An Objection to Universal Jurisdiction: The Problem of Domestic Criminal Codes
Let us take stock. I have argued that humankind at large, via international institutions such as international criminal tribunals, has primary jurisdiction over war- related crimes; when international institutions cannot, will not, or ought not to punish, third-party states have the right to punish war criminals even though they are not connected by nationality or territory to those crimes. It is precisely because the right to punish is divorced from territoriality that the principle is aptly labelled ‘universal jurisdiction’. In effect, it consists in turning on its head the complemen?tarity principle as set out by the Rome Statute. According to the principle, you recall, sovereign states have primary jurisdiction over the punishment of wrongdoers, and the international criminal court has jurisdiction when states cannot or will not punish. On my account, by contrast, states have jurisdiction over those crimes only when international judicial institutions will not go or when it is better from the point of view of justice after war (for reasons mentioned a couple of paragraphs ago) that they should not go. In doing so, however, states would simply act on behalf of humankind—and not exercise a prerogative of sovereignty.
The principle of universal jurisdiction as I have expounded and defended it might seem vulnerable to a number of objections, only one of which I will address here.42 According to that objection, given that the rationale for the principle is that violations of human rights are of concern to humankind at large, there is no reason to exclude from its remit such violations as are committed in peacetime and in purely domestic contexts—for example Austrian citizen Josef Fritzl’s repeated crimes of rape and imprisonment against his daughter Elizabeth and the seven children she bore him. And yet to hold that, in principle, an institution like the international criminal court, as opposed to the Austrian state, has primary jurisdiction over him, is to undermine individuals’ jointly held right to punish wrongdoers for what are essentially domestic crimes albeit violations of human rights. Given that the Austrian state was deemed able to prosecute and punish him in compliance with basic standards of due process and fairness (which by all accounts it did), to regard international institutions as having ultimate jurisdiction in this case is tantamount to holding that the Austrians are not (jointly) immune from interference with the operation of their criminal law system. Put bluntly, it is a violation of their sovereignty over their own territory.
From a cosmopolitan point of view, the claim that the conferral of a given right stands in tension with sovereignty rights is not in and of itself worrisome. After all, sovereignty rights are largely instrumentally valuable as a way to bring about justice; and to the extent that they have value above and beyond that, they are severely constrained by the moral imperative to bring about justice across borders. That said, the objection sheds light on how best to understand crimes which fall within the primary jurisdiction of international institutions or the secondary jurisdiction of states. Members of a territorially bounded and sovereign community have the right to punish human rights violations, which they entrust to state officials, in so far as doing so is instrumental to the realization of punitive justice across borders. This allows for some leeway in the design of domestic criminal law. The distinction I drew in ch. 4 between the declarative and the specifying clauses of a peace agreement helps explain why. Declarative clauses—for example, a clause stipulating that belligerents will respect each other’s rights of self-government—merely state belligerents’ independently justified rights and duties vis-a-vis one another, and are binding on belligerents in virtue of a general duty to do what justice requires. By contrast, specifying clauses—for example, a clause specifying that elections will be 
held in belligerent community A within three years of the conclusion of the peace process—render more specific such rights and duties. By parity of reasoning, a clause in a criminal code which deems it an offence to (e.g.) have sex with someone who is not in a position validly to consent is declarative in so far as it merely states the independently justified obligation not to rape another person. But what counts as valid and invalid consent remains to be established. Within a reasonable range, there are broadly equivalent (morally speaking) conceptions of what counts as invalid consent which it is up to the relevant political community (be it a sovereign territorially bounded state, or a state within a federal union) to formulate. Not protecting children as young as nine from having sexual intercourse forced upon them by adults is a violation of those children’s basic human rights to sexual integrity and autonomy—such that anyone who commits such act ought to be deemed guilty of rape. Contrastingly, setting the age of sexual consent at seventeen rather than sixteen is not (I think) a moral imperative given that as far as we know, whether they are better able to consent at seventeen than at sixteen is a tough call. Put differently, having (consensual) sex with a sixteen-year-old is not a human rights violation; but nor is making it a criminal offence to prevent someone from having sex with a sixteen-year-old a human rights violation. It is therefore a prerogative of a political community whose members have the right and power, via their officials, to settle on the age of consent at sixteen or seventeen via their criminal code.
To see why we can thus block the objection to universal jurisdiction, consider someone—call him Andrew—who knowingly and willingly has sex with a sixteen- year-old. In Turkey, he would be guilty of rape; in England, he would not. The worry elicited by the objection is that Andrew would be liable to punishment under the Turkish penal code for having sex with a sixteen-year-old English girl in Britain and that Turkey would have a prima facie right to punish Andrew, even though neither he nor his partner is Turkish. But the principle of universal jurisdiction does not have this implication. For so long as England’s criminal code issues in a reasonable interpretation of the age at which teenagers can be presumed validly to consent to sex, its citizens have the protected power via their officials to enshrine such an interpretation in its criminal code and to punish Andrew accordingly; they also enjoy immunity from being interfered with, in this endeavour, by Turkey. In that sense, England’s power to punish Andrew can go hand in hand with an immunity from interference by other punishing parties.
Those considerations do not impugn my defence of the reversal of the complementarity principle. For it still remains the case that humankind has primary jurisdiction over all human rights violations. Thus, the principle of universal jurisdiction as I construe it does hold adults who knowingly have sex with a nine-year-old guilty of a gross human rights violation and liable to punishment at the hands of humankind at large. From a cosmopolitan point of view, to repeat, this is not to be regretted. For given the nature of their crime, those individuals are not wronged by the imposition of punishment; nor are their compatriots wronged by other states’ power to refuse to regard their refusal to punish those wrongdoers as dispositive. Moreover, in practice, the principle of universal jurisdiction so conceived does not entail that UN-mandated police squads should whisk off any suspected paedophile to The Hague or to any country whose criminal code explicitly prohibits paedophiliac acts and is properly enforced. There are in fact very good, morally directed pragmatic reasons to resist such a move. What the principle does entail, though, is that should such an individual be liable to prosecution and should it be possible to apprehend him without incurring such widely disproportionate costs (be they financial or reputational), punishment would not wrong him.
-  For other objections and rebuttal thereof, see Chehtman, The Philosophical Foundations ofExtraterritorial Punishment, 134—9.
-  Another way to put my point is this. (See, e.g., Duff, Punishment, Communication, andCommunity, 64ff.) Some offences are mala in se—offences irrespective of time and space. Others aremala prohibita—offences in virtue of being made so by a given legal system. Offences which are malaprohibita ought to be divided into two categories: specifications of mala in se (along the lines suggestedin the main text) on the one hand, and wholly new offences on the other hand. My claim here, in