Adil Ahmad Haque, Professor of Law and Judge Jon O. Newman Scholar, Rutgers School of Law, Newark
Alexander Greenawalt (“Sasha” to those lucky enough to know him) begins his article, International Criminal Law for Retributivists, with Mirjan Damaska’s famous question: “What is the point of international criminal justice?” This general question conceals a number of more specific questions:
- What is the point of international crimes? Why not rely on domestic criminal law to prohibit serious violence?
- What is the point of international criminal tribunals? Why not leave the prosecution of international crimes to national courts?
- What is the point of international criminal prosecution? Why prosecute international crimes committed in some conflicts but not others, some international crimes committed in a given conflict but not others, and some of those responsible for a given international crime but not others?
- What is the point of international criminal punishment? Why not deal with international criminals through lustration, compensation, or the exchange of apology and forgiveness?
- What is the point of international criminal sentencing? Why punish some international criminals more severely than others?
In principle, these questions might have different answers. The point of international crimes might be expressive, the point of international tribunals preventive, the point of international sentencing retributive, and so on.
As I read him, Sasha believes that all such questions rest on a false premise. There is no aspect of international criminal justice that has a single point; all serve multiple purposes and are subject to multiple constraints. In each domain, the relevant actors must simply weigh the various relevant considerations—retributive, expressive, or consequentialist—on both sides of a given policy issue.
On Sasha’s view, the fact that most international criminals deserve punishment is simply one good reason among others to create new international crimes; establish new tribunals; prosecute particular conflicts, crimes, or defendants; punish particular defendants; or set particular sentences. In each domain, there may be opposing, non-retributive reasons not to proceed—to conserve resources, to permit conflict resolution, to facilitate post-conflict reconciliation, and so on. The right course of action therefore depends on the overall balance of competing reasons. Since the balance of reasons is a contingent matter, Sasha concludes that retributivists are not committed to any particular position on any of the pressing policy questions facing the ICL regime. Retributivists might even support general amnesty for international crimes if that is what the balance of reasons supports.
In one way, Sasha is clearly right. The retributive duty to punish wrongdoers is not absolute. To that extent, the right answer to any ICL policy question indeed depends on the relative strength of relevant non-retributive considerations.
Nevertheless, I reject the view that moral responsibility for past wrongdoing is merely one good reason to punish among others, a reason that competes on equal terms with good reasons not to punish. On my view, only retributive reasons affirmatively justify punishment. The expressive value and consequential benefits of punishment contribute to the justification of punishment only indirectly, by offsetting the expressive disvalue and consequential costs of punishment. The expressive value and consequential benefits of punishment are not reasons to punish (“favorers”) but merely reasons to discount certain reasons not to punish (“defeaters of defeaters”). Retributive reasons to punish then compete with the remaining, undefeated expressive and consequentialist reasons not to punish. Put another way, retributive reasons compete, not with the “gross” costs of punishment, but only with the “net” costs of punishment. On my view, retributive reasons can still be outweighed, but they serve a unique and indispensible function in the justification of punishment.
Why do only retributive reasons directly count in favor of punishment? Ordinarily, it is morally wrong to intentionally harm other people. However, people can make themselves morally liable to suffer certain harms for certain purposes. Importantly, the reasons that make it permissible to harm someone also limit the amount of harm that may be inflicted as well as the purpose for which it may be inflicted.
For example, those who are morally responsible for an unjust threat to others (i) forfeit their right not to be (ii) defensively harmed as a causal means of averting that very unjust threat (iii) so long as the harm they suffer is not substantially greater than the harm they threaten. Notice that forfeiture, purpose, and proportionality share a common foundation. Similarly, on my view, those who are morally responsible for an unjust harm (i) forfeit their right not to be (ii) punitively harmed as a constitutive means of punishing that very unjust harm (iii) so long as the harm they suffer is not substantially greater than the harm they inflicted (discounted by their degree of fault).
In contrast, Sasha embraces “negative retributivism”, the view that those who are morally responsible for an unjust harm (i) forfeit their right not to be (ii) instrumentally harmed as a causal means of producing various intrinsic goods (iii) so long as the harm they suffer is not substantially greater than the harm they inflicted discounted by their degree of fault. This account of moral liability to punishment has always struck me as incoherent. We should instead accept that moral responsibility for past wrongdoing is the basis, purpose, and limit of moral liability to punishment.
Sasha at times suggests, not only that past wrongdoing is merely one good reason to punish among others, but also that past wrongdoing is not a particularly strong reason to punish. At one point, Sasha considers the possibility that retributive reasons may play only a “tie-breaking” role. In this passage, Sasha refers not to (imaginary) situations in which we know that the future costs and benefits of punishment will be on a par, but rather to (common) situations in which we cannot reliably predict the future costs and benefits of punishment. However, if we cannot reliably predict that the benefits of punishment will outweigh the costs, then we probably also cannot predict that the benefits of punishment plus the (apparently minor) retributive value of punishment will outweigh the costs. So even this tie-breaking role seems illusory.
ICL detractors like to point out that the Special Court for Sierra Leone (SCSL) cost over $200,000,000 and punished only 9 defendants of international crimes. Surely, the argument goes, we should not spend over $20,000,000 simply to give one wrongdoer the punishment he or she deserves—not when we could use that money to give many children the education they deserve, many patients the treatment they deserve, and so on.
Such arguments are particularly effective when they take on board the “consequentialist retributivism” with which Sasha sometimes flirts. On this view, deserved punishment is an intrinsic good that we should try to maximize. The retributive value of the SCSL is just the amount of deserved punishment it produced. Since the SCSL only produced around 350 years of deserved incarceration (at a cost of over $500,000 per year), it can seem like a waste of money.
Fortunately, consequentialist retributivism is false. I have argued elsewhere that deserved punishment is not an intrinsic good. Certainly, it should not be maximized. Consider the widely accepted principle that international tribunals should prosecute those “most responsible” for international crimes. Evidently, following this principle will not maximize deserved punishment. Since the “most responsible” are mortal, there is a physical limit to how many units of deserved punishment we can inflict upon them. Indeed, generally we can produce more units of deserved punishment by punishing a handful of minor participants than by punishing the mastermind they served. Yet such a distribution of deserved punishment would be unjust.
The distributive component of retributive justice remains largely unexplored. I suspect that retributivists generally should first seek to vindicate the rights of as many victims as possible and only then seek to vindicate these rights as fully as possible. Forced to choose, generally we ought to punish fewer of those responsible for more wrongs rather than punish more of those responsible for fewer wrongs. In general, it is more just to (partially) vindicate the rights of more victims rather than (more fully) vindicate the rights of fewer victims. There may be extreme cases in which full vindication of fewer victims can take precedence over partial vindication of more victims. But it is the vindication of rights, not deserved punishment, that retributivists should seek, not to maximize, but to justly distribute.
It follows that ICL ought to prioritize those “most responsible” for international crimes because generally this will vindicate the most victims of the most serious wrongdoing. Since the SCSL vindicated the rights of tens of thousands of victims of extreme wrongdoing, it was more than worth the cost. At a few thousand dollars per victim vindicated, this was justice at a bargain.
Sasha calls his view “good reason retributivism”. We could instead call it “one good reason retributivism” or “weak reason retributivism”. We could call my view “only reason retributivism” or “strong reason retributivism.” On my view, the point of international criminal justice is to ensure just retribution for serious wrongdoing. Non-retributive considerations will inevitably intrude but they will always remain, in an important sense, beside the point.