the use and abuse of criminal process clearly accelerated the interim government’s own crisis of legitimacy.

Legitimacy and Impunity

The Mubarak regime lost its legitimacy in large part by abusing the criminal law and displacing it with emergency powers. By contrast, Egypt’s interim government risked its legitimacy in part by failing to use the criminal law against members of the Mubarak regime. Some elements of the interim government were complicit in some of the very crimes for which Mubarak and others face prosecution, and this fact no doubt casts the delays of prosecutions in a particularly suspicious light.

In some cases, particularly cases of transitional justice, a regime may decline to prosecute and punish certain crimes for which it is not responsible; which it did not commit, facilitate, or encourage; and which may have occurred prior to the regime’s rise to power. In some such cases, prosecution and punishment of these crimes might reignite the armed conflict or communal violence that led to their commission; drain the state of resources necessary to fulfill other demands of justice (particularly socio-economic justice); or overwhelm the legal system’s capacity to deal with new crimes. In such circumstances, a regime might be justified in failing to deliver criminal justice, and political legitimacy generally is not undermined by justified conduct. But there is little evidence that such justifying circumstances obtain in the Egyptian case.

Some Egyptians responded to the delayed prosecutions by calling for individual members of the interim government to step down, thereby challenging the political legitimacy of their continued rule. Others, however, were more concerned that continued demonstrations challenging the legitimacy of the interim government will damage the economy and further disrupt ordinary life. How were they, and how are we, to compare the positive requirements of criminal justice with the costs and risks of political upheaval? Is it even coherent to overthrow a regime for failing to enforce its own laws? It is to these questions that we now turn.

Revolution and Vigilantism

There is a view in contemporary criminal law theory according to which the norms of criminal justice belong to political morality but not to ordinary morality and apply to the state but not to ordinary citizens. This view surfaces routinely in discussions of state punishment and vigilantism. In various influential writings, Antony Duff, Guyora Binder, and Malcolm Thorburn argue that any attempt to justify legal punishment by reference to general moral principles applicable to individuals would equally justify vigilantism and therefore reduces to absurdity. For example, Duff writes that

if the demand [for punishment] is a moral demand of justice, then anyone could in principle satisfy it by imposing the right quantum of suffering on the guilty, so that even private enterprise “punishment” inflicted by a vigilante group could constitute just punishment. The state’s claim to exclusive authority to punish breaches of the criminal law could then be justified only as a part of a wider justification of the state’s monopoly over the use or authorization of coercive force. (Duff 2011)

Duff clearly thinks this picture of criminal justice is implausible and concludes instead that ‘criminal punishment is an aspect of the state, and is thus a political enterprise’ (Duff 2011, p. 6). Only the state has the political standing to punish; if the state fails to punish then private individuals cannot stand in its place to correct its omission. Later, Duff adds the following:

It is not enough to say, for instance, that as moral agents we have a proper interest in seeing that retributive justice is done and the standing to ensure that it is done, for criminal punishment is a dimension not of our personal, informal moral lives, but of our political lives as members of a state; punisher and punishee must be related by the existence of some appropriate political structure that gives them these roles. (Duff 2011, p. 7)

In these passages Duff makes three central claims: that ‘punishment is a relational activity (its simple manifestation is not A being punished, but A being punished by B)’; that the relation between punisher and punished is a political relation; and that this political relation arises from and exists within the political structure of the state.3
Similarly, Thorburn writes that

[i]f… the criminal justice system is simply a tool for bringing about retributive justice, then what reasons might we have to object if private citizens were to take it upon themselves to hold trials and to punish offenders? It seems that there is no reason in principle why we should object: so long as the vigilante ‘gets it right’ and punishes only those who have committed moral wrongs (and does so for that reason and in proportion to the offender’s desert), it seems that [retributivists] would have no principled reason to object to this practice. (Thorburn 2011, p. 10)

[H]owever, the law in virtually every common law jurisdiction is deeply hostile to vigilantism. It is not just that the balance of reasons usually favours a state monopoly of legitimate violence; rather, it appears to be a central commitment of the legal system to maintain this state monopoly. (Thorburn 2011, p. 11)

Thorburn argues, instead, that the state is justified in punishing individuals in order to protect and preserve the equal distribution of freedom established by its laws. On this view, vigilantes arrogate to themselves an unequal freedom to punish other citizens.
Finally, Binder writes that

[i]f retributive punishment rests on the obligations of individuals to punish all and only those who deserve it, then it would seem to condone some instances of lynching… .

Nevertheless, I think most of us would say that this kind of gratuitous vigilante justice would be very wrong, an act of murder. But why? Is it morally wrong? If so, how can it be morally right for state officials to execute the criminal based on the same considerations? I think the answer is that vigilante justice is not morally wrong, and that legally authorized punishment of the guilty is not morally right. The wrong of vigilante justice is a political wrong and the right to punish conferred by law is a political right…. The problem with our hypothetical lynching is not with the consequence that the murderer gets punished, but with the institution doing the punishing. The defect is not of morality, but of legitimacy. (Binder 2002, 327–328).

Binder proceeds to argue that both Bentham and Kant offered only justifications of state punishment that do not apply to individuals (Binder 2002, pp. 348–349).

One problem with the statist or institutionalist view reflected in the quotations above is that such a view precludes not only vigilantism but also revolution for the sake of criminal justice. If principles of criminal justice arise only from, exist only within, and apply only to states then it would be irrational, in at least two ways, for ordinary people to overthrow a state in order to achieve criminal justice. First, on the statist view, reasons of criminal justice apply to states but not ordinary citizens. Citizens have no reason to pursue criminal justice, and by overthrowing the state revolutionaries would destroy the only agent with reasons to pursue criminal justice. More importantly, revolution would destroy the political structure from which principles of criminal justice arise and within which they exist. Revolution for the sake of criminal justice would prove self-defeating since success would cancel or annul the very principles on which the revolution is based.

Revolution for the sake of criminal justice is rational only if reasons of criminal justice arise and exist independently of the state and directly apply to ordinary people. Of course, individuals ordinarily respond to such reasons by supporting the legal system of an existing just state. However, if a state systematically fails to secure criminal justice then individuals may have reasons of criminal justice to overthrow the existing state and create a new state committed to criminal justice. Since revolution for the sake of criminal justice appears rational, reasons of criminal justice must directly apply to ordinary people independently of the existence of the state.

To frame the issue more precisely, it may be helpful to deploy John Gardner’s distinction between reasons that are ‘personal in respect of conformity’ in the sense that they are reasons for particular people to do particular things and reasons that are ‘personal in respect of attention’ in the sense that other people have no reason to ensure that those particular people do those particular things (Gardner 2007, 131). For example, Gardner argues that the fact that I made a promise gives me a reason that is personal in respect of conformity because only I can keep my own promise; that fact also gives others a reason that is impersonal in respect of attention because (Gardner believes) everyone else should take reasonable steps to see to it that I keep my promise. I take Duff, Binder, and Thorburn to claim or assume that criminal justice generates reasons to which only the state can conform, but to which everyone (or at least all citizens) should attend. On this view, criminal justice provides citizens with reasons to insist that the state do what criminal justice requires of it, but not with reasons to do criminal justice themselves.

This view certainly succeeds in rendering vigilantism unjustifiable, but it has the side-effect of rendering revolution for the sake of criminal justice unjustifiable as well. For if the state should fall then it will necessarily fail to conform to the reasons to which only it can conform. If criminal justice provides reasons of this kind then the very idea of overthrowing the state for failing to provide criminal justice is self-defeating. A reason that is personal with respect to conformity cannot provide a reason to eliminate the only agent that could conform to that reason. On this view, the fact that our state failed to punish serious wrongdoing gives us no reason to overthrow and replace that state, any more than the fact that I broke my promise gives you a reason to prevent me from ever fulfilling my promise, promise to do what I promised to do, and then do that very thing.4 Even if you fulfill your promise you cannot fulfill my promise.

True, a successor state would have reasons to achieve criminal justice, but these would be new reasons summoned into existence upon the creation of the new state, not the old reasons on which the predecessor state failed to act and on which individuals were supposedly precluded from acting.5 Yet if criminal justice can provide reasons for revolution these must be reasons that exist prior to the revolution, on which the unjust state fails to act and on which individuals decide they must act. Revolutionaries might anticipate the reasons to which a future state must conform, but they cannot attend to that state’s conformity to those reasons before that state comes into existence. Reasons to attend are derivative of reasons to conform; the former cannot exist apart from the latter, and the latter cannot exist without the agent to whom they are addressed. The only reasons on which we can act are those to which we can conform as well as those to which others can conform and to whose conformity we can attend.

Indeed, if a revolution results in the total dissolution of an unjust state then there will be a period in which there will be no state to whose conformity to reasons anyone could attend. Revolutionaries may find themselves in an anomalous situation in which the reasons of criminal justice that seemed to justify revolting against the old state cease to exist upon its overthrow and no new reasons of criminal justice arise until the formation of a new state. So just as individuals have no reasons to overthrow a state for failing to prosecute and punish its own crimes they also have no reasons to create a new state in order to prosecute and punish the crimes of the prior regime. On the view we are considering, revolution for the sake of criminal justice appears unjustifiable as a matter of logic, quite apart from matters of fact.

If only the state can fulfill the requirements of criminal justice then criminal justice may give ordinary citizens strong reason to insist that the state fulfills these requirements, but criminal justice can give citizens no reason to fulfill these requirements themselves, either directly (through vigilantism) or indirectly (through revolution and the creation of a new state). Citizens might have other reasons for overthrowing a state that does not provide criminal justice and for creating a new state that will provide criminal justice. But these will not be reasons of criminal justice, which by hypothesis are reasons for existing states to provide criminal justice and for citizens to see to it that existing states provide criminal justice. For example, if an existing state’s failure to pursue criminal justice leaves ordinary citizens vulnerable to ordinary criminals then those citizens will have both self-interested and altruistic reasons to overthrow that state and replace it with a new state that, as a result of pursuing criminal justice, will produce greater security for individual citizens. But even these citizens will have no reasons of criminal justice for overthrowing and replacing the existing state, for criminal justice asks of them only to insist that existing states conform to reasons of criminal justice.

By contrast, both vigilantism and revolution assume that criminal justice gives individuals reason to see to it that justice is done even if it is not done by an existing state. In other words, both vigilantism and revolution assume that reasons of criminal justice are impersonal both in respect of conformity and in respect of attention. To that extent, the justifiability of vigilantism and the justifiability of revolution stand or fall together. Duff, Binder, and Thorburn argue that only the state can deliver criminal justice because if individuals can deliver criminal justice, either by themselves or by means of the state, then the reasons that justify punishment could also justify vigilantism, a conclusion they regard as absurd. I have tried to show that their position leads to a far greater absurdity. The moral intelligibility of revolution for the sake of criminal justice shows that criminal justice generates reasons to which individuals can both attend and conform.6 If states are to cancel or outweigh those reasons then the state must itself deliver enough criminal justice to make both revolution and vigilantism not worth their risks and costs. The state’s monopoly on the use of force cannot be merely asserted but must be earned.

Before moving on, let us consider how the statists might respond to the challenge posed above. Duff might respond by arguing that reasons of criminal justice apply not to states but to political communities who in turn authorize states to act on the community’s behalf. On this view, a political community might replace an existing state with a new state in order to better (indirectly) discharge its duties of criminal justice. This view, more communitarian than statist, seems like an improvement but runs into a parallel difficulty. For it seems entirely rational for individuals to replace an existing political community for the sake of criminal justice, for example by introducing a new political culture and mode of political engagement, and this suggests that reasons of criminal justice apply directly to individuals.

Finally, statists might argue that reasons of criminal justice are not personal in respect of conformity to the state that exists at the time a crime is committed, but rather to any state with an appropriate relationship with the crime. They might then argue that a post-revolutionary state has such an appropriate relationship with crimes committed under the pre-revolutionary state. This view would still reject vigilantism, since only states can conform to reasons of criminal justice, but might allow for the rationality of revolution for the sake of criminal justice. Clearly, such a view requires a persuasive account of the relevant relationship between a state and a crime that explains both why states but not individuals have duties of criminal justice and also why post-revolutionary states enjoy such a relationship with pre-revolutionary crimes. I am skeptical that such a persuasive account is possible, but we will have to wait and see.

Criminal Justice and the State

None of this is to say that vigilantism is preferable to legal process, or that permanent revolution is preferable to social peace. Evidently, delivering justice in all its forms requires a level of coordination and cooperation that the state can enable but which individuals acting without legal processes and political structures would struggle to achieve. States are far better equipped than either individuals or social groups to provide rational fact-finding, principled sentencing, and adjudicative finality. Particularly when criminal and victim are members of different social groups, imperfect information and in-group bias can give rise to false accusations, disproportionate punishments, and spiraling cycles of revenge (Haque 2005). On some accounts, it was to escape these pathologies that criminal law was established in the first place.7 Certainly, Egyptians generally demanded criminal prosecution of those responsible for killing and torturing protesters, although there were some private reprisals against police and security officials as well.

Moreover, when the state delivers justice it thereby relieves individuals of the moral burden of trying to do so by themselves, bearing the attendant costs and taking the attendant risks. Just as the modern state acts as a conduit for individual duties of humanitarian assistance and distributive justice, thereby freeing individuals to concentrate their attention on their own lives and their own flourishing, so too should the state relieve individuals of their duties to pursue criminal justice. This, I believe, is the best reply to those who accuse the criminal law of ‘stealing’ private conflicts from victims and their families (Christie 1977). For if victims and their families are left to seek justice themselves then they cannot concentrate on healing their injuries, mourning their losses, and moving past their suffering and on with their lives. Private pursuit of justice generally requires one to sustain one’s anger, outrage, and resentment, since such emotions provide the strong motivation necessary to willingly incur future costs to punish past wrongdoing. Yet such reactive attitudes can be psychologically debilitating, for they require that victims and their families continue to live in the past, constantly revisiting their loss to refresh or reinforce their desire for justice. States do not bear such psychological burdens and this partly explains why it is normally preferable for the state to take responsibility for delivering criminal justice.8

Even so, it is worth pausing to remember that criminal law long predates the modern state, not to mention liberalism and democracy. Under many feudal and imperial arrangements, criminal law derived as much from customary and religious norms as from directives promulgated in the name of the ruler. The law managed and organized but did not yet seek to displace these earlier forms of social organization and preexisting sources of public values. Crimes against the ruler were prosecuted by agents of the ruler, but crimes against persons and property were largely prosecuted by victims and their families. Legal process harnessed, but did not seek to displace, the moral reasons of individuals. We have already noted two benefits of the state’s gradual takeover of the criminal process: it avoids the instability of group vengeance and relieves individuals of the burdens of seeking justice. In addition, the stable and cohesive moral communities that made it possible to base large parts of the criminal law on custom and religion have changed irreversibly in most parts of the world. This transformation also carries substantial benefits, since it allows a level of group diversity and individual freedom that could not exist under a criminal law that sought to fully implement community and religious values.

In recent years, Antony Duff has sought to explain how criminal justice should be realized in a modern, liberal, democratic state. This is a vitally important inquiry, but one that we should take care to frame correctly. For example, Duff writes that

We must remember that criminal punishment is an aspect of the state, and is thus a political enterprise. If we are to understand its meaning and rationale, we must ask what proper state purposes it could serve and how the state could be warranted in imposing such burdens on its citizens; to answer such questions, we must clearly engage in political theory. (Duff 2011, p. 6)

I have already observed that criminal punishment is now an aspect of the state but has been an aspect of earlier forms of social organization and that the role of the state in criminal punishment (adjudicator, executioner, prosecutor) has changed over time. Ideally, we should seek a theory of criminal law that recognizes criminal law as an evolving social enterprise extended over space and time and that contributes to a critical understanding of criminal law’s evolution as well as its current shape and trajectory.

From an historical perspective, the relationship between the state and its citizens should not be the starting point of criminal theory but rather the mid-point. The question we should ask is how the rise of the modern state, as well as its ongoing transformations, should affect the conceptual structure and substantive content of criminal justice. For example, the criminal law of a premodern, tribal society might criminalize adultery, declaring it a public wrong that implicates the core values of the community. If there is no police force to investigate, rules of evidence that make proof all but impossible, and harsh penalties for false accusation, then the prohibition may send its intended moral message even if the crime is almost never punished. But if the same law is adopted by a modern state, with the power to monitor its citizens’ movements and communications, prove guilt through circumstantial evidence or coerced confessions, and cultivate informants, then the law gives the state another tool with which to control its citizens, blackmail them, deprive them of their dignity, their liberty, or even their lives. Even a deeply conservative society that views adultery as a violation of its basic values should remember that its basic values also may be threatened by a modern state empowered to police private conduct. In particular, conservative political parties such as the Muslim Brotherhood would be foolish to expand the power of the state in a misguided moral crusade.

From a normative perspective, a liberal theory of criminal law should tell us not what kind of criminal law a liberal state should create but rather how a liberal state should adapt, modify, and curtail the criminal law it inherits. As many scholars have observed, liberal theory generally does not concern itself with questions of moral wrongdoing, fault, responsibility, blame, or desert. However, as Thaddeus Metz has argued, liberal theory is nonetheless fully compatible with a criminal law structured by such moral concepts because a liberal state may act for the sake of individual rights either by protecting these rights or by expressing respect for these rights (Metz 2007). As I have argued elsewhere, the core of retributive justice is best understood in terms of vindicating the rights of victims (Haque 2005). In any event, our theory of punishment need not derive from our theory of the state; it is enough that the two peacefully coexist.

A Duty to Punish

When is it morally justifiable to overthrow an unjust state, despite the risks and costs, for the sake of criminal justice? So far we have seen that individuals have reasons to pursue criminal justice by supporting the legal system of a just state (generally), by overthrowing and replacing unjust states (rarely), or by taking justice into their own hands (hardly ever). The next step is to get some sense of how much force or weight such reasons carry, and how their force or weight compares with the costs and risks associated with revolution.

It would seem that the reasons for revolution in response to the state’s failure to prosecute or punish criminals are only as strong as the reasons to prosecute and punish criminals in the first place. On one view, the reasons to prosecute and punish are ordinary reasons that compete with opposing reasons on an even playing field. On another view, there is a duty to prosecute or punish. Duties, in turn, may be understood as protected reasons that not only favor an action but also exclude some opposing reasons from competition (unless, perhaps, those opposing reasons cross some threshold of cumulative weight). The strength of a duty, on this view, is a function of the strength of the reason protected and the exclusionary scope of the reason doing the protecting. If the first view is true then a state that fails to prosecute or punish contrary to the balance of reasons acts unjustifiably. If the second view is true then a state that fails to prosecute or punish contrary to the balance of non-excluded reasons acts not only unjustifiably but unjustly as well. Other things equal, we have stronger reason both to avoid and to prevent an act if it is both unjust and unjustifiable than if it is unjustifiable but not unjust, since if the act is unjust then at least some opposing reasons are excluded from counting against it. An unjustifiable act is made still more unjustifiable if it is also unjust. So if there is a duty to prosecute and punish then the case for revolution in the name of criminal justice becomes considerably stronger.

Whether there is a duty to prosecute and punish criminals, or merely ordinary reasons to do so, is of course the subject of considerable debate, not least between retributivists and consequentialists.9 Of course, some retributivists assert only that past wrongdoing provides a good reason to punish wrongdoers, a modest position that some scholars (Berman 2011; Cahill 2007) suggest collapses into a suitably ecumenical consequentialism. But real retributivists are positive deontic retributivists who believe that past wrongdoing generates a duty to punish wrongdoers (Haque 2005). The duty to punish can, of course, be outweighed if the bad consequences of punishment exceed some threshold of gravity. Opposing reasons can only be excluded up to a point, and when they reach that point they must be confronted directly. But a duty that contains a threshold is still not so easily lost amidst the crowd of ordinary (often consequentialist) reasons.

If duties are constituted by protected reasons then it should be possible to determine if a reason is also a duty by asking whether it excludes at least some opposing reasons at least up to a point. For example, the fact that criminal justice carries administrative and financial costs does not ordinarily count against it, and must simply be accepted as the price of doing justice. But if the administrative costs of the criminal justice system threaten to bankrupt the state and deprive it of the funds needed to care for the old, educate the young, house the poor, heal the sick, and defend the nation, then it would be justifiable for the state to breach its duty and let some wrongdoers go unpunished or punish them less than they deserve. Similarly, the fact that deserved punishment causes wrongdoers to suffer ordinarily provides no reason for amnesty or abolition. But if the suffering that deserved punishment causes wrongdoers is torturous, cruel, inhuman, or degrading (as it often is, due to prison violence, inadequate medical care, and overcrowding) then this suffering indeed counts against punishing at all. The duty to punish is not absolute—it can be cancelled or overridden—but it is there all the same.

Earlier it was argued that reasons to punish are not personal to the state. If the state fails to act on these reasons then we must, unless our acting on these reasons would be contrary to the balance of reasons that apply to us. If the state’s reasons to punish constitute a duty to punish, and if the state fails to discharge that duty, then that duty falls to us to discharge either directly (through vigilantism) or indirectly (through revolution) unless doing so would be contrary to the balance of non-excluded reasons that apply to us.

From what has already been said, we can see that it is not desirable for the duty to punish to pass from states to individuals and social groups. The moral burden that would fall on individuals, particularly victims and their families, would be considerable. In addition, the transfer of responsibility will often prove extremely volatile, as families, tribes, clans, as well as ethnic and sectarian groups all compete to take up what the state has laid down. Indeed, it is the state’s capacity to relieve these burdens and end this competition that provides the strongest justification for the state’s control over the criminal process.

We can also see why the failure to punish is so destructive of political legitimacy. If the state fails to discharge the duty to punish then it loses one of the chief rationales for its very existence. If the state cannot displace individual retaliation and supplant group vengeance, and if these forces cannot be allowed to operate freely, then the case for overthrowing the existing state and forming a new one is very strong indeed.

Legitimacy and Fault

While criminal justice is neither necessary nor sufficient for political legitimacy, criminal injustice substantially undermines political legitimacy. The Mubarak regime compromised its legitimacy by committing criminal acts, abusing and displacing the criminal process, and by failing to discharge its duty to punish wrongdoing—a duty that then fell to ordinary Egyptians. The strength of this duty helps justify the decision to overthrow the Mubarak regime despite the risks and costs of revolution.

The impact of political legitimacy on criminal justice is more complicated. I shall argue that political legitimacy is not very relevant to the justification of criminalization, but may be highly relevant to the justification of criminal adjudication. While Egyptians have a moral obligation to obey the morally justified laws of an illegitimate authority, an illegitimate authority may be incapable of justly administering even a just criminal law. As I shall argue, this incapacity provides yet another reason to overthrow illegitimate authorities for the sake of criminal justice. However, I begin, perhaps unexpectedly, by discussing the relevance of political legitimacy to criminal defences. By focusing on this somewhat neglected dimension of the broader problem we shall gain a fresh perspective on the whole.

To frame the issue, consider George Fletcher’s position that moral fault should be legally required for criminal conviction only if a true political theory demands such a requirement:

In short, the political precedes the moral. It is only when a political theory makes reference to a moral question that the latter can become relevant in the criminal law. This thesis is grounded in the simple fact that the criminal law addresses the state’s authority to intervene in people’s lives. That authority must first be justified as a matter of political theory before one turns to the criteria, including perceptions of morality, that might enter into the use of the state’s power. (Fletcher 2007, p. 154)

As we have seen, the opposite is closer to the truth. The moral obligation to pursue criminal justice applies to us all. We need some political theory to explain when we should allow the state to discharge this duty on our behalf (generally), when we should abolish an existing state and create a new state that will better discharge this moral duty (rarely), and when we should take matters into our own hands (hardly ever).

Fletcher’s claim that the political precedes the moral is undermined by his statements that ‘in contemporary writing on political theory… neither criminal law nor criminal procedure receives much attention’ (Fletcher 2007, p. 151) and that ‘[p]unishment is not justly imposed unless the wrongdoer is also culpable or blameworthy for the wrongdoing’ (Fletcher 2007, p. 264). These statements are difficult to reconcile: political theory tends to ignore criminal law; only a political theory can make moral questions relevant to criminal law; yet we already know what the relevant moral questions are (is the wrongdoer ‘culpable or blameworthy for the wrongdoing’) and how they should be answered (by reference to ‘the requirements of intention, recklessness, or negligence as well as the recognition of excuses that negate personal culpability, including insanity, personal necessity, duress, and mistake of law’ Fletcher 2007, p. 264). Fletcher examines four political theories—liberalism, libertarianism, communitarianism, and loyalty—without deriving from them any particular criteria for ‘moral selection’ (Fletcher’s phrase for liability to punishment). Yet before we know what justifies ‘the state’s authority to intervene in people’s lives’ as a matter of political theory, we already know at least some of ‘the criteria, including perceptions of morality, that might [indeed, must as a matter of justice] enter into the use of the state’s power.’ In other words, our principles of moral selection are not derived from our political theory but rather constrain our selection of political theories. So there is at least one important sense in which the moral precedes the political and considerations of justice precede considerations of legitimacy.

There have been some scholars who have more consistently maintained that criminal defences should be shaped by political morality rather than ordinary morality. For example, H.L.A. Hart famously argued that the mental elements of crimes as well as the leading excuses (such as duress) and denials of responsibility (such as insanity) are necessary not in order to protect the morally blameless from undeserved punishment but rather to limit the coercive power of the state by making its application to any particular individual dependent, to the greatest extent possible, on the choice of that individual to break the law. Liability to punishment should therefore turn not on an individual’s moral culpability, responsibility, or fault but rather on her capacity and fair opportunity to conform her conduct to the law (Hart 1968). Hart thereby answered the question of moral selection by reference to the political values of individual freedom and limited government rather than to the moral values of blameworthiness and desert.

Hart’s commitment to the values of the rule of law is admirable, but he was wrong to conclude that these values provide the best account of excuses and denials of responsibility. No doubt, the state is under an affirmative obligation to make the law clear, public, simple, prospective, and so forth, and in these ways provide its subjects with a fair opportunity to conform to the law. Indeed, the rule of law makes an indispensible contribution to political legitimacy because the control over state violence that the rule of law provides each individual goes a long way toward ensuring that life under even an unjust state is morally preferable to the arbitrary violence of a life without law. Moreover, if the state fails to implement the rule of law it may be appropriate to bar the prosecution of a blameworthy defendant simply to prevent the state from exercising lawless power. The state may forfeit its right to punish by violating the rule of law. But a state that satisfies all the requirements of the rule of law must nevertheless acquit a defendant who lacks culpability, acts under duress, or is mentally ill. Hart’s account of such defences succeeds only if a defendant who lacks moral responsibility for breaking the law necessarily lacks the capacity to obey the law and if a defendant who breaks the law without moral fault necessarily lacks a fair opportunity to obey the law. If these criteria were to yield divergent results then it would remain unjust to punish a defendant who lacks moral responsibility or moral fault even if she could be said to have the capacity and fair opportunity to obey the law.

Even a legitimate state cannot deny criminal defendants the opportunity to claim self-defence, duress, or insanity without risking serious injustice. A legitimate state may be able to create wrongs, as we shall see, but it cannot create fault. Criminal justice has internal constraints that resist manipulation by states eager to increase their power.10 In principle, a state might permissibly limit the scope of certain defences or place a greater burden on defendants to prove that they satisfy the elements of these defences. But these are grave decisions that can only be justified if they will reduce undeserved acquittals much more than they will increase undeserved convictions. This standard of justification, in turn, reflects the fact that the duty to punish the morally guilty is strong but the duty not to punish the morally innocent is stronger.11

Legitimacy and Criminalization

Having established the priority of ordinary morality over political morality with respect to criminal defences, we now turn to the justice of criminalization by an illegitimate state. It is tempting to assume that an illegitimate or conditionally legitimate regime such as Mubarak’s cannot justly enact criminal legislation applicable to those under its effective control. However, this temptation should be resisted. Even if it is better for political authority to pass to another regime or to dissolve entirely, this does not entail that every exercise of illegitimate authority is also unjust. As Michael Philips observes,

[e]ven a military government imposed by conquest or an unpopular coup should pass and enforce laws against murder, rape, arson and so forth. It would be morally objectionable, e.g. were an occupying force to fail to protect those under its control in these ways. It follows that as long as a government holds power such a government has an obligation to punish in some cases. (Philips 1986, p. 414)

Philips is quick to add that a government that lacks legitimacy has only a conditional obligation to pass and enforce such laws. The government has a categorical obligation to surrender power to its morally superior political opponents; it is only on the condition that the government fails in its categorical obligation that it has the obligation to pass and enforce such laws (Philips 1986, pp. 414–415).12 What is significant about these laws is that they criminalize serious moral wrongs, specifically serious violations of individual rights. In the criminal law we would describe these crimes as mala in se, acts that are wrong apart from or independently of the law.

Surprisingly, Philips does not explore whether an illegitimate regime might also be conditionally obligated (or even just conditionally permitted) to pass and enforce laws criminalizing conduct that is not morally wrongful prior to or independent of the law. Presumably, individuals deserve to be punished for committing such mala prohibita offences only if they have a strong moral obligation not to commit those offences. As John Gardner argues, ‘obligation-imposing legal norms are sometimes morally justified, and when they are they create moral obligations as well’ (Gardner 2010, p. [12]). So, in principle, individuals may deserve to be punished for violating morally justified legal obligations.

Does the moral justification of a legal obligation turn on the political legitimacy of the regime imposing the legal obligation? In many cases it does not, since the moral justification for many legal obligations is that by conforming to them we will better avoid committing acts that are mala in se and that even an illegitimate regime has a conditional obligation to prevent. Many driving laws are morally justified on this basis, for by conforming to them we will better avoid accidentally killing and maiming other people. Statutory rape laws may be morally justified along similar lines. Now, it might be objected that many such laws could be drafted differently (a different speed limit, a different age of consent) and that a politically illegitimate regime has no right to impose its preferred rule upon its subjects. But this objection is mistaken. In general, an action or rule is justified if it is supported by undefeated reasons, even if the same reasons would also favor (or even more strongly favor) some alternative action or rule. So a legal obligation to drive below a certain speed does not lose its moral justification simply because an alternative speed limit would help us avoid accidents as well or even better (although a speed limit loses its moral justification if it causes more accidents than it prevents or otherwise does more harm than good).

Other examples of mala prohibita offences raise more complex issues. For example, it is a crime to fail to pay your taxes even if some or even most of your tax contribution will pay for goods and services that will benefit others but will not benefit you. Whether your legal obligation to pay such taxes is morally justified depends in part on which political theory, specifically which theory of distributive justice, is true. For example, your legal obligation to pay such taxes is more likely justified if some form of liberalism is true than if some form of libertarianism is true, acknowledging the diversity of views within each tradition. So individuals who fail to pay such taxes are more likely to deserve punishment if liberalism is true than if libertarianism is true. It follows that we cannot evaluate the justice of many criminal laws if we rely solely on criminal law theory and moral philosophy and ignore political theory.

Nevertheless, I would argue that although it is very important for criminal law theorists to read widely and deeply in political theory, not every criminal law theorist must become a political theorist or, for that matter, a criminologist, sociologist, psychologist, anthropologist, or neuroscientist. The moral justification of criminal legislation often turns on philosophical or empirical claims about which criminal theory as such has nothing to say. The primary task of the criminal law theorist is to explain how the moral justification of criminal laws would be affected if different philosophical or empirical claims were true. To actually evaluate these claims and deliver a final verdict on the criminal laws under scrutiny is admirable, in part because it is slightly perilous. We should expect every criminal law theorist to engage with political theory; but it seems too much expect every criminal law theorist to engage in political theory.

So while some mala prohibita offences are morally justified only if by avoiding them we will better avoid committing ordinary moral wrongs, others are morally justified only if they implement (even imperfectly) principles of distributive justice. Some may object that, since such principles may be implemented in a variety of ways, an illegitimate state has no right to force us to implement such principles in any particular way. But this invites the familiar response that a legal obligation may be morally justified even if an alternative legal obligation would also be morally justified. If a legal obligation implements principles of distributive justice then it remains morally justified even if an alternative legal obligation would do so as well or better.

It also might be objected that it is unjust for an illegitimate state to select one theory of distributive justice among others for enforcement through the criminal law. Of course it is unjust to enforce a false theory of distributive justice through the criminal law. But there is nothing inherently unjust about enforcing a true theory of distributive justice through the criminal law. Now, it may be unfair to enforce any theory of distributive justice through the criminal law if that theory was selected through an unfair procedure. And it seems safe to say that most illegitimate states use unfair procedures in most of their political decisions. But it also seems safe to say that some legitimate states use unfair procedures with some regularity (for example, in the United States each state, regardless of its population, is entitled to equal representation in the Senate). Conversely, it is not inconceivable for an illegitimate state or political entity to decide some issues using fair procedures (for example, the military government or occupying force described by Philips might allow some domestic or local issues to be resolved by an elected parliament or assembly). So there is no necessary connection between political illegitimacy and procedural unfairness.

It is nevertheless worth asking whether we have a moral obligation to follow morally justified but unfairly enacted legal obligations. I am inclined to think that we do. The law generally requires only that we do as the law demands, not that we do as the law demands because the law demands it. So if the law’s demands are morally justified then we should do what the law demands even if the law’s demanding it of us is unfair. Similarly, we should follow the morally justified laws of an illegitimate authority even if that authority has no claim to our loyalty or allegiance. We should follow such laws not because the authority is morally justified in demanding this of us but because not doing what the authority demands is morally unjustified. An illegitimate authority can therefore be the source of moral obligations even though it is not the ground of those moral obligations. It therefore seems that even if Egyptians were justified in overthrowing the Mubarak regime they remain morally obligated to obey any morally justified laws the regime enacted before it was overthrown.

Democratic Legitimacy

We have seen that individuals are morally obligated to obey the morally justified laws of an illegitimate state. We next ask whether laws that would be morally unjustified if enacted by an illegitimate state might be morally justified if enacted by a democratic state. For example, could a democratic Egyptian government permissibly enact criminal laws that the Mubarak regime could not permissibly enact? In other words, does the scope of permissible criminalization depend on the legitimacy of the regime?

Recently, some scholars have proposed that a legitimate state, specifically a democratically legitimate state, may justifiably pass and enforce criminal laws that are not otherwise morally justified. On this view, democratic enactment itself can provide sufficient moral justification for a legal obligation to make that legal obligation a moral obligation as well. If true, this view would entail that one form of political legitimacy, democratic legitimacy, is sometimes sufficient to justify criminal laws. It would follow that democratically legitimate regimes could justly pass and enforce laws that undemocratic and illegitimate regimes could not justly pass or enforce. If this is the only difference between democratic and undemocratic regimes, so far as criminal justice is concerned, then it is interesting but not terribly exciting. I doubt the activists I met in Egypt braved tear gas, bullets, and police batons so that otherwise morally unjustified laws may be justly passed and enforced and so that laws that have nothing to be said for them except that they were democratically enacted will carry the force of moral obligation. Presumably they took to the streets in part so that morally unjustified laws will be repealed or not passed and morally justified laws passed and enforced. But, to be fair, at the time I did not think to ask them.

Interestingly, Dan Markel argues that citizens of democratic states are morally obligated to obey laws that are ‘dumb but not illiberal’ and democratic states are justified in punishing citizens who break such laws (Markel 2012). By contrast, citizens of undemocratic states have no such moral obligations and may not be punished for breaking such laws. On Markel’s account, a law is ‘dumb’ if it is intended to benefit citizens but proves ineffective or misdirected; a law is ‘not illiberal’ if it does not violate liberal rights or principles of equality. The moral basis of the obligation to obey such laws and of the punishment for breaking such laws is that these laws are democratically enacted.

Intuitively, our strongest reason to oppose criminal laws that are ‘dumb but not illiberal’ and seek their repeal is that individuals will be unjustly punished under those laws. Counter-intuitively, Markel’s view entails that we have no such reason, because he thinks it just to punish individuals under criminal laws that are dumb but not illiberal. This makes it unnecessarily puzzling why we should oppose such laws and seek their repeal. If such laws make the majority happy and unjustly harm no one then it is not clear that their repeal should be a moral priority. Yet the repeal of such laws is a moral priority, and combatting overcriminalization should be a political priority, in democracies and elsewhere, primarily because such laws enable unjust punishment.

Of course, democracies sometimes enact laws that are neither dumb nor illiberal, laws that are morally justified quite apart from their democratic enactment. Presumably, we are doubly morally obligated to obey such laws since they are both democratically enacted and otherwise morally justified.13 Perhaps we have two obligations to obey the law, one grounded in the law’s democratic enactment and the other grounded in the law’s other moral justifications, or perhaps we have one obligation to obey the law with two grounds, each imparting their own moral force.

In either case, it would seem to follow that, on Markel’s view, otherwise morally justified laws impose stronger moral obligations and justify proportionately greater punishment if they are democratically enacted rather than undemocratically enacted. This would mean that democracies justly may be more repressive (they may criminalize more behavior) and more punitive (they may punish the same behavior more severely) than non-democracies. If this result is implausible then it casts substantial doubt on the premise that individuals justly may be criminally punished for breaking laws that are democratically enacted but not otherwise morally justified.14 It therefore seems that a democratic Egyptian government will be governed by the same principles of just criminalization that governed (though they did not guide) the Mubarak regime.

Before moving on, I should note that the democratic process remains the best and fairest mechanism for choosing among morally justified laws. A society considering several possible laws on a given subject, each of which is morally preferable to no law at all, generally should adopt the law favored by the majority of its members or their representatives. In particular, members who favor a different possible law should respect the law favored by the majority. Of course, one may respect a law while seeking its amendment or replacement. Nevertheless, it is excessively utopian to respect a law only if it is better than any possible alternative. Generally, we should respect a law so long as it is better than nothing.


We have seen that political legitimacy does not substantially affect the scope of just criminalization. However, I shall argue that political legitimacy may substantially affect the possibility of just adjudication. If an illegitimate state cannot justly enforce even a just criminal law then this provides an additional reason to overthrow such states and replace them with legitimate states capable of both enacting and enforcing a just criminal law.

If the purpose of the criminal trial is only to determine whether an individual is responsible for a moral wrong for which she lacks justification or excuse, and to sentence individuals to punishments they deserve, then there is no obvious reason why an illegitimate authority cannot administer just criminal trials. However, several important scholars believe that the criminal trial has an intrinsic moral value that an illegitimate authority may not be capable of realizing.

For example, Antony Duff famously argues that the purpose of a criminal trial is to hold individuals accountable for the public wrongs they commit to the public whose core values those wrongs infringe (Duff 2011). Individuals must answer the charges against them through their plea of guilty or not guilty and may have to answer for their actions by offering a justification or excuse. Duff argues that this communicative exchange between the accused and the polity is intrinsically valuable and is not merely a means of determining guilt or innocence. If Duff is right, then the Mubarak regime’s decision to use State Security Courts, as well as the government’s decision to use military courts, undermined their legitimacy by failing to deliver what a legitimate regime presumably would: public trials featuring a communicative exchange between defendants and representatives of the polity. Moreover, if only a legitimate state can represent the polity in criminal proceedings then an illegitimate state necessarily lacks the moral standing to speak for the polity and adjudicate, convict, and punish on its behalf.

An illegitimate state could, of course, provide the accused with various procedural protections: an impartial tribunal, notice of charges and an opportunity to be heard, and the full array of familiar defences. However, the resulting communicative exchange will not be between a citizen and the polity of which she is a member, but rather between an individual and the power to which she is subject. An illegitimate state could still accurately and fairly determine guilt and innocence, but the communicative dimension of criminal justice will be lost. Interestingly, an illegitimate state might incorporate lay juries into the trial process. Perhaps then the accused will be held accountable to the polity of which she is a member, albeit by a state that does not legitimately represent that polity.15 In a legitimate state, lay juries may be unnecessary because state officials including judges represent the polity. But in an illegitimate state a lay jury may be necessary to ensure accountability to the polity.

There is a stronger position, however, namely that the normative effect of criminalization is to make wrongdoers accountable to the state, not to the polity, for the moral wrongs they commit. This seems to be the position of Scott Hershovitz (2011), who argues that when a legitimate state criminalizes conduct its subjects owe an obligation to the state not to engage in that conduct; if a subject violates that obligation she is accountable to the state much as a promisor who violates an obligation to her promisee is accountable to the promisee.16 On this view, we are not accountable to the state for our pre-legal moral wrongs against other people, only for our legal wrongs against the state. This view appears to have the implication that an illegitimate state lacks standing to hold individuals accountable either for wrongs against the state (since we owe no obligation to an illegitimate state) or for wrongs against other people (since we are not accountable to the state for these wrongs).

Perhaps an illegitimate state could hold wrongdoers accountable to their victims (or their victim’s loved ones) for violating their moral obligations to those victims.17 We saw above that an illegitimate state could hold wrongdoers accountable to the public by introducing lay juries. But how could an illegitimate state hold wrongdoers accountable to individual victims? If such accountability would require victims or their loved ones to adjudicate guilt then there is no way for an illegitimate state to have a just criminal procedure. Perhaps wrongdoers could be made accountable to their victims in other ways, by giving victims more influence in prosecution (for example, by allowing victims and their loved ones to legally challenge the state’s failure to prosecute, or to approve plea bargains) or punishment (for example, by allowing victims or their loved ones to waive punishment, recommend sentences, or testify during sentencing). Perhaps those who propose such reforms, including in liberal democracies, implicitly challenge the legitimacy of the state. But if such procedures are insufficient to hold wrongdoers accountable to their victims, or if sufficient procedures would be unjust, then we would have to conclude that an illegitimate state simply cannot justly enforce morally justified criminal laws.

Yet does not an illegitimate state have a conditional responsibility to protect those under its effective control from serious moral wrongs? Indeed it does, but perhaps an illegitimate state could discharge its conditional responsibility through a system of preventive detention, one that does not adjudicate guilt, issue moral condemnation, or administer punishment but that simply removes dangerous wrongdoers from society for so long as they can be proven to remain dangerous. Preventive detention does not provide criminal justice, which illegitimate states cannot justly administer, but rather basic security, which illegitimate states retain a conditional right and obligation to provide.

Preventive detention is probably inherently unjust, but may be justifiable if it would prevent the detained wrongdoers from inflicting substantially greater harm on others.18 The inability of an illegitimate state to justly administer the criminal law therefore sets up a potential moral conflict. Evidently, our first duty with respect to moral conflicts is to prevent such conflicts from arising. This is because our general duty to avoid unjust acts implies a duty to prevent circumstances from arising that would render unjust acts justifiable. The circumstance that arguably renders preventive detention justifiable is the inability of illegitimate authorities to justly administer the criminal law. Legitimate authorities, by contrast, can justly administer the criminal law and therefore are unjustified in implementing preventive detention. We may therefore have one more reason (indeed, one more duty) to overthrow illegitimate authorities and replace them with legitimate authorities. Through revolution we would avoid the moral conflict between the injustice of preventively detaining criminal suspects and the justifiability of protecting crime victims. We would instead create a legitimate state that can protect crime victims by justly administering a just criminal law.

If illegitimate authorities have a conditional justification for preventive detention, they thereby gain no advantage over legitimate authorities. The justification is only conditional, and illegitimate authorities remain under a categorical obligation to surrender power. Legitimate authorities can justly administer the criminal law while illegitimate authorities cannot. And if illegitimate authorities are justified in unjustly using preventative detention then the avoidance of this justifiable injustice provides one more reason to overthrow illegitimate authorities and replace them with legitimate authorities.


On the morning of August 3, 2011, a judge in Cairo called on the first of ten criminal defendants in the courtroom to answer the charges against him. The charges included the killing of protesters as well as corruption. The defendant, Hosni Mubarak, denied the charges completely. The other defendants, including Mubarak’s two sons, a former Interior Minister, and six senior police officers, did the same (Shadid 2011). Finally, it seemed, Mubarak and his confederates would have to answer for their crimes. One Egyptian newspaper declared it ‘The Day of Judgment’. Such declarations proved premature. Ten months later, Mubarak was indeed convicted as an accessory to the murder of 240 protesters. At the same time, many officials accused of ordering and carrying out these murders were acquitted, triggering yet another round of protests (Kirkpatrick 2012).

Egypt’s newly-elected president, Mohamed Morsi, promised during his campaign to launch a new round of criminal prosecutions once elected. At the time of this writing, it remains to be seen whether Morsi will follow through on his pledge. What we have seen, I hope, is that to the extent that Egyptians rose up to both prevent and punish the crimes of his regime they did so with good reason. Mubarak’s regime compromised its legitimacy by committing criminal acts, by abusing and displacing the criminal process, and by failing to punish serious wrongdoing. Once the regime lost its legitimacy it also lost its capacity to justly hold individuals accountable either to the polity or to the state through criminal trials. This incapacity provided an additional reason to overthrow the regime and replace it with a regime capable of justly administering a just criminal law.

The legitimacy of Egypt’s new government remains contested, and may turn on its own crimes, its own perversions and subversions of the criminal process, and its own tolerance for impunity. If the laws enacted and enforced by the new government are morally justified then Egyptians have a moral obligation to obey those laws. If the election of a new parliament and president leads to a fully democratic government, that government can only morally obligate its citizens to obey morally justified laws and cannot punish its citizens unless they are morally responsible and morally at fault for violating those laws. Such a democratic government will have an easier time maintaining its legitimacy, since regular elections make it possible to replace failed leaders through legal processes rather than by popular revolt. It is in this sense that democratic election is a better legitimacy-preserving mechanism than it is a legitimacy-conferring mechanism. But in any case, Egypt’s new government will prove legitimate only so long as its rule remains morally preferable to a second revolution. We shall see.


  1. 1.

    I should add that any moral obligation to obey the otherwise morally justified legal duties imposed by a conditionally legitimate state is opposed by a range of competing reasons to disobey, including the use of disobedience as part of a strategy to deprive the state even of conditional legitimacy.

  2. 2.

    Additionally, large-scale embezzlement by the regime and widespread extortion by the police were crimes in themselves and also compounded the socio-economic injustices afflicting Egyptian society.

  3. 3.

    Punishment is indeed a relational activity, although I have argued that in its central case punishment is a triadic relationship in which the punisher owes the victim of wrongdoing a duty to punish the perpetrator (Haque 2005).

  4. 4.

    Of course, the fact that my promise created expectations may give you a reason to do what I promised to do in order to satisfy those expectations. But my promise is not your reason for action.

  5. 5.

    For example, international law generally obligates new states to pay the debts and honor the treaties of their predecessor states. The predecessor state obligation terminates with the predecessor state, but the new state incurs a new obligation to do what the predecessor state was obligated to do.

  6. 6.

    Victor Tadros reminds me that Kant, who inspires Binder and Thorburn, generally denied the moral permissibility of revolution. My impression is that Kant’s position on revolution is not widely accepted, even among Kantians. However, if Binder and Thorburn share Kant’s position on revolution then they are in worse shape than I previously imagined.

  7. 7.

    For the classical dramatization see Aeschylus (1969). For a philosophical exposition see Gardner (1998, p. 31) (‘The blood feud, the vendetta, the duel, the revenge, the lynching: for the elimination of these modes of retaliation, more than anything else, the criminal law as we know it came into existence.’).

  8. 8.

    I first explored this argument in Haque (2009). For a very different view of the moral burdens borne by private punishers see Harel (2007).

  9. 9.

    Note, however, that it is possible to argue that there is a duty to punish in order to achieve general deterrence. See Tadros (2011).

  10. 10.

    Compare Husak (2008)(describing internal constraints on criminalization).

  11. 11.

    Some scholars argue that the moral asymmetry between positive and negative obligations does not apply to state conduct. See Sunstein and Vermeule (2005) and Lee (2011). For my response, see Haque (2007, pp. 628–635).

  12. 12.

    It is also worth noting that individual officials may have competing reasons not to enforce the laws of a conditionally legitimate state, including the use of non-enforcement as part of a strategy to deprive it of even conditional legitimacy. However, these competing reasons are more likely to be decisive regarding crimes that are mala prohibita rather than mala in se.

  13. 13.

    At one point, Markel expresses doubt that moral obligations are additive in any relevant sense. Unfortunately, Markel’s example of non-additive obligations (two promises, one forgotten, to the same person to perform the same act) muddies the waters. If I promise two different people to perform the same act then I have stronger reasons to perform that act than if I promise only one person, and I require stronger reasons to justify violating both promises than to justify violating only one. More clearly still, if you promise your spouse not to commit some minor marital wrong (shirking your household duties, for example) then you have an additional duty to avoid that wrong and would need stronger reasons to justify committing that wrong.

  14. 14.

    Victor Tadros reminds me that we may have other reasons to obey morally unjustified laws, for example if failing to do so will encourage other people to disobey morally justified laws.

  15. 15.

    Similarly, the defendant in a tort action is held accountable by the court to her victim.

  16. 16.

    I write ‘seems’ because Hershovitz at one point writes that ‘in most developed democracies, laws that prohibit murder reflect legitimate exercises of authority precisely because it is valuable for people be accountable to one another through the state for serious moral transgressions of that sort’ (Hershovitz 2011, p. [19], emphasis added). This passage seems to imply that, even in a legitimate state, individuals are not accountable to the state; instead, the state holds individuals accountable to one another. But this position seems at odds with almost everything else that Hershovitz says.

  17. 17.

    By way of comparison, it seems that an illegitimate state could justifiably hold a tortfeasor accountable to her victim for violating her duty to her victim.

  18. 18.

    Victor Tadros has suggested to me that an illegitimate state may also be justified in punishing for the sake of general deterrence. However, I expect Tadros would agree that deterrent punishment is substantially harder to justify than preventive detention. As Tadros argues elsewhere, it is harder to justify harming someone to prevent future wrongdoing by others, for which they will not be responsible, than to justify harming someone to prevent them from committing future wrongs themselves, for which they will be responsible. Moreover, it is harder to justify harmfully manipulating someone than to justify harmfully eliminating the threat they pose. See Tadros (2011). Since general deterrence is substantially harder to justify than preventive detention it is highly unlikely that an illegitimate state would be even conditionally justified in punishing for the sake of general deterrence.



Thanks to Michael Cahill, Glenn Cohen, Antony Duff, David Gray, Dan Markel, and Victor Tadros for comments on an earlier draft.


  1. Aeschylus. (1969). The Eumenides. In D. Grene & R. Lattimore (Eds.), Aeschylus I: Oresteia: Agamemnon, The Libation Bearers, The Eumenides. Chicago: University of Chicago Press.Google Scholar
  2. Associated Press. (2011). Egyptians protest, demand justice after Mubarak. New York Times, July 8, 2010,
  3. Berman, M. (2011). Two kinds of retributivism. In R. A. Duff & S. P. Green (Eds.), Philosophical foundations of criminal law. Oxford: Oxford University Press.Google Scholar
  4. Binder, G. (2002). Punishment theory: Moral or political? Buffalo Criminal Law Review, 5(321), 327–328.Google Scholar
  5. Cahill, M. (2007). Retributive justice in the real world. Washington University Law Review, 85, 815.Google Scholar
  6. Christie, N. (1977). Conflicts as property. British Journal of Criminology, 17, 1–15.Google Scholar
  7. Duff, R. A. (2011). Retrieving retributivism. In M. D. White (Ed.), Retributivism: Essays on theory and policy. Oxford: Oxford University Press.Google Scholar
  8. Fletcher, G. P. (2007). The grammar of criminal law. Oxford: Oxford University Press.Google Scholar
  9. Gardner, J. (1998). Crime: In proportion and in perspective. In A. Ashworth & M. Wasik (Eds.), Fundamentals of sentencing theory: Essays in honour of Andrew von Hirsch. Oxford: Oxford University Press.Google Scholar
  10. Gardner, J. (2007). Complicity and causality. Criminal Law and Philosophy, 1, 127.CrossRefGoogle Scholar
  11. Gardner, J. (2010). Law and morality. In J. Skorupski (Ed.), The Routledge companion to ethics. London: Routledge.Google Scholar
  12. Hammer J. (2011). Egypt: Who calls the shots? New York Review of Books, August 18, 2011,
  13. Haque, A. A. (2005). Group violence and group vengeance: Toward a retributivist theory of international criminal law. Buffalo Criminal Law Review, 9, 273.CrossRefGoogle Scholar
  14. Haque, A. A. (2007). Torture, terror, and the inversion of moral principle. New Criminal Law Review, 10, 613.Google Scholar
  15. Haque, A. A. (2009). Sharing the burdens of justice. In P. H. Robinson, S. Garvey, & K. Ferzan (Eds.), Criminal law conversations. Oxford: Oxford University Press.Google Scholar
  16. Harel, A. (2007). Why only the state may inflict criminal sanctions: The argument from moral burdens. Cardozo Law Review, 28, 2629.Google Scholar
  17. Hart, H. L. A. (1968). Punishment and responsibility: Essays in the philosophy of law. Oxford: Oxford University Press.Google Scholar
  18. Hershovitz, S. (2011). The authority of law. In A. Marmor (Ed.), The Routledge companion to philosophy of law. London: Routledge.Google Scholar
  19. Husak, D. (2008). Overcriminalization. Oxford: Oxford University Press.Google Scholar
  20. Kirkpatrick, D. (2011). Egypt: protests against military spread beyond Cairo to New Cities. New York Times, December 23, 2011,
  21. Kirkpatrick, D. (2012). New turmoil in Egypt greets mixed verdict for Mubarak. New York Times, June 2, 2012.Google Scholar
  22. Law 162/1958, Qanun bi Sha’n Halah al-Tawari’ [Law Concerning the State of Emergency], as amended.Google Scholar
  23. Lee, Y. (2011). Deontology, political morality, and the state. Ohio State Journal of Criminal Law, 8, 385.Google Scholar
  24. Markel, D. (2012). Retributive justice and the demands of democratic citizenship. Virginia Journal of Criminal Law, 1, 1.Google Scholar
  25. Metz, T. (2007). How to reconcile liberal politics with retributive punishment. Oxford Journal of Legal Studies, 27, 683–705.CrossRefGoogle Scholar
  26. Philips, M. (1986). The justification of punishment and the justification of political authority. Law and Philosophy, 5, 393.CrossRefGoogle Scholar
  27. Reza, S. (2007). Endless emergency: The case of Egypt. New Criminal Law Review, 10, 532.Google Scholar
  28. Shadid A. (2011). At Mubarak Trial, stark image of humbled power. New York Times, August 3, 2011,
  29. Sunstein, C. R., & Vermeule, A. (2005). Is capital punishment morally required? Acts, omissions, and life–life tradeoffs. Stanford Law Review, 58, 703.Google Scholar
  30. Tadros, V. (2011). The ends of harm: The moral foundations of criminal law. Oxford: Oxford University Press.CrossRefGoogle Scholar
  31. Thorburn, M. (2011). Constitutionalism and the limits of the criminal law. In R. A. Duff, L. Farmer, S. Marshall, & V. Tadros (Eds.), Structures of criminal law. Oxford: Oxford University Press.Google Scholar

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  1. 1.NewarkUSA

Personalised recommendations