CRIMES AGAINST HUMANITY AND THE LIMITS OF INTERNATIONAL CRIMINAL LAW

. Crimes against humanity are supposed to have a collective dimension with respect both to their victims and their perpetrators. According to the orthodox view, these crimes can be committed by individuals against individuals, but only in the context of a widespread or systematic attack against the group to which the victims belong. In this paper I offer a new conception of crimes against humanity and a new justiﬁcation for their international prosecution. This conception has important implications as to which crimes can be justiﬁably prosecuted and punished by the international community. I contend that the scope of the area of international criminal justice that deals with basic human rights violations should be wider than is currently acknowledged, in that it should include some individual violations of human rights, rather than only violations that have a collective dimension.


CRIMES AGAINST HUMANITY AND THE LIMITS
OF INTERNATIONAL CRIMINAL LAW (Accepted 30 September 2011) ABSTRACT. Crimes against humanity are supposed to have a collective dimension with respect both to their victims and their perpetrators. According to the orthodox view, these crimes can be committed by individuals against individuals, but only in the context of a widespread or systematic attack against the group to which the victims belong. In this paper I offer a new conception of crimes against humanity and a new justification for their international prosecution. This conception has important implications as to which crimes can be justifiably prosecuted and punished by the international community. I contend that the scope of the area of international criminal justice that deals with basic human rights violations should be wider than is currently acknowledged, in that it should include some individual violations of human rights, rather than only violations that have a collective dimension.

I. THE NOTION OF CRIMES AGAINST HUMANITY
In this article I offer a new conception of crimes against humanity and a new justification for the international prosecution of these crimes. An important implication of the view I offer is that the category of crimes against humanity should be expanded to cover crimes that are not currently covered by it under international law. This in turn would have important consequences in terms of the types of crime that can be justifiably prosecuted and punished by the international community. Let me start by saying something about the structure of crimes against humanity as they are currently understood. The Rome Statute lists a series of particularly serious crimes, such as murder, enslavement, torture, rape and enforced prostitution, and then specifies that these acts are to be considered crimes against humanity when they are 'committed as part of a widespread or systematic attack directed against any civilian population, with knowledge of the attack'. 1 Thus, in order to count as a crime against humanity an act must meet two conditions: first, the act must fall within the list of inhumane acts itemized in the statute; second, the act must fulfill the so-called 'contextual element', i.e. must be committed as part of a wider attack against a specific group. 2 We can identify four salient features of the notion of crimes against humanity as it is currently understood 3 : (1) they constitute particularly odious offences. These crimes are so barbarous as to violate the human dignity of the victims; (2) they are international crimes. These crimes concern the international community, rather than just the domestic political community, and therefore trigger international intervention. It is permissible for the international community to trump state sovereignty in order to punish them 4 ; (3) they have a policy element: they are committed, instigated, or at least tolerated, by a state, a de facto authority or a politically organized group. (4) they have a collective element: they target victims qua members of a group.
This is not to say that crimes against humanity cannot be committed by individuals, nor is it to say that wrongdoers must be moved by the desire to target their victims qua members of the group. The idea is rather that crimes against humanity can be committed by individuals only to the extent that they can be seen as part of a widespread or systematic attack. This requires wrongdoers to be aware that their victims belong to a group that is targeted by the wider attack. It is in this sense that crimes against humanity are group-based. They can be committed against individuals only insofar as the group to which these individuals belong is also attacked, who need to live in politically organized groups. When these politically organized groups use their power, perversely, to attack rather than to protect their citizens, this fundamental aspect of the victims' humanity is assaulted; and this assault is so serious that the international community cannot leave it unpunished. 8 This view seems to me largely convincing as an account of the legal notion of crimes against humanity as it has developed since Nuremberg and it is not my intention to question it. The category of crimes against humanity became prominent after the end of World War II, when the international community decided for the first time to prosecute political and military leaders for crimes committed against their own subjects. The main problem at the time was that these crimes were not prohibited by international law, which only proscribed crimes committed against foreign populations. The category of crimes against humanity was introduced precisely to remedy this lacuna and, in the words of the British Chief Prosecutor Sir Hartley Shawcross, finally put 'a limit to the omnipotence of the State'. 9 Broadly speaking the same rationale underlies the introduction of this category in the statutes of the ICTY, the ICTR and the ICC, so in focusing on the idea that crimes against humanity are those that involve a 'perversion of politics', the account offered by Luban and Vernon does capture the central feature of these crimes as they are currently understood. However, there seems to be an important tension within this understanding of the notion of crimes against humanity. In Adil Haque's words, international criminal law is frequently portrayed as the strong arm of the international human rights regime, an instrument designed to safeguard the dignity of each human person. There is an important truth to this characterization: international crimes involve many of the most grotesque violations of individual rights human beings inflict and endure. Yet … the law governing crimes against humanity and genocide frames the acts and fates of individuals against broader and darker patterns of group perpetration and group victimization. It is only within the context of group violence that international law attributes individual responsibility for wrongdoing and vindicates the rights of the victims. As this passage suggests, the collectivist nature of crimes against humanity is at odds with our common understanding of human rights as those rights that have the function of protecting the dignity of each human person. The point of acknowledging that certain rights have the status of human rights is to vindicate the idea that the individual enjoyment of these rights should not be affected by states' failure to protect them and enforce them. Since these rights protect human dignity and allow individuals to lead minimally decent lives, they should be guaranteed whether or not states are willing and able to do so.
This conception of human rights has important implications with respect to the question of how we should understand the relationship between international and domestic political order. Before the rise of the human rights discourse individuals were under the exclusive jurisdiction of the State of which they were nationals and where they lived. No other State could interfere with the authority of that State, which in a way had a sort of right of life and death over those individuals. Beyond national boundaries individuals could only be taken into consideration qua citizens of a foreign State. 11 The significance of the doctrine of human rights is to be found in the shift from this scenario to one in which individuals were 'no longer to be taken care of, on the international level, qua members of a group, a minority, or another category. They began to be protected qua single human beings '. 12 Yet this focus on individuals seems to be at odds with the collectivist nature that crimes against humanity are supposed to have in virtue of their contextual element. The function of the contextual element is to ensure that only when someone is murdered or tortured in connection to a wider attack on the group of which she is part can the international community intervene to prosecute and punish the crime (at least if domestic courts fail to do so). If the same crime is committed against an individual outside the context of a wider attack on a group, what we have is a domestic crime, over which national courts have exclusive jurisdiction.
In this paper I will argue that the way in which we should try to overcome this tension is by expanding the scope of the notion of crimes against humanity. The scope of the area of international criminal justice that deals with serious human rights violations should be wider than is currently acknowledged, in that it should include some individual violations of human rights, rather than only violations that have a collective dimension.
In particular, I will defend the following three theses: first, crimes against humanity are those that 'deny their victims the status of human being'. Second, crimes of this kind properly concern the whole of humanity and thus have an international dimension. This means that they trigger international punishment, where this can be inflicted either by an international institution (such as the ICC, the ICTY or the ICTR) or by a domestic court claiming universal jurisdiction.
Third, the crimes listed in the Rome Statute and other international statutes deny their victims' status of human being even when they have no collective element or policy element, i.e. even when they are committed in isolation from a widespread or systematic attack against the group to which the victim belongs. These 'individualized' crimes are to be considered crimes against humanity and therefore have an international dimension, in addition to their traditional dimension of domestic crimes.

II. CRIMES AGAINST 'HUMANNESS'
Any account of crimes against humanity will have to provide an answer to two questions: one is the conceptual question of how we should understand the notion of crimes against humanity. In other words, what do we mean when we label certain crimes as 'against humanity?' The other is the normative question of what justifies the international prosecution of these crimes. In other words, why does the international community have the right to prosecute and punish crimes against humanity? Normally states have exclusive jurisdiction over crimes committed within their borders, but in the case of crimes against humanity, this right can be trumped by the right of the international community to punish them. What is the justification for this right? 13 The two questions are obviously connected, since the reasons why crimes against humanity ought to be prosecuted by the international community, rather than by the domestic community, will depend on those features that distinguish them from domestic crimes. But this is precisely why we should not lose sight of the distinction. It is important that we bear in mind the implications of labeling something as a crime against humanity. In this section I suggest a new answer to the conceptual question. In the next section this answer will be employed to tackle the normative question.
There are two main ways of understanding crimes against humanity: as crimes 'against humanness', i.e. crimes that somehow violate the core humanity that we all share, or as crimes 'against humankind', i.e. crimes that harm not only their direct victims, and possibly the political community, but all human beings. 14 According to the former understanding, what is at stake when crimes against humanity are committed is the violation of a value, humanness, which is offended by these crimes. According to the latter, what is at stake is an attack against a set of individuals, humankind, which is harmed by them. What I intend to offer here is a version of the first view. My suggestion is that each of the acts listed in the definition of crimes against humanity of the Rome Statute and other international documents constitutes a serious attack on the human dignity of the victims -one that can be said to deny their status of human being. 15 It is for this reason, I submit, that these crimes concern the international community.
What does it mean that crimes against humanity 'deny their victims' status of human being? 'I will suggest that these crimes deny their victims' status of human being because they violate some basic human rights of their victims. Crimes involving human rights violations of this sort constitute an assault on the humanity of the victims, and it is in this sense that they are to be considered 'against humanity'.
In developing my argument I will appeal to the traditional conception of human rights as those rights that we have simply in virtue of our humanity, so let me start by briefly outlining this conception. 16 According to this view, the role of human rights is to protect the dignity attached to the status of human being. All human beings are supposed to enjoy these rights no matter where they live, no matter what their social or economic condition is, and no matter whether these rights are included in the constitution of their state. These rights belong to human beings as such. A full discussion of this conception is obviously beyond the scope of this paper, but let me highlight what I take to be its two key features before I go on to explain how this is relevant for the notion of crimes against humanity that I intend to defend.
The first feature is the idea that human rights are pre-institutional. Individuals possess them in a pre-political state of nature, and this is why they do not depend for their existence on their being embodied in any political institutions or on their being recognized in any political doctrines. 17 The second feature is the claim that human rights are ultimately to be justified with respect to the idea of the dignity of human persons. Of course there is some disagreement as to how we should understand both the concept of 'dignity' and the concept of 'human' in this context (not to mention the disagreement over what these rights and their implications are); still human dignity is the notion that this conception typically appeals to.
The notion of dignity plays a double role here. Firstly, human dignity is the value that explains why all human beings can be said to have human rights: it is in virtue of their intrinsic dignity, however we understand the notion, that human beings are in possession of these rights. Secondly, human dignity constitutes the ultimate value that human rights are supposed to protect. These rights protect human dignity by placing limits on how human beings can be treated. The idea here is that in order to lead a life of dignity, human 16  beings need to be protected against certain 'standard threats'. 18 Human rights protect them against these threats, so that they can have a minimally decent life. Thus the reason why all human beings possess human rights, independently of their social relationships and undertakings, is that all human beings are entitled to live a life of dignity, and this is possible only to the extent that they are not treated in certain ways. Human rights provide individuals with protections against being treated in these ways. We can now see in what sense crimes against humanity 'deny their victims the status of human being' in virtue of the fact that they violate some of their most important human rights. Any time that these rights are violated the humanity of the victim is 'denied' in the sense that the victim is treated as if she did not have those basic protections that all human beings are entitled to simply in virtue of the fact that they are humans. The victim is treated 'as if she was not human' because she is treated as if she did not enjoy those protections that automatically attach themselves to every human being and that are meant to shield human beings against serious abuses of their dignity and their fundamental interests.
Jean Hampton has suggested that there are actions that 'morally injure' their victim, in the sense that the victim is treated in a way which is precluded by her value. These actions represent the value of their victim as less than the value that should be accorded to her, because they deny the entitlements which are generated by that value. 19 I contend that this is how we should understand crimes against humanity. Violations of fundamental human rights deny that their victims have the status of human being, in that they treat the victims as if they did not have those basic protections that all human beings have simply in virtue of their being humans.
It should be noticed here that I call 'right violations' those cases in which acting against someone's right is not morally justified and 'right infringements' those cases in which acting against someone's right is morally justified. When human rights are infringed the humanity of the victim is not denied because, to the extent that the agent acts with moral justification, the victim is not treated as if she did not enjoy those protections that every human being is entitled to. Rather, those protections are recognized, but justifiably overridden. When this is the case, no crime against humanity is committed. 20 One obvious objection should be considered at this point: the class of violations of human rights seems to be much wider than the class of crimes against humanity. The list of human rights that appear in the Universal Declaration and other international documents includes not only rights to physical security (e.g. the right not to be tortured), but also civil and political rights (e.g. freedom of expression), social rights (e.g. non-discrimination for minorities) and, more controversially, minority rights and environmental rights. However, the lists of crimes against humanity that appear in the Rome Statute and other international statutes is limited to crimes such as torture, murder and rape, and we would not want this list to be expanded to include violations of rights such as the right to paid holidays. But if I am right that crimes against humanity are those that treat the victim 'as if she was not human', where this simply means that some of the human rights of the victim are violated, shouldn't we conclude that any human rights violation is a crimes against humanity?
The reply to this objection is to point at the common distinction between basic and non-basic human rights, and claim that crimes against humanity are committed only when the former are violated. 21 Probably the best known way of characterizing the notion of basic human rights (such as the right to life or to bodily integrity) is by appealing to Henry Shue's idea that they are those rights that we need to enjoy in order to enjoy all other rights. 22 Other philosophers however, have focused on different aspects: for example, Carol Gould suggests that these are 'rights to the conditions that are minimally necessary for any human action whatever'. 23 These formulations cannot be discussed here for reasons of space. In any case, what matters for the purposes of my discussion is the underlying idea that they seem to share, namely that basic rights are rights to 20 Here I adopt Jeff McMahan's formulation of the infringing/violating distinction, which is slightly different from the classic formulation by Judith Jarvis the conditions that are necessary for a minimally decent life. Whereas non-basic human rights (such as the right to paid holidays) are rights to the conditions required for a minimally happy or a flourishing life, basic rights 'specify the line beneath which no one is to be allowed to sink', and thus constitute 'everyone's minimum reasonable demands upon the rest of humanity'. 24 Ideally all human beings should have a happy life, but it is certainly possible to live a life of dignity which is not a happy life. This is why the list of acts that constitute crimes against humanity is limited to violations of basic human rights (i.e. crimes such as torture, rape and murder): only when the rights to the conditions necessary for a minimally decent life are violated is the human status of the victim denied in the sense I have specified above. The constituent acts listed in international criminal statutes are those that involve violations of these rights. 25

III. TWO KINDS OF WRONGS
I will now turn to the normative question of what justifies the international prosecution of crimes against humanity. According to the view presented in the previous section, crimes against humanity are those which deny their victims the status of human being, i.e. those crimes that violate basic human rights of their victims. 26 But even if we accept this account, we still need to address the question of what justifies trumping national sovereignty in order to prosecute and punish these crimes. Why should not states have exclusive jurisdiction over them, as they do in the case of domestic crimes?
One natural way of trying to account for the right of the international community to override national jurisdiction in order to 24 Basic Rights, Notice that although I tend to agree with those who complain about the excessive proliferation of human rights (it is a mistake to think that all important rights are thereby to be considered human rights), I do not intend here to claim that non-basic human rights are not genuine human rights. However, to the extent that these are rights to the conditions required for a minimally happy or a flourishing life, rather than to a minimally decent life, they are to be considered human rights only in a derivative or secondary sense. This is why we should not expect the notion of crimes against humanity I advance to cover violations of these rights. 26 In the rest of the paper I will keep using the expression 'deny the humanity of the victim' in the technical sense specified above, i.e. to refer to those acts that violate basic human rights of the victim. Other types of acts (for example, discriminatory or unfaithful conduct) are sometimes said to 'deny the humanity of the victim' in a different sense. According to my view these acts do not constitute crimes against humanity because they do not target the basic protections that we need in order to have a minimally decent life (unless they also violate basic rights).

CRIMES AGAINST HUMANITY AND THE LIMITS OF ICL
punish crimes against humanity is by appealing to the idea that these crimes somehow harm the whole of humanity, rather than the members of the domestic political community. This strategy is developed by Larry May, who argues that in the same way in which states have the right to punish domestic crimes because they harm the domestic political community, the international community has the right to punish international crimes because they harm the international community. 27 Elsewhere I have argued that this conception is problematic, and that we have reason to abandon a harm-based model of crimes against humanity in favour of an accountability-based model. The justification for the international prosecution and punishment of crimes against humanity does not lie in the fact that they harm the interests of the international community, but rather in the fact that wrongdoers are accountable for these crimes to the members of the international community (rather than just to their fellow citizens). 28 This view relies on a conception of criminal law defended by Antony Duff, according to which the function of criminal law is to specify those wrongs that citizens are accountable for to their fellow citizens. Crimes are 'public wrongs' not in the sense that they harm the political community, but rather in the sense that they are the kind of wrongs for which we are accountable to the members of our political community. This means that those who commit these wrongs are answerable for what they have done to the whole polity, rather than just to their victims. Punishing them is the way in which the political community calls them to account for these wrongs. 29 In this paper I will not attempt to defend this view (nor will I rehearse my arguments against the harm-based account of crimes against humanity), but will simply rely on it in order to explain why the international community has a right to punish crimes against humanity. I will suggest that in the same way in which we are answerable to the members of our polity for domestic crimes, we are answerable to the international community for crimes against humanity. Just like domestic punishment is the way in which wrongdoers are called to account for the former, international punishment is the way in which they are called to account for the latter.
The accountability-based model relies on a relational notion of responsibility as answerability, according to which being responsible is being answerable for something, to someone, in virtue of our occupying a certain role. In Antony Duff's words, '[t]o be responsible, is to be answerable; answerable to a person or body who has the right or standing to call me to account; and I am thus answerable in virtue of some normatively laden description, typically a description of a role that I satisfy'. 30 In virtue of our occupying different roles and sharing different affiliations we are responsible (i.e. answerable) to certain specific groups for certain specific matters, and it is the roles we occupy that determine what we are responsible for to whom. For example, as a teacher, I am responsible to my colleagues for complying with the rules of my department, whereas as a son I am responsible to the members of my family for taking care of my parents when they are sick. These responsibilities correlate to a right that members of my family and members of my department have to call me to account for my failing to attend to my duties as a colleague and as a son respectively. My colleagues have no right to call me to account for not spending more time with my parents when they are sick, because I am answerable to them only for those responsibilities I have as a member of the department. Similarly, I am not accountable to my family for having missed the deadline of a funding application, because I am accountable to them only for those responsibilities that I have as a member of my family.
This conception of responsibility provides a justification of domestic criminal responsibility once we focus on the role of citizen: as citizens we are answerable to the other members of the political community in relation to those wrongs that violate the foundational values of our political community. These wrongs are 'public' in the sense that they are not just private matters between the offender and the victim, but rather concern all the members of the political community. 31 However, saying that these are wrongs for which we are accountable to the members of our polity is not yet saying that they are criminal wrongs. Being accountable to A in relation to x means only that A has the standing to call me to answer for my behaviour relating to x, and to censure me if my behaviour has been wrongful, but there are many different ways in which we can be called to account, and different degrees and forms of censure that are appropriate. Many of the wrongs for which we are answerable to our polity call for criticisms and censure, but not for the specific kind of censure that punishment is; for the condemnatory force of punishment is so great that only those wrongs that cross a given threshold of seriousness ought to be properly censured in this way. 32 Thus, domestic criminal wrongs are a subset of those wrongs for which we are accountable to the polity, namely those wrongs serious enough to justify their inclusion in the criminal law.
As I have indicated, my intention is to apply this framework at the international level in order to address the normative question of what justifies the prosecution and punishment of crimes against humanity. Whereas Duff's account focuses on the responsibilities attached to our identity as citizens, I intend to consider the responsibilities attached to our identity as human beings. I want to suggest that in the same way in which there are wrongs for which we are accountable to our fellow citizens in virtue of our membership in the polity, there are wrongs for which we are accountable to our fellow human beings in virtue of our membership in the wider community of humanity. And as in the case of domestic responsibilities, when these wrongs cross a given threshold of seriousness, the appropriate way of censuring them is by way of punishment. In other words, in the same way in which domestic punishment is justified as a way of calling wrongdoers to account for those wrongs that they are answerable for in virtue of their being 31 Answering for Crime, p. 52. 32 Grant Lamond, 'What Is a Crime?', Oxford Journal of Legal Studies 27 (2007): 609-632.
MASSIMO RENZO members of the polity, international punishment is justified as a way of calling wrongdoers to account for those wrongs that they are answerable for in virtue of their being members of the community of humanity. 33 Defending this view requires clarifying which wrongs we are answerable for to the domestic political community and to the community of humanity respectively. In order to do this I will start by distinguishing between two kinds of wrongs: wrongs that consist in our failure to discharge duties that we have in virtue of our membership in the political community and wrongs that consist in our failure to discharge duties that we have independently of such membership, i.e. pre-institutional duties that all individuals have to each other independently of any social or political relationship. This distinction is only partially overlapping with the traditional distinction between mala prohibita and mala in se, for mala prohibita are normally defined as conduct that is wrongful in virtue of its being criminalized 34 or in virtue of its being legally regulated, 35 whereas here I am interested in conduct that is wrongful in virtue of our membership in the political community, whether or not it is criminalized or regulated by the law.
For example, many believe that in virtue of our membership in the political community -at least one grounded in liberal valueswe have special redistributive responsibilities towards fellow citizens. If so, it would be wrong not to redistribute some of our resources to compatriots, independently of whether there are laws regulating this area of conduct. More interestingly for the purposes of this paper, consider the relationship between theft and the existence of a system of private property. The moral reasons that we have not to steal depend on our being part of political societies that have adopted a system of private property (whether or not theft is explicitly criminalized or regulated by the law). Indeed, were it not for the existence of a socially created system of moral rules protecting private prop-

CRIMES AGAINST HUMANITY AND THE LIMITS OF ICL
erty, taking an object over which someone else claims control would not even count as stealing. Absent such a system of rules -for example in a Hobbesian, but also in a Kantian, state of nature -we would have no stringent moral reasons to refrain from appropriating things over which others claim control. 36 Consider now by contrast crimes that involve the violation of basic human rights, such as torture or rape. These crimes are different from theft in one fundamental respect. We have a moral obligation to refrain from committing them simply in virtue of the respect that we owe to others qua human beings. Our obligation to refrain from these crimes does not depend on the existence of the state or of any system of social rules prohibiting these conducts. Even if we lived in a state of nature, or in any other social context that did not prohibit torture or rape, we would have an obligation to refrain from committing these crimes against other individuals, for having this obligation depends on our being their fellow human beings rather than on our being their fellow citizens.
Here I do not intend to commit myself to any specific account of what grounds the obligations we owe to all human beings by virtue of our common humanity. 37 My aim is rather to discuss how the distinction between these two kinds of wrongs is relevant for the question of what we are accountable for to our political community and to the community of humanity respectively. The natural answer is that we are accountable to the polity for failing to discharge the duties we have in virtue of our membership in the political community, while we are accountable to the community of humanity for failing to discharge the duties we have independently of such membership. This solution seems compelling, as it elegantly associates the group to which we are answerable for u-ing with the group whose membership grounds our duty not to u: since it is in virtue of our membership in the polity that we have a duty not to steal, we are accountable to the other members of the polity for stealing; since it is in virtue of our membership in the group of humanity that we have a duty not to rape, we are accountable to the 36 Apart of course from special transactions that we might enter into with other individuals (if transactions of this kind were possible at all). Admittedly things would be different in a Lockean state of nature, for Locke includes ownership of property in his list of natural rights. 37 An approach particularly well-suited to the view of crimes against humanity that I offer is provided in Stephen Darwall, The Second-Person Standpoint.
whole of humanity for committing the crime of rape. This seems plausible for the same reason why we plausibly think that we are responsible to the other members of the family for our failure to discharge the duties that we incur qua members of the family, or to the members of our department for our failure to discharge the duties that we incur qua members of the department.
Less plausible would be to shift from this claim to the stronger claim that we are accountable to the polity only for those wrongs that consist in our violation of the duties we have in virtue of our membership in the political community. We should not conflate the role in virtue of which we are answerable to someone for u-ing with the role in virtue of which we have a duty not to u, for the two do not necessarily coincide. Duff, for example, acknowledges that our duty not to commit crimes such as murder or rape does not depend on our membership in the political community -we have such a duty simply qua human beings -and yet insists that since these crimes violate the basic values of the polity, we are accountable for them to the political community. I think he is right about this and below I explain why. For now, the point I want to make is merely that we should reject the simple structure according to which: (1) we are accountable to the political community only for wrongs that violate the duties we have qua citizens and (2) we are only accountable to the international community for wrongs that violate the duties we have qua human beings.
Duff's position, on the other hand, would be problematic if it were not limited to the claim that we are accountable to the political community for crimes such as murder or rape, but consisted of the stronger claim that we are only accountable to the political community for these crimes (when they do not have a group-based component). While Duff does not explicitly make this move, his discussion of crimes for which accountability to humanity can be invoked revolves exclusively around crimes committed as part of a widespread or systematic attack against a civilian population, suggesting that it is only for these crimes that we can be called to answer to humanity. 38 But why should that be the case? Why are we not answerable to the whole of humanity for individual cases of murder and rape? If we are answerable to other family members for failing to discharge the duties we have qua sons or parents, and to our colleagues for our failure to discharge the duties that we have qua members of the department, why are we not answerable to our fellow human beings for failing to discharge the duties we have qua human beings (of which the duty not to murder and not to rape are clear examples)? Absent a convincing answer to these questions, I believe we should conclude that endorsing the relational conception of responsibility commits us to the view that we are.
The structure of domestic crimes and of crimes against humanity is therefore similar. In both cases we are dealing with serious wrongs for which we are accountable to the members of a certain group in virtue of our membership in the group. The difference between the two is that in the case of domestic crimes the relevant group is our political community, 39 while in the case of crimes against humanity the relevant group is humanity. While (as I explain below) Duff would be right in suggesting that there are reasons why we should answer for crimes such as murder or rape also to our fellow citizens, rather than only to the international community, he offers no reasons to support the conclusion that we are only answerable to the former and not to the latter.

IV. THE REJECTION OF THE GROUP-BASED REQUIREMENT
So far I have introduced two theses: first, crimes against humanity are those crimes that deny their victims the status of human being, where this means that they violate basic human rights; second, crimes against humanity are those that properly concern the whole of humanity, where this means that they are wrongs for which we are responsible (i.e. accountable) to the whole of humanity. While the second thesis relies on a conception of responsibility developed by Duff, the first is new and needs to be further developed.
Of course what is new is not the idea that crimes against humanity involve serious human rights violations. What is new is 39 This obviously does not imply that domestic crimes can be committed only against fellow citizens. Clearly citizens of state A can commit domestic crimes against citizens of state B -an example being an Italian citizen who steals a French citizen's car in France. The point is rather that this is a crime only to the extent that France has adopted a system of private property and a law against theft. In this case the wrongdoer can be answerable both to Italy (as per the 'active personality principle') and to France (as per the 'passive personality principle'): he is answerable to Italy as a citizen and to France as a guest (I discuss this problem in my 'Responsibility and Answerability in the Criminal Law'. For a discussion of overlapping responsibilities, see below, Sect. VI). the idea that it is on this aspect that we should focus in giving an account of what crimes against humanity are and in justifying their international prosecution -namely on the fact that violations of basic human rights constitute a type of wrong for which we are answerable to the whole of humanity. This goes against the orthodox view, which identifies in the contextual element the distinctive feature of crimes against humanity.
According to the orthodox view, the same crime (say torture, enforced prostitution or rape) can be either a crimes against humanity or a regular crime, depending on whether it is part of a systematic or widespread attack or not. If a Muslim woman is raped as part of a wider attack against her religious community, this is a crime against humanity; if a woman is raped on a Saturday night in London this is just a domestic crime. According to the view that I am presenting, on the contrary, any case of rape, torture, enforced prostitution and so on is a crime against humanity, no matter whether committed as part of a wider or systematic attack or not. These are crimes against humanity because they deny their victims the status of human being. Thus, my third claim, which follows from the two above, is that crimes against humanity are not necessarily committed as part of a widespread or systematic attack against the group to which the victim belongs.
If the view that I have developed so far is convincing, we have reasons to expand the scope of the notion of crimes against humanity, rejecting the idea that they should necessarily have a collective dimension or a policy element. If we accept that the key feature of crimes against humanity is that they constitute a serious attack on human dignity, which involves denying their victims' status of human being, there is no reason to consider crimes against humanity only those offences which are linked to a wider attack on a civilian population.
Take the cases of torture or rape. Both torture and rape constitute an attack on human dignity of the sort just described. They clearly do so even when these crimes present no connection whatsoever to any wider attack on a civilian population. When someone is tortured or raped, her status of human being is being denied, whether or not the crime is linked to a wider attack. The reason why philosophers such as Vernon and Luban consider only crimes that do present such link CRIMES AGAINST HUMANITY AND THE LIMITS OF ICL as crimes against humanity is that their accounts focus on the political dimension of these crimes. As I have acknowledged, there are good reasons to follow this line if our aim is to account for how the category of crimes against humanity is currently understood under international law. However, my aim here is a different one. I want to suggest that there are also good reasons to expand the scope of this category in order to include violations of individual human rights. My claim is that when these violations are so serious as to deny the humanity of the victims (i.e. when basic human rights are violated), wrongdoers are accountable for their crimes to the international community rather than simply to the national community. 40 In the remaining part of this paper I will develop further the nature and implications of my proposal, but let me first quickly address three possible objections to it. First, it might be argued that after all my view is also group-based, since individualized crimes can be ultimately understood as crimes that target the group of humanity. 41 This however would be misleading. Remember that group-based crimes are those that select the victims qua members of a group, and thus target simultaneously the victims and the group. 42 But in domestic cases of murder or rape, victims are clearly not targeted as part of a wider attack against the group of humanity. There simply is no attack against the group of humanity that these crimes can be seen as part of. 43 40 Again, this paper intends to provide neither a defense of the notion of human dignity, nor a defence of the notion of human rights. My argument is conditional on accepting these two notions as morally significant. This should not be a particularly controversial assumption, given the crucial role that they are normally recognized to play in international law and, particularly, as a foundation of the notion of crimes against humanity. Notice, however, that my view can also be accepted by those who reject the centrality of these notions, once reformulated along the following lines: whatever the empirical properties are on which our most basic rights supervene, we should have a special criminal category that recognizes that violations of these rights are acts for which the perpetrator is accountable to all persons. I am grateful to Jeff McMahan for pressing this point. 41 Thanks to Larry May for raising this objection. 42 See above, Sect. I. 43 It might be argued that when a woman is raped the group being targeted is the group of women, but this would be misleading too in my view. Although most rapists certainly have a mistaken conception of women's moral worth, they do not necessarily see their actions as part of a wider attack against women as a group, which explains why individualized cases of rape are not crimes against humanity according to the orthodox view. Some feminists might disagree and object that it is a shortcoming of the orthodox view that it does not recognize the group-based nature of this crime when perpetrated against women [see, for example, Catharine A. MacKinnon, 'Women's September 11th: Rethinking the International Law of Conflict ', Harvard International Law Journal 47 (2006): 1-31, at p. 22], but even if we were to concede this, it would not constitute an objection to my view. My contention is not that crimes against humanity cannot be group-based, but rather that they can be individualized. As long as we acknowledge the existence of at least some basic rights violations that are not group-based, my view does constitute a genuine alternative to the orthodox conception of crimes against humanity. Second, saying that some crimes should be considered crimes against humanity even when they do not include a contextual element is obviously not saying that there is no difference in seriousness between crimes that include this element and crimes that do not. Although we are accountable to the whole of humanity for all crimes that involve the violation of basic human rights, it might well be that those including a policy element and a collective element are more serious, and therefore deserve harsher punishment. This is why the view advanced in this paper is compatible with the idea that crimes that have a political dimension of the sort described by Luban and Vernonmost notably, mass atrocities perpetrated by state officials or by members of an organized political or military structure -are particularly serious. Crimes that include a collective element and a policy element do have a distinctive component, which should be considered when it comes to their prosecution and punishment. My contention is simply that they are not the only crimes that constitute a serious assault on the humanness of the victim and, consequently, not the only crimes for which wrongdoers are accountable to the international community.
However, let me also point out that the idea that crimes committed as part of a widespread or systematic attack are necessarily more serious than crimes that lack such a collective component is more controversial that we might initially think. In fact, there is at least one sense in which the reverse seems to be the case. As Mark Drumbl has convincingly argued, whereas normal cases of murder or rape are typically instances of deviant behaviour in the times and places they are committed, participation in collective atrocities, particularly by lower-level cadres and 'minor players', is often the product of conformity to social norms widely perceived to be morally binding (hence the 'banality' of this evil). 44 Of course this does not exculpate the perpetrators of these crimes, but suggests that normal cases of murder or rape typically exhibit a higher level of culpability to the extent that they are often (although admittedly not always) the product of an autonomous choice to pursue evil. Participation in collective atrocities, by contrast, is often the product of exceptionally strong social pressure and of 'a weakened sense of autonomy and independence '. 45 Finally, it might be objected to my view that expanding the scope of the notion of crimes against humanity in the way I suggest would be counterproductive, in that only recently, after long debates and diplomatic bargains, could an agreement finally be reached as to its definition. Trying to change this notion now might weaken this agreement, and thus it would be best to introduce a new category of crimes to cover individualized crimes that deny the humanity of their victims, rather than expanding the category of crimes against humanity already in place. 46 This is an important point, and I agree that the question of how to best implement my proposal deserves careful consideration. Still, even if it turned out that in fact there are pragmatic reasons to use a different label for individualized crimes that violate basic human rights, nothing would change as to the substance of my proposal: these crimes would be 'against humanness' no less than group-based crimes, and we would still have valid reasons to make them subject to international punishment.

V. AN IMPERSONAL DEMAND OF JUSTICE?
I have suggested that all crimes that deny their victims' status as human beings (i.e. all crimes that violate basic human rights of their victims) should count as crimes against humanity, and thus concern the whole of humanity. It follows that, at least in principle, every crime that violates basic human rights, no matter whether linked or not to a wider attack on a civilian population, should be dealt with by international criminal law. This however does not mean that all individualized crimes against humanity will have to be prosecuted by international courts. As I will argue in the next section, there are pragmatic as well as principled reasons why it is better to let national courts deal with crimes of this sort. An obvious pragmatic reason is that national courts are usually better placed to investigate and prosecute crimes committed in their states' territory, whereas the most important principled reason is the importance of respecting states' sovereignty. Still, at least in principle, these crimes are the international community's business, and this has important implications as to who can justifiably punish them and how. 46 Thanks to David Luban and Leif Wenar for raising this problem in conversation.

MASSIMO RENZO
Before we go on to discuss these problems, however, let me briefly compare the view I am advancing here with the position defended by Andrew Altman and Christopher Wellman in their article 'A Defense of International Criminal Law'. 47 According to Altman and Wellman everyone has a duty to prevent and punish human rights abuses, and this is why international tribunals are concerned, at least in principle, with any violation of fundamental rights that states are not willing or able to prosecute. It is only because of pragmatic and principled reasons of the kind I have just mentioned that international tribunals are ultimately justified in intervening only in the case of widespread or systematic violations. This view at first sight might look similar to the one I am suggesting, so it is worth briefly clarifying why the two are different.
As I have explained, the view I defend relies on a conception of the criminal law according to which crimes are those wrongs that properly concern the public. Crimes against humanity are to be understood as those crimes that properly concern the whole of humanity, i.e. as those crimes that involve wrongs for which we are accountable to all human beings and that are serious enough to be dealt with by international criminal law. Altman and Wellman, on the contrary, appeal to a more radical legal moralistic perspective that grounds the justification of criminal law in an impersonal demand of justice that the guilty be punished. The idea underlying their approach is that the proper function of the criminal law is to attain retributive justice and that any institution that successfully delivers retributive justice is ipso facto legitimate. 48 David Luban, who adopts a similar approach, provides an excellent formulation of this view: the legitimacy of international tribunals comes not from the shaky political authority that creates them, but from the manifested fairness of their procedures and punishments. Tribunals bootstrap themselves into legitimacy by the quality of justice they deliver, their rightness depends on their fairness.

CRIMES AGAINST HUMANITY AND THE LIMITS OF ICL
Thus, according to this view, any court able to realize retributive justice is in principle equally legitimated in dealing with any crime, no matter where it has been committed or who is involved in it.
While some of the conclusions I defend in this paper are similar to those reached by philosophers endorsing a legal moralistic perspective of the kind defended by Altman and Wellman as well as by Luban, it is important to notice that in my case these conclusions are reached by a completely different route. I do not appeal to an impersonal conception of justice, nor do I claim that tribunals can bootstrap themselves into legitimacy by delivering retributive justice. My suggestion of expanding the notion of crimes against humanity to include individualized crime is not grounded in the idea that human rights violations deserve to be punished and that anyone able to 'deliver champagne-quality due process and fair, human punishments' 50 has a right to do so. My theory is grounded in a specific answer to a question of responsibility; namely the question of what we are responsible for to all other human beings qua human beings. In considering crimes like torture, rape or enforced prostitution, I start with the question: 'who are we responsible to for committing these crimes?' I follow Duff in understanding this question as a question of answerability: the question of who we should answer to for committing crimes like murder and rape. But while Duff seems to assume that we are responsible for these crimes to the other members of our political community, unless they are part of a widespread and systematic attack against a civilian population, 51 I believe that we are responsible for them to the whole of humanity. So, while agreeing with Duff on the idea that wrongdoers should answer for these crimes to the public, I disagree with him on what should count as 'the public' here.
Whether or not this view is plausible, it should be recognized as clearly different from the view that demands of justice are impersonal in the sense that they are addressed to anyone who is in a position to fulfil them. The reason why my position, with respect to certain crimes, leads to conclusions similar to those reached by the impersonal view, is that I claim that the community to which we are responsible for crimes like torture, rape or enforced prostitution is 50 Ibid. 51 See above, Sect. III. the whole of humanity. My suggestion however, in a sense runs exactly in the opposite direction to the one advanced by Altman and Wellman as well as by Luban.
According to the impersonal view, the raison d'être of international criminal law is ultimately to punish those rights violations that states should punish, but that they fail to punish. As Luban himself notices, this means that the legitimacy of international criminal law ultimately depends on its capacity to occupy the void created by states' failures. In a world where states did their job, there would be no place for international criminal law, for in such a world all criminal law would be domestic law. 52 My account of crimes against humanity, on the contrary, rests on the distinction between wrongs for which we are responsible only to our fellow citizens and wrongs for which we are responsible to the whole of humanity (since they are so serious as to deny the status of human being to their victims). According to this account, even in a world where states did their job in punishing wrongdoers, we would need international criminal law. For even in this world crimes against humanity would not be the exclusive business of the domestic political community, but of a wider community: the community of all human beings.

VI. THE DOUBLE-LAYERED DIMENSION OF CRIMES AGAINST HUMANITY
The main worry about including individualized crimes in the category of crimes against humanity seems to be that if we do so 'there is no way to limit international crimes and prevent them from completely overlapping with domestic crimes'. 53 This objection however does not work against the view presented here. Since not every domestic crime violates basic human rights, thereby denying the victim's status as a human being, not every domestic crime can be said to concern humanity. The position I suggest does not run any risk of conflating international crimes with domestic crimes, because it starts precisely by distinguishing between those wrongs for which we are responsible to the domestic political community and those wrongs for which we are responsible to the international community.
Still, my view would be implausible if its adoption required that any individual case of torture, rape or enforced prostitution should be prosecuted by an international court. Although these crimes involve serious human rights violations and are among the most horrible forms of violence that can be inflicted on human beings, we normally think that they should be dealt with by national courts when they are committed in isolation from a wider attack on a civilian population. Thus, it is important to say something about whether, and how, the conception of crimes against humanity that I am suggesting would change our current understanding of the forum in which these crimes are to be prosecuted.
The first thing to notice here is that saying that individualized cases of torture, rape or enforced prostitution should fall within the scope of international criminal law is not saying that they ought to be prosecuted by international courts. The same is true of groupbased crimes already recognized as crimes against humanity under international law, many of which are prosecuted in national courts, and there is no reason to think that things should be any different in the case of individualized crimes. The important role that national courts should play in administering international criminal justice is indeed acknowledged in the statutes of the ICTY, the ICTR and the ICC, in spite of the differences between them. 54 I have already stressed that there are both pragmatic and principled reasons to let national courts deal with individualized crimes against humanity, at least in first instance. Firstly, national courts are usually better placed to investigate and prosecute crimes committed in their states' territory. 55 Secondly, states' sovereignty plays a crucial role in determining the structure of political life both at the national and the international level, and therefore ought to be respected whenever possible. Thirdly, the scope of crimes such as homicide, rape or torture is a contested matter. There are many permissible 54 The main difference is that while the ICTY and the ICTR have primacy over national courts (and thus can formally request any national jurisdiction to defer investigation or on-going proceedings at any stage), the ICC is based on the principle of complementarity (i.e. it acknowledges that national courts enjoy priority in the exercise of jurisdiction, but has a right to step in if they prove unable or unwilling to do justice). On the role of national courts in enforcing international criminal law see Drumbl,Atrocity,Punishment,and International Law,'International v. National Prosecution of International Criminal Law',in Antonio Cassese (ed.) ways to shape these offences, and there may be democratic, social and cultural reasons in favour of particular definitions in particular contexts. 56 There is, however, a fourth important problem that requires closer examination here, and on this I shall focus in the rest of the section.
The problem goes back to the distinction between wrongs for which we are accountable to our fellow citizens and wrongs for which we are accountable to our fellow human beings. It is this distinction, once combined with my suggestion that all violations of basic human rights fall within the second group, that seems to lead to the implausible conclusion that domestic criminal law should not cover any crime that involves the violation of basic human rights. In other words, domestic criminal law should only deal with crimes like theft, but not with crimes like rape, torture or murder.
To show why this is not the case I will try to further elaborate the relational conception of responsibility defended by Duff. Let me start by stressing that when I distinguished between the responsibilities we have to the members of our political community and those we have to all human beings I did not claim that the two classes are mutually exclusive. Following Duff, I have suggested that we have different responsibilities to different groups, and that these responsibilities depend for their content and direction on the role to which they are attached. However, this is not to say that each of our responsibilities is owed to one and only one group of people. Quite the opposite: most of the time our responsibilities are entangled, and the groups to which we are responsible overlap. This is partly due to the fact that our roles are not mutually exclusive (since they also tend to overlap to a great extent), partly due to the related fact that we might be accountable for our behaviour to different audiences inasmuch as our behaviour falls simultaneously within multiple descriptions.
So, for example, as a teacher I am responsible to my students for not preparing my lectures adequately, but I am also responsible to their parents and to the faculty. And not only to them. I am certainly responsible to all those tax payers whose money is partly invested to fund university programs. All these groups have a right to call me to account, although in different ways and for different reasons, in 56 To take an obvious example, any definition of rape will rely on a specific conception of sexual integrity and on a specific way of understanding consent.

CRIMES AGAINST HUMANITY AND THE LIMITS OF ICL
relation to my failing to deliver adequate lectures to my students. I am answerable to my students for providing them with a poor education; to the faculty for compromising the quality of the department; and to the other tax payers for wasting their money.
Notice that this is a case in which I am responsible for my behaviour as a teacher to different audiences. The responsibilities that I have as a teacher are to different groups (my students, the faculty, the tax payers), but they are all attached to one and the same role. Matters become more complicated when the same responsibilities are entailed by different roles that we occupy. Imagine that I claim from my department reimbursement for fictitious expenses or that I inflate my travel expenses. This is a wrong for which I am responsible to my university, which has the right to call me to account (and sanction me) for it. But it is also a wrong for which I am responsible to my political community (fraud is obviously covered by domestic criminal law). In this case I am responsible to two different groups for the same wrong in virtue of the fact that I occupy two different roles: the role of citizen and the role of university member.
It is important to notice that this does not make my obligation not to commit fraud any stronger. It is not as if members of the University of York have a stronger duty not to commit fraud against the University than non members. Rather, my membership adds a dimension to the structure of responsibilities of which I am part. In virtue of the fact that I occupy a role that others do not occupy, there is a further group of people to which I am answerable for the wrong in question.
The same kind of overlap is at stake in the case of the responsibilities that as members of the political community we have to our fellow citizens and the responsibilities that as human beings we have to our fellow human beings. While it is true that the responsibilities that we have to our fellow citizens depend on our being members of the same political community, this is not to say that these responsibilities cannot include any of those pre-institutional responsibilities that we have to all human beings as such. Any adequate theory of domestic criminal law will have to account for the fact that we are responsible to our fellow citizens for crimes like murder or rape. Duff's theory relies on a particular interpretation of the notion of liberal community, 57 but we do not need to follow him this way. We can be fair-play theorists, political contractualists, or adopt any other account of domestic criminal law. 58 The point is that for these crimes we are also responsible to all human beings. As in my example above, we are responsible to two groups (one of which is a subset of the other) in virtue of our memberships in both: qua human beings we are responsible for these crimes to all human beings; qua members of the political community we are responsible for them to our fellow citizens. This is to say that when we break domestic laws against murder or rape we are responsible first to the whole of humanity, for failing to discharge the duties we have qua human being; second to our fellow citizens, for failing to discharge the duties we have qua members of the political community. Again, we should not think that this adds any force to our obligations not to murder or rape. These obligations do not become any stronger than they would be in a pre-political condition because of the fact that we live in a political organization that declares murder or rape as a domestic crime. Rather, this adds a dimension to the structure of responsibilities of which we are part, in that we are now answerable both to the whole of humanity and to our political community for these wrongs. Both groups have a right to call us to account for them, although for different reasons.
Thus, the difference between these crimes and crimes like theft is the following: if we happened to live in a political community with no private property we would have no obligation not to take goods over which others claim control, and nobody would have a right to call us to account for that. If we lived in a political community where no prohibition against murder or rape were in place (if such a political community were possible at all), we would not be accountable to our polity for committing these crimes, but we would still be accountable to the whole of humanity. Of course, since humanity includes the members of our polity, strictly speaking we 57 See Duff, Punishment, Communication, and Community. 58 The account we adopt will partly depend on our account of political obligation, i.e. of what justifies the obligations that we have toward the members of our political community. My views on the subject can be found in 'State Legitimacy and Self-Defence', Law and Philosophy 30 (2011) would be accountable to the latter too, but only qua fellow human beings rather than qua fellow citizens. Our political community could not call us to account for these crimes.
This 'twofold dimension' of crimes against humanity is well captured by Antonio Cassese, who describes international crimes as follows: they constitute criminal offences in domestic legal systems … in that they infringe municipal rules of criminal law. In addition, they have an international dimension, in that they breach values recognized as universal in the world community and enshrined in international customary rules and treaties. 59 This double-layered dimension is precisely the one which is presupposed by the view I am suggesting, although according to my view the order of the two layers described by Cassese should be reversed. The underlying offence has not a municipal, but rather an international character, because crimes like murder, rape or torture are first of all crimes that deny the victims' status of human being. We are primarily responsible for these wrongs to all human beings. In addition to this international dimension, however, these offences also have a municipal dimension, which depends on the fact that political communities normally make their members accountable for them.
Finally, notice that although this further dimension does not affect the strength of our duty not to commit crimes such as murder, rape or torture, it does affect who should call us to account for them. While all human beings have the right to call me to account for these crimes, the particular subset of humanity that constitutes my political community has stronger grounds to do so because I am responsible to its members not only as a fellow human being, but also as a fellow citizen. This further dimension, which is added to the structure of responsibilities of which I am part because of my membership in the polity, explains why my political community has a stronger claim to call me to account when I commit individualized crimes against humanity that are also domestic crimes. This in turn provides a further reason why domestic courts should take precedence in punishing crimes of this sort. 59  Let me conclude by outlining the implications of the conception of crimes against humanity that I have presented. I have suggested that given the pragmatic and principled reasons considered above, states should enjoy precedence in the exercise of jurisdiction over individualized crimes against humanity. Still, since we are responsible for these crimes to all human beings, it is important that when they are prosecuted and punished by national courts, we see these courts as acting in the name of the whole of humanity, rather than just in the name of a particular political community. Thus according view, when an Italian court is prosecuting someone for a regular crime, the court is acting only in the name of the Italian polity, but when an Italian court is prosecuting someone for crimes such as murder, rape or enforced prostitution, the court is doing so also in the name of the whole of humanity.
A second implication is the following. We have seen that in the case of regular domestic crimes the defendant is responsible to his own fellow citizens and only to them. This is why if domestic courts fail to prosecute him, no other court can step in. The defendant is accountable to nobody but the members of his polity for these crimes. The case of someone prosecuted for crimes like murder, rape or enforced prostitution however, is different. Since the defendant is also accountable for these crimes to the international community, if his national courts fail to prosecute him, international courts (or other national courts claiming universal jurisdiction) are at least in principle justified in doing so. In other words, according to the conception presented here, all individualized crimes against humanity are covered by the complementarity principle. 60 This has important consequences in the foreseeable future, and could have even greater implications in the long term. As to the former, according to the current view, if a citizen of state A commits an individualized crimes against humanity in state B, and then flees to state C, he can only be punished by A or B. The role of C is only to extradite the wrongdoer if state A or state B asks for it. On the contrary, according to the view I am suggesting, state C has the right to claim universal jurisdiction and punish the wrongdoer, in case A 60 See above, footnote 54.

CRIMES AGAINST HUMANITY AND THE LIMITS OF ICL
or B fails to prosecute her (provided of course that C is in a position to collect evidence and guarantee a fair trial). Thus, wrongdoers would not be able to get away with crimes like murder, rape or enforced prostitution simply because their states (or the state of their victims) fail to prosecute them once they have fled.
Think, for example, of the case of an Italian citizen who commits a sex-crime in Thailand and then flees to France. It might happen that, although there is clear evidence that the crime has been committed, both Italy and Thailand fail to prosecute the wrongdoer (Italy might not be keen in prosecuting one of its citizens for a crime committed abroad, whereas Thailand might not have an interest in a prosecution that might discourage sex-tourism). Whereas according to the orthodox view the role of France is merely to extradite the wrongdoer in case Italy or Thailand requests it, according to my view France in this case could claim the right to prosecute and punish the crime.
To this it might be objected that giving France this right, in spite of the fact that Italy does not want to prosecute, might create diplomatic tensions between the two countries. But there is no reason to believe that these tensions would be any stronger than the tensions currently generated by refusals to extradite wrongdoers (a common phenomenon under international law). Since the latter do not seem to be posing an insurmountable problem to international stability, it is not clear why the former should. 61 Indeed, I contend that there might be cases in which it is not desirable to extradite the wrongdoer, even if his country (or the country of the victim) does ask for extradition. For instance, when the wrongdoer would be at risk of being persecuted or suffering human rights violations if extradited. In these circumstances my view would allow France to exercise the right to punish the wrongdoer, in spite of the fact that the crime has been committed elsewhere and that no French citizens were involved. 61 Indeed, arguably tensions are more likely to occur when state A refuses to extradite a citizen of B than when state A punishes a citizen of B for a crime committed elsewhere. In the first case, the desire of B to do justice is frustrated by A. By contrast, the second case is not necessarily a case in which B's desire not to prosecute the wrongdoer is thwarted by A. For while failure to prosecute certainly presupposes a lack of intention to prosecute, it does not necessarily presuppose an intention not to prosecute. (We could think, for example, of a scenario in which B simply does not have enough resources to investigate and prosecute all the crimes that B would like to prosecute, particularly when wrongdoers have fled abroad).
As for the long term consequences of adopting my theory, doing so would open up the possibility that one day, when institutions of international criminal justice will have greatly developed, individuals might be able to bring a case of rape or murder that has not been prosecuted domestically to an international court and ask for justice. If the current view of crimes against humanity is correct, this would be not only undesirable on pragmatic grounds, but unacceptable on moral grounds, since isolated cases of murder, rape or enforced prostitution that take place within the territory of the state are exclusively the state's business. Whereas if the view I suggest is correct, we should look favourably at such possibility (and even struggle to bring it about), since these crimes are not exclusively the business of domestic justice, although we currently have both pragmatic and principled reasons to treat them as if they were.
The conceptual change that I am suggesting certainly involves a drastic expansion of the notion of crimes against humanity as it is currently understood in international law. However, this should not necessarily be considered a problem. The notion of crimes against humanity is very much in the process of being defined and has already gone through several important changes. For example, the nexus with an armed conflict is no longer required. Similarly, the requirement that victims of crimes against humanity could only be civilians was eventually dropped. Finally, it is now accepted that crimes against humanity need not be perpetrated by actors acting on behalf of a government. Admittedly all these changes came from the tribunals' decisions, but there is no reason to think that philosophers should not contribute to redefining the notion of crimes against humanity and the normative implications that this notion should have. Indeed, this is a task that they have neglected for too long.