Journal of Controversial Ideas

Some news presentations for the general public that I have seen on the issue of possible genocide in Gaza consider how to determine intention. Often the proposals seemed inadequate. Determining intention is important because the Genocide Convention refers to “acts committed with intent to destroy, in whole or in part, a national, ethnic, racial or religious group as such.”1 The presentations I have seen make no use of a common test used by philosophers to determine intention in any context (not merely war). This surprised me and I believe it might be helpful to consider the test, using hypothetical rather than real-life cases. In addition, in the news presentations for the general public that I have seen there has been no consideration of a possible problem with the Genocide Convention itself. I shall try to correct this omission, again using hypothetical rather than real-life cases.

Before considering these matters, it may be helpful to consider some of the Convention in more detail.

Finally, intent should be distinguished from motivation (as it sometimes is not).4 For example, one may intend to kill another person from various motives, such as hatred or kindness (if the person is suffering) or from a desire to help others (e.g., because the person’s death is useful to acquire organs for transplantation for others). Further, if an act that harms is motivated by hatred, this need not mean the harm is intended. For example, suppose someone hates a group of people but on a given occasion is only intending to kill one of them and is afraid of being distracted from that by going after the others. However, they foresee some of the others will be killed. If they did not hate all the people, they would be more careful about harming others as a side effect. Causing indiscriminate harm can be wrong but not because it involves an intention to harm. It may rather be that the harm is indiscriminate because hatred makes there be no intention not to harm even if there is no intention to harm.5

The aim in this section is to show that an act can have multiple intentions and that the Counterfactual Test could help identify intentions; in particular, the fact that there is a military intention does not mean there isn’t also a genocidal intention.6

The Counterfactual Test is meant to distinguish between (i) an outcome that is intended and foreseen to occur and (ii) an outcome that is merely foreseen to occur, holding constant the probability of occurrence of the outcome. The test asks us to consider whether if, contrary to fact, an act would not cause a particular outcome, that would give an agent at least some reason not to do the act and/or to change the act to one that led to the outcome. (It is important to note that in the counterfactual scenario in which the particular outcome would not occur, an agent may continue to do the original act despite having some reason not to do it. This is because there are other reasons that still support acting. Hence, a willingness to continue doing the act in the counterfactual case in the absence of a certain outcome does not show the outcome was not intended in the actual case in which the outcome does occur.) If the answer to the Counterfactual Test’s question about the counterfactual case is yes – there is some reason not to do the act in the absence of the outcome – then the outcome when it actually occurs was intended; if the answer is no in the counterfactual case, then the outcome was not intended even if it was foreseen as certain to occur in the actual case. Here is a simple example: Suppose a state is to bomb its opponent’s combatants and it is expected that some civilians will also be harmed. Then the state’s agents find that, contrary to what was expected, the civilians will not be harmed. Would that give them a reason not to bomb or to bomb another target that was just as militarily effective but also harmed civilians? If the answer to these questions is yes, then they intended to harm civilians. If the answer is no – it makes no difference to the reasons they have for acting or for switching to another act – then the harm to civilians was merely foreseen (even with certainty) but not intended. In the latter case, the foreseen outcome is shown to play no role in the agent’s achieving their aims because the absence of the outcome would be of no significance to how the agent acts.

Applied to the determination of genocide, the question posed by the Counterfactual Test would be: If military actions failed to produce, for example, the destruction in whole or part of an ethnic group, would that be a reason not to continue that military action or to switch to a different military action that did help cause the destruction? If the answer is yes – it would be a reason – the destruction would be intended. If the answer is no, the destruction might still occur but it wasn’t intended even if it was certainly foreseen.

Therefore, one way to prove that there was genocide would be to show that, in a series of decisions about the course of military action, acts that were judged unlikely to help destroy an ethnic group in whole or in part were not undertaken, while alternative acts, no more militarily effective, that were judged to be likely to destroy an ethnic group in whole or in part were done instead. This way of proving genocide is in addition to proving it by finding cases in which some acts that helped to destroy the ethnic group had no military purpose at all. It has been said that when a country’s attack on group members can be characterized as an effect of the country’s defensive military strategy, the country can deny that the attacks are to destroy the group “as such.”7 But the example I have given shows that this is not true. This is because if the country switched its military strategy from one that did not destroy the group in whole or in part to a strategy, no more militarily effective, that did, then intent to destroy the group members “as such” could be demonstrated, even though their deaths were an effect of the defensive military strategy. The Genocide Convention does not require that genocidal intent be the sole intent of an act only that it be an intent of an act.8

It would be harder to find out whether an act’s not helping destroy an ethnic group provided a reason not to do the act or to switch to another act when the agent continued the original act because there were other reasons to continue doing it. This is because one rarely has access to a state’s reasons independently of the effect they have on whether an action is undertaken. Nevertheless, it might be possible to find out about reasons per se if there are records of how decisions were made about proceeding with military actions. Suppose it was found that when such decisions were being made, one reason raised for not proceeding was that there was little chance that the ethnic group would be destroyed in whole or part. However, in the decision-making process weightier reasons were raised for proceeding with a reminder that there was a chance that the ethnic group would be destroyed in whole or in part. If, contrary to what was expected, the group’s destruction occurred, the decision-making process could support the conclusion that the destruction was intended as such, despite its having been unlikely.9

In the absence of such a record of decision-making, the fact that there are, or at least that there are thought to be, many reasons to go forward with an act, makes it harder to prove that there was an intent to destroy the ethnic group as such. For example, suppose a state bombs military targets knowing this will not help achieve its military aims and these bombings will also destroy part of an ethnic group. Suppose it could be shown that the bombings had the intended non-military effect of extending the war so as to keep a wartime leader in power despite the bombing not being militarily useful. That additional wrongful intention makes it harder to show that the militarily useless bombings were also undertaken to destroy the ethnic group as such.

In this section I have tried to show that an act can have multiple intentions and that the Counterfactual Test could help identify intentions; in particular, the fact that there is a military intention does not mean there isn’t also a genocidal intention.

A problem with the Counterfactual Test is that it could sometimes incorrectly imply that there is an intention when there is none. For example, suppose that bombing a military facility will also cause civilian deaths and it is only because the community will be grieving these civilian deaths that they won’t quickly rebuild the military facility that was destroyed. If the facility were quickly rebuilt, there would be no point in bombing it to begin with. Hence, if it were known that the deaths would not occur, the bombing would be called off.10 According to the Counterfactual Test this indicates that the deaths would be intended because their not occurring would provide a reason not to bomb. I think this is the wrong conclusion. In this case, I suggest, the act of bombing is done because the deaths will occur but that is not enough to conclude that the bombing is done with the intent to produce the deaths. (Here is an example to help grasp the distinction: I may decide to go on living because I will not be in pain but this does not mean I decide to go on living in order not to be in pain. That is, not being in pain can be a condition of my going on living – I would not go on living if I were in pain – but that does not mean not being in pain is an aim I have for going on living.) In the bombing case, the deaths could be a side effect of the bombing that helps sustain (not produce) the intended military success already produced in destroying the military facility. That is, the deaths keep the facility from being quickly rebuilt. It could be shown that the deaths are not intended if the agents who bomb the military facility would not do anything additional (however easy, risk free, and undetectable) beyond what was necessary to destroy the military facility to ensure that the civilians are killed if, counterfactually, they found out after bombing that destroying the military facility did not also kill the civilians. In this counterfactual case, if they had intended the deaths, they would do something additional (that is easy, risk-free, and undetectable) to ensure the deaths that goes beyond having bombed the facility. If they would not do this, they did not intend the deaths when they bombed even if their expecting the deaths was a condition of their bombing the military facility.

Hence, in using the Counterfactual Test one must keep in mind the distinction between acting because of an effect that will be produced and acting in order (with intent) to bring about that effect. One must use an additional test to distinguish “because of” from “in order to.” (I did this by imagining whether the agents would do anything in addition to bombing the military facility to kill the civilians if it turned out after they bomb the facility that, contrary to expectations, bombing hadn’t killed the civilians).

Having discussed a test meant to identify intention and a problem with it, we now come to considering a possible problem with the Genocide Convention itself. This problem is its focus on intention. Taking this problem seriously could mean that looking for ways to determine whether an intention to commit genocide is present is not, after all, important, at least for deciding whether a crime has been committed.

Article I of the Genocide Convention says: “Genocide … is a crime under international law.” This means that, according to the Convention, genocide, as previously defined as an act whose intent is to destroy in whole or part certain groups as such, is a crime on its own independently of whatever else may be occurring. However, consider the following hypothetical case: Suppose soldiers setting off a bomb that will kill unjust enemy combatants is militarily necessary for the defense of their country but the bomb also causes proportionate civilian deaths. Setting off the bomb is justified despite the civilian deaths. Now further suppose that (i) the civilian deaths amount to eliminating a large part of an ethnic group and (ii) it is discovered that the only thing that prompted the soldiers to set off the bomb that kills the unjust combatants was the intention to eliminate a large part of the ethnic group as such. In particular, suppose the soldiers were lazy and would not have set off the bomb that kills the combatants as they should have for the defense of their country had they not discovered that setting off the bomb would fulfill their intention to eliminate part of the ethnic group as such. They may or may not also want to take advantage of the “cover” for destroying part of the group provided by killing the combatants without which the attackers would certainly face punishment. In this case, even though the act is “committed with intent to destroy, in whole or in part, a national, ethnic, racial or religious group as such,” it should not be criminalized. The act itself also does not seem morally impermissible even if the intention for the act is morally wrong. This is so even though the only intention for setting off the bomb is to eliminate a part of a relevant group as such. Setting off the bomb is not a crime or morally impermissible because there are objective factors that justify it, namely it is necessary to succeed in defending the country against unjust combatants without harming a disproportionate number of civilians.11 These justifying factors are present even though the soldiers see the factors as mere side effects of doing what achieves their purpose of destroying in whole or part an ethnic group as such.

It is important to see that this is not a case in which the fact that the action has the effect of defending the soldiers’ country makes it hard to determine that there is also a genocidal intent for the act (as in some cases discussed earlier). Even though it is clear that the soldiers’ intent is genocidal and that defense of their country is a mere pretext, their act (it is being claimed) should not be a crime and is not even a morally impermissible act. The reasons for this apply quite generally in non-war cases as well. Suppose that some innocent person is being criminally attacked on the street. A bystander was not going – and given the risk to them did not have a duty – to help stop the attack by the necessary means of injuring the attacker until they saw that the attacker was their enemy. They only do what has the effect of stopping the attack because it involves them injuring their enemy when it would otherwise be morally and legally impermissible to do so.12 Doing the act that is necessary to stop the attack on the innocent person is not a crime from which the bystander should refrain simply because it would be done with a wrongful intention. This is so even if it is the sole intention for the act. This is because the factors that justify injuring the attacker are present even if it is not for the sake of those factors that the bystander acts. The wrongful intention is not enough to make the otherwise permissible act either legally or morally impermissible. Or suppose the bystander sets off a smoke bomb that alone can allow the innocent person to escape but whose smoke also injures the attacker. That the bystander sets off the bomb only in order to injure his enemy (the attacker) foreseeing the escape as a side effect does not make his act a crime.

If this is so, then there seems to be a problem with the Genocide Convention. For it incorrectly implies that the lazy soldiers’ act is a crime solely because the act is intended to destroy part of the ethnic group as such. It would be interpreted to imply this even if there were also another intention for the act; in the case I have described there is no other intention. I think the intention for which the act is done should not make the act a crime, and it does not even make the act morally impermissible. This is consistent with the agent who does the act for the wrong reason being criticizable for doing so.13 It is also consistent, I think, with an act that is morally and legally impermissible on other grounds (for example, there is disproportionate harm to civilians) being a more serious wrong in virtue of the bad intention with which it is done.

That the intention to destroy part of an ethnic group as such does not make an act done with that intention a crime, or even morally impermissible, could also be shown even more simply. Consider a hypothetical case where the unjust combatants who would justifiably be killed in an attack are themselves a large part of the ethnic group. Again, suppose that soldiers who carry out an attack on the combatants are lazy and wouldn’t have done so to defend their country; their aim is to destroy in whole or part the ethnic group to which the combatants they attack belong. That they have a wrong intention for their act does not, I think, make their act of killing unjust combatants that is necessary for the defense of their country a crime or even a morally impermissible act.

It might be thought that if this is so, it is because the soldiers in these cases would otherwise fail through laziness to do their duty to defend their country from the unjust attack. However, suppose that the effort required to do what stops the unjust combatants would go beyond what duty requires, making what is done beyond the call of duty. Then non-lazy soldiers who nevertheless acted as was necessary to defend their country would do something supererogatory (morally and legally permissible but not required) and morally exemplary. Suppose the lazy soldiers would also do the act that requires the great effort beyond duty but only in order to destroy in part the ethnic group as such to which the combatants belong. Their act should be no more a crime than that of the non-lazy soldiers. However, given the lazy soldiers’ intentions, their making the great effort would not make their action morally exemplary, unlike what would be true of soldiers whose intention was only to defend their country.

I have argued that genocidal intent does not make acts that are not otherwise crimes into crimes.14 This is contrary to the what the Genocide Convention, which is concerned with designating a crime, says. If what I have argued is correct, the Genocide Convention could be improved if it were revised so that it does not criminalize acts undertaken with the intention to destroy, in whole or part, a national, ethnic, racial, or religious group as such when there are other factors that justify those acts even if the factors that justify the acts are, it is known, not those for the sake of which the acts are done, even in part.15 If the Convention is revised in this way, it will reduce the need to consider ways to identify the presence of a genocidal intention, at least for the purpose of declaring an act a crime. The focus will shift to finding out whether there are other factors that justify or rule out the acts aside from the intentions of the agents who do them. For example, we discussed a case in Section 2 in which a military strategy was switched from one in which no ethnic group would have been destroyed to an otherwise identical military strategy in which an ethnic group would have been destroyed. This was supposed to show that there was a genocidal intention along with a military intention. On this ground the Genocide Convention as currently phrased would prohibit the switch. However, the switch could also be prohibited if the Convention is revised as suggested because the switch violates the necessity condition on the use of military force in war. This is because the second military strategy is no more successful than the first in achieving legitimate military aims and it harms more civilians. In this case, some other factor besides genocidal intention can show that the switch is prohibited. I also suggested above that an act that is morally and legally impermissible on other grounds such as disproportionate harm to civilians might be a more serious wrong in virtue of the bad intention with which it is done. If this were so, finding out whether there is genocidal intent might be relevant to determining the seriousness of a crime even if it is not relevant to determining that there is a crime.

In this paper I considered how the Counterfactual Test might help in determining whether an act was committed with genocidal intent even when defensive military action was also involved. I then considered a problem with the Counterfactual Test and how one might avoid that problem. Finally, I showed how recent philosophical work on the relation between intention and moral and legal permissibility can bear on the correctness of the Genocide Convention itself. If what I have argued is correct, more attention should be paid to the soundness of the Genocide Convention itself than to how to meet the requirements for applying it (e.g., how to identify the presence of a genocidal intention).