The claim that war-related crimes warrant punishment does not tell us who, exactly, of all the participants in such crimes, is liable to punishment. In this section, I argue that rank and file combatants, and not just leaders and generals, are liable to punishment for participating not just in what the laws of war labels war crimes and crimes against humanity, but also and more controversially, in the crime of aggression.
Officers, Combatants, Leaders, Citizens: The Problem of Participation in War-Related Crimes
As we saw in s.7.2, an agent W is liable to punishment for a wrongdoing W only if he has committed W (the conduct—or actus reus requirement) and if he was morally culpable for so doing (the fault—or mens rea—requirement).
War-Related Crimes and the Actus Reus Requirement
Let us apply that point to war-related crimes, starting with the actus reus. Some war-related crimes are committed by combatants each acting on their own, as when a combatant deliberately kills or tortures some innocent civilians or prisoners of war in his care. In those cases, it is easy to hold that the combatant meets the conduct requirement. However, far many more crimes are committed by combatants acting jointly, as when an entire platoon takes part in the deliberate killing of civilians, or when a regiment mounts an attack on an illegitimate target, or when a whole army unwarrantedly invades another country. More widely still, unjust wars are waged by armies with the assistance, forced or voluntary, of large sections of civilian populations. Finally, crucial decisions in war are taken by agents—the belligerents’ leaders—who do not themselves directly harm the enemy. The defence of punitive liability which I sketched earlier must thus account for two features of war-related crimes. First, they are committed by a multitude of agents whose respective individual contributions differ widely. Second, leaders seemingly commit a lesser wrongdoing and thus fare comparatively well—relative to ordinary combatants—at the bar of the conduct requirement (they are not the ones who kill and rape). Yet they are the most likely to intend the commissions of the harms which they are instigating or, at least, the most likely to be reasonably expected to know that those harms will ensue as a result of their decisions; they are also the most likely to be reasonably expected to know that those harms constitute war- related crimes. Accordingly, they fare comparatively badly at the bar of the fault requirement. By contrast, ordinary combatants commit the more grievous wrongdoings (they kill and rape) and fare comparatively badly—relative to leaders—at the bar of conduct, and yet they often do not intend to commit the larger harm of which many of their individual acts are a part; or they are the least likely to be reasonably expected to know both that they are committing the larger harm and that in so doing they commit war-related crimes.
Taken together, those two features of war-related crimes generate the following dilemma, in which one cannot honour both actus reus and mens rea. If one confers greater weight on conduct when determining liability to punishment, leaders will be punished rather lightly compared to ordinary combatants and civilians, which, given that they stand in a relationship of authority to them, seems unjust. But if one confers greater weight on fault, combatants will be punished rather lightly compared to their leaders, which, given that their acts (killing, raping, etc.) are more directly harmful to their victims, also seems unjust.
How then can we best justify the intuition that leaders, some combatants, and indeed civilians who assist them are liable to some form of punishment for war- related crimes? The task is relatively simple for crimes such as the deliberate targeting of civilians or the deliberate torture of prisoners of war. When those crimes are committed by several combatants jointly, they can each easily be described as taking part in a wrongful collective enterprise. Thus, combatants who killed civilians during the My Lai massacre are each properly described not as ‘each killing a few civilians’ but as ‘taking part in a massacre’, which might carry out a stiffer sentence than ‘ordinary murder’. Moreover, if the massacre is ordered by a commanding officer who did not himself take part in it, or if it happens without his orders but thanks to his negligent failure to intervene, it would be appropriate to hold him as accountable for it as his men: for to the extent that they acted upon his orders or that he could have stopped them but did not, his contribution to the death of those civilians is highly significant. Likewise, mutatis mutandis, with torture.
Those are micro-threats—which together form the macro-threat of the war itself. In view of the vast numbers of agents involved and how varied their individual contributions, establishing individual liability to punishment tout court, and to which punishment in particular, for that macro-threat is harder. Suppose that A’s army wrongfully invades B’s territory. Combatants and civilians on A’s side who contribute to that invasion are properly described as participants in that war crime. However, combatants actually carry out the invasion, whereas civilians’ participation typically takes the form either of ordering and planning the invasion (as high-ranking officials), or of providing the material means for its execution
(as taxpayers, weapons factory workers, etc.) Allowing a small part of one’s income to be directed, through taxation, towards the war effort is not as grievous a wrong as planning, let alone ordering, an invasion. At the same time, the former is still a wrong. Moreover, among combatantsA themselves, some actually kill a number of combatantsB who were not liable to be killed, while others, e.g., carry out reconnaissance and intelligence-gathering missions as preparation for the invasion. In this particular case, killing non-liable agents is more grievous a human rights violation than spying. Carrying out reconnaissance missions in preparation for an invasion, however, is wrong in a way that spying to satisfy one’s curiosity is not. And so on. 
Aggressive war is not the only macro-threat individual liability for which is hard to establish. So is genocide. As the ICTY recognized in its judgement against Goran Jelisic, an individual acting on his own who kills (e.g.) members of a particular ethnic group can be charged with genocided8 However, such a crime usually cannot occur without the participation of huge numbers of agents—from those who instigate it at the highest level, to those who carry out its constituent acts of killing, to the camp guards who oversee the prisoners before they are executed, to the clerks who maintain records, the train drivers who ferry victims to their final destination, the manufacturers who supply the gas, bullets, or machetes with which they are killed, the citizens whose taxes fund all of the above, and so on. In order to apportion liability to punishment among all those agents, we need to refine our understanding of a collective enterprise and participation therein.
In CW, drawing on Christopher Kutz’s account of complicity, I defended the following view: to the extent that a combatant or a civilian intentionally contributes in various ways to an unjust war and that he can reasonably be expected to know that other agents, combatants, and civilians are also contributing and share an understanding of what they are doing, that combatant or civilian can be said to share individual responsibility for the occurrence of the war—as can any other combatant or civilian in a similar epistemic and agential situation.!® Irrespective of their specific individual contributions, those agents can under those conditions be described as taking part in the war or phase of war. If the war or phase of war is unjust—in the sense that they consist in the violation of some other party’s human rights—all participants are liable to some form of punishment. The phrase ‘some form of punishment’ is crucially important. Participation in an unjust war, where participation involves some contribution, constitutes the conduct in virtue of which they are liable to punishment simpliciter. Which kind of punishment depends on what they actually did in pursuit of that collective venture.
-  See Larry May’s three-volume treaty on punishment for war wrongdoings (see fn 9), which ineffect seeks to provide a solution to that dilemma. His solution consists in privileging the mens rea to the detriment of the actus reus. I set out my reasons for disagreeing with him below.
-  For discussions of the liability of high-ranking civilian officials for the crime of aggression, withreference to the Nuremberg trials, see S. Levinson, ‘“Responsibility for Crimes of War”’, Philosophy &Public Affairs 2 (1973): 244—73. See also Walzer, Just and Unjust Wars, ch. 18; May, Aggression andCrimes Against Peace.
-  See Jelisic (IT-95-10-T), Judgement, 14/12/1999, art. 100. At http://www.icty.org/sid/7712(accessed on 4/6/2014). 1® Cosmopolitan War, 29—31. See C. Kutz, Complicity: Ethics and Law for a Collective Age(Cambridge: Cambridge University Press, 2000); as well as C. List and P. Pettit, Group Agency—ThePossibility, Design, and Status of Corporate Agents (Oxford: Oxford University Press, 2011). For a richand nuanced account of different kinds of complicity, see Goodin and Lepora, On Complicity andCompromise. The account of complicity provided by Kutz strikes me as more helpful than criminal lawdoctrines of complicity and, for that matter, of conspiracy (not least because those doctrines get miredin the difficulties of distinguishing co-principals from accomplices).