A Journal of Philosophy, Applied to the Real World

Unjust Wars Worth Fighting For

School of Law, University of Warwick


I argue that people are sometimes justified in participating in unjust wars. I consider a range of reasons why war might be unjust, including the cause which it is fought for, whether it is proportionate, and whether it wrongly uses resources that could help others in dire need. These considerations sometimes make fighting in the war unjust, but sometimes not. In developing these claims, I focus especially on the 2003 Iraq war.


The laws of war, like the criminal law, have many ambitions. Here are three of the most important: to guide conduct, to guide courts in determining which conduct ought to be publicly condemned, and to guide officials in determining whom to punish. These ambitions sometimes come apart. We sometimes wish to condemn wrongdoers, but we do not wish to prevent them from acting wrongly, or to prevent others from doing the same thing. This seems paradoxical. If something is wrong, why not prevent and deter it? Surely it is better that the world contains less wrongdoing. Not always. Sometimes wrongdoing makes the world a better place: not better than it could be, were we able to encourage the wrongdoer to do what is best instead of acting wrongly, but better than it would be were the wrongdoing not to occur.


Racists Three Options. The lives of fifty white people and fifty non-whites are in peril. Derek, a racist, is the only person capable of saving anyone. He has three options—

Rescue no one;

Rescue only the whites, causing Derek to lose his foot.

Rescue everyone, causing Derek to lose his foot.

Derek, let us suppose, is not required to rescue everyone. He would not act wrongly if he did nothing, for rescuing anyone comes at a very high cost to him. Nevertheless, it would be wrong for him to rescue only the whites. If Derek saves the whites he must save the non-whites as well. This is because if he chooses to save the whites, he can also save the non-whites at no extra cost to himself. If Derek is willing to bear the cost of losing a foot for the sake of rescuing the whites, he must also rescue the non-whites. This is so simply in virtue of the fact that if he could rescue 50 people at no cost to himself, he must do so. If he has chosen to rescue the whites at the cost of a foot, he can rescue 50 people at no additional cost to himself. That is sufficient to require him to do so.1

Now suppose that we can’t get Derek to rescue everyone. Derek is inclined to rescue only the whites, and if he rescues the whites there will be no way of getting him to rescue the non-whites as well. What should we do? Well, whilst it is wrong for Derek to rescue only the whites, and it is not wrong for Derek to rescue no one, we would prefer it that he rescues only the whites to rescuing no one. In that case, we should encourage Derek to do the only thing that is wrong and rescue the 50 whites.

So here is an interesting fact about this case: rescuing only the whites and not the non-whites is the only wrongful act amongst Derek’s three options. But that is also the act that we ought to encourage him to perform. Sometimes, we have good reason to encourage people to act wrongly. We can understand this as follows. It is because Derek has the option of rescuing the non-whites at no additional cost that rescuing the whites alone is wrong. But whilst Derek has that option, we lack it. Our options are either to cause Derek to rescue the Whites or to rescue no one. Between those options, rescuing the Whites is permissible and preferable. The fact that Derek has an option that we lack does not render our act of ensuring that the Whites are saved wrong.

In the light of this, consider how the law ought to respond to circumstances like these, assuming that they sometimes occur. We face a dilemma. On the one hand, we owe it to the non-whites and their families to condemn Derek for his failure to rescue the non-whites. But on the other hand, we don’t want to deter those in Derek’s position from rescuing the whites. Passing a law warranting condemning and punishing Derek for his failure risks deterring people in Derek’s position from rescuing anyone. After all, given that it would not have been wrong for Derek to rescue no one, we can hardly condemn and punish Derek for rescuing no one. If Derek will be condemned and punished for rescuing the whites, and he is not at all inclined to rescue the non-whites, Derek may well prefer to rescue no one. But that is the last thing that we want. We would rather see people in Derek’s circumstances lose their feet if that helps to secure the saving of many lives. We would prefer it even more, of course, that those in Derek’s position would rescue everyone. But in a world of racists that may not be a result that we can secure. We might conclude, then, that whilst Derek is liable to be condemned and punished for failing to rescue the non-whites, we ought not to actually condemn and punish Derek, nor to indicate that we will do so by prohibiting rescuing only the whites in law.

Fortunately, dilemmas of the kind just discussed are rare in the real world of domestic criminal law. Not so in war. It is not at all uncommon that people act wrongly in declaring war, and in participating in wars, and yet that we ought not to condemn and punish them for their actions. We might even have good reason to assist those who wrongfully declare, organise, and participate in wars to help them achieve their aims. This provides one reason why the laws of war ought to reflect the morality of war only crudely: we do not always wish to deter people from acting wrongly, if permitting or encouraging their wrongful actions serves our ambitions to protect people against being harmed.

I will explore these ideas by focusing on the Iraq war that commenced in 2003. Let us call those who invaded Iraq the Invaders. Let us call those fighting against the invasion the Resisters. Many people believe it was unjust for the governments of the Invaders to declare a war against Iraq and to orchestrate the invasion. The fact, if it is a fact, that this was wrong does not establish that it was also wrong for individual Invader soldiers to go to war. I will explore a range of reasons offered for the view that declaring and orchestrating the war was wrong, and consider both a) whether this makes it wrong for Invader soldiers to go to war; and b) whether we ought either to have assisted or deterred individual Invader soldiers from going to war.

The war in Iraq might be thought wrong because it was conducted for the wrong reason—to get oil rather than to prevent serious rights violations. I explore this consideration in Section I, concluding that even if the war was wrong for this reason it does not supply a reason against participating in the war. In Section II, I consider whether the war was wrong because it was conducted with the intention of imposing democracy on a foreign country. I suggest that this is not itself a reason against going to war, and could not render it wrong. However, the war may have been wrong if it was conducted in order to establish unjust political institutions in Iraq that would better serve illegitimate western interests. Again, though, even if this rendered the war wrong, it would not supply a reason against participating in the war if the institutions imposed were less unjust than those they replaced. Section III is concerned with the question whether the war was wrong in virtue of the number of casualties it caused. This, I suggest, is the most powerful argument against the permissibility of the war. However, it also does not support the verdict that it was wrong for combatants to participate in it, as long as their own contributions are not disproportionate. Section IV considers an objection to the conclusions established in earlier Sections—that participating in unjust wars is typically wrong because doing so encourages future unjust wars. It is shown that this objection is not necessarily decisive, both because participating in unjust wars might not have these consequences, and because it is not always permissible to allow some people to die in order to deter the future deaths of others. Section V is concerned with a less familiar reason why going to war in Iraq might have been wrong—because the resources used for war could better be spent on non-military humanitarian aid. It is suggested that this may well have undermined the humanitarian case for war, but that this does not necessarily supply a reason for combatants not to engage in war. It may also not supply a reason to deter humanitarian wars, or participation in them.

Overall, there is no simple account of the permissibility of participating in unjust wars, or the wisdom of preventing such participation. If we are required to assess each individual act performed during a war on its own merits, and I believe that we ought to do this, whether a person acts wrongly in participating in a war depends on the particular contributions that he will make to the war. This can sometimes justify participating in unjust wars, including wars that lack a just cause, wars conducted for the wrong reason and disproportionate wars. And that is because the reasons that apply to those who decided to go to war, given the options available to them, need not be reasons that apply to potential participants in wars. Hence, sometimes it is permissible for individuals to join unjust wars, and sometimes we ought not to deter them from doing so even if doing so is wrong.

The War for Oil

The main reason offered by Invader governments for going to war in Iraq was preventive. Political leaders of the Invader governments claimed that going to war was necessary to provide security against the potential threat that Iraq posed to the international community, either through military action or by assisting terrorists. Few believe that Iraq posed an imminent threat to other nations. But that fact may not count decisively against war, at least in theory if not in law.2 What counts against this rationale for going to war is that war may well have been unnecessary to pursue the Invaders’ defensive aims, even given a charitable judgement of the Invader’s assessment of the likelihood that Iraq possessed weapons of mass destruction. At best, defensive aims could provide only part of the justification for going to war along with humanitarian goals.

Regardless of whether there were defensive grounds for going to war, though, some believe that the Invaders acted unjustly in going to war because they had the wrong intentions. Whilst Invader governments sought to claim that they had purely preventive aims, or pointed to a combination of preventive and humanitarian goals, many suspected that at least part of the reason why war was declared against Iraq was to secure cheap oil. Suppose that the Invaders went to war to secure oil rather than for defensive or humanitarian reasons. Would that render it wrong for the Invaders to declare war on Iraq?3

I) Intentions and Permissibility

It might seem obvious that if the Invaders went to war for oil, and not to achieve preventive or humanitarian goals, going to war was wrong. Not everyone, though, will draw this conclusion. For not everyone accepts that an action can be wrong in virtue of the bad intentions of the person performing that action. Some claim that the permissibility of warfare depends not on the intentions of participants, but rather on what they will in fact achieve. If the Invaders would prevent Iraq from attacking other countries and prevent the Iraqi government from killing their own civilians, these facts can render the war permissible, even if the Invaders did not intend to secure these ends, and were only motivated to secure cheap oil.

One prominent scholar who endorses such a view—that all that matters is what one will in fact achieve, and not what one’s intentions are—is Frances Kamm. For Kamm, if the Invaders went to war for cheap oil, we have good reason to criticize their characters and motivations. We also have reason to think that the meaning of the war was unappealing. But these facts do not make going to war wrong.

Kamm also considers the following question: does it make a difference to the permissibility of going to war that those going to war do so only on condition that they will secure preventive and humanitarian goals? For example, suppose that the invaders go to war in order to secure cheap oil. But they are willing to pursue this ambition only because securing it will also avert threats that Iraq would otherwise pose, and would prevent the killing of civilians by the Iraqi government. Does their greedy motive, in that case, render their action wrong? (see Kamm 2011, 119 – 124)4 She concludes that acting for the wrong reason does not render the declaration of war wrong. It is not wrong for the Invaders to go to war to secure oil as long as the preventive and humanitarian causes that they might have fought for were sufficiently important to justify the war. Furthermore, she argues that declaring war may have been permissible even if the warmongers would have gone to war regardless of the achievement of preventive and humanitarian goals.5

I doubt that Kamm is right to think that permissibility is independent of intentions. The fact that the war could have been justified on preventive humanitarian grounds, if indeed it could, is insufficient to render their conduct permissible. The fact that these ends would be secured might render the war justifiable—it has the potential to be justified. But the acts of the Invaders were in fact justified only if they acted for the right reasons. If their acts of war were not justified, they were wrongful.6

There is a great deal to say about the difficult and important question whether and when a person’s intentions can make a difference to whether her act was wrongful.7 Here I will only respond to the two arguments that Kamm offers for her view. She considers: (Kamm 2011, 119-130)8

Weden Oil Case: Suppose it is permissible for some country called Weden to begin a war against Germany to stop its invasion of Norway and also its genocide of Norwegians. However, neither Weden nor any other country is interested in starting a war for these purposes, but Weden knows that if it does stop the aggression and genocide, Norway will favour Weden in the sale of its oil resources. Getting such resources is not an aim that could justify Weden in starting a war, but that it will get resources is also not a reason against starting a war that stops aggression and genocide. Suppose Weden intervenes and stops aggression and genocide, but it does so only in order to get access to the resources that Norway will grant it, and it would not have intervened had it not had this aim.

Kamm believes that as long as Weden in fact stops the aggression and genocide, it permissibly goes to war. This is true even if stopping aggression and genocide did not motivate Weden in going to war. Furthermore, she thinks that it would not be wrong for Weden to go to war even if they would have been willing to go to war simply in order to get the resources in circumstances where they would not have prevented aggression and genocide (Kamm 2011, 128).

One potentially important difference between this case and the Iraq case is that Weden’s intention—to be favoured by Norway in the sale of oil resources – is not a bad intention in itself. In contrast, it is plausible that if the Invaders went to war for oil, they did so to unjustly secure cheap oil for themselves when others were entitled to control the resources. Perhaps it is important whether one’s intentions are inherently bad. (See, further McMahan 2009, 345).

But even acting on intentions that are not inherently bad may be insufficient to render Weden’s actions permissible. Kamm offers two arguments to support her view. We can call the first Scanlon’s Argument.9 The argument draws on the idea that the primary role that the judgement that an act is wrong plays is in our practical reasoning. When we decide that an act is wrong, we also decide that we have decisive moral reasons not to perform that act. Following Scanlon, Kamm argues that in most cases, when we are deciding whether we ought or ought not to perform some act, we do not consider the intentions with which we will act. We consider other properties of the act, such as whether the act will cause another person to be harmed, or prevent harm.

This argument, though, does not seem decisive. One reason is this. When we determine how to act in cases where there are powerful considerations that might militate in favour of and against acting, we do not normally begin with a blank sheet of paper, and address all of the considerations together. Rather, we have a particular thing in mind that we want to achieve, and we determine whether it would be permissible to achieve that thing given the reasons against doing it.

For example, on seeing the Iraqi government killing its civilians, we might wish to do something to prevent this occurring in the future. However, preventing this will be very costly both to our own combatants, and to Iraqi soldiers and civilians. The question that we face is whether we have sufficient reason to do what we want to do, given the costs. When we reason like this, we have already determined the intentions with which we will act, for we are already focused on the goal of preventing the loss of life. We do not need to reason about the intentions we will act with. We start with an aim, and then consider whether pursuing this aim is to be done all things considered. If we determine that it is, and we decide, we form an intention to pursue the aim.

In the Weden Oil Case, though, Weden deliberates in a quite different way. Its question is not whether it would be justified to secure its aim of preventing genocide and aggression by going to war, but rather whether it would be justified in securing its aim to secure cheap oil by going to war. But if that is the question, we should ask how it could possibly justify its aim to secure cheap oil given that killing people is necessary to secure the aim.

It is true, of course, that good things—the prevention of aggression and genocide—will in fact occur if it goes to war. Because of these facts Weden’s acts were justifiable—they were capable of justifying the decision to go to war. But Weden cannot appeal to these facts to justify its decision to go to war, given that these facts played no role in Weden’s decision. Justifying a decision involves showing that the considerations that featured in that decision rendered it permissible. And if that is right, Weden performs unjustified acts of killing. Their acts of killing are unjustified because the facts that could have played a role in rendering the decision to kill justified did not play such a role. This draws on the plausible idea that a fact that plays no role in guiding a decision cannot justify that decision. It is a fact that could have justified it, but as it played no causal role in it, it did not do so.10

In other words, the way in which Scanlon’s Argument assesses moral deliberation is very artificial. It does as though we begin with the question whether going to war is justified. But in deliberating about war we do not begin with the question whether to go to war. We begin with some other aims, such as the aim to prevent aggression or the aim to secure cheap oil. We then work out whether we would be permitted to secure these aims given the cost. Whether we are justified in securing these aims depends on how good the aims are, and how heavy the costs are in pursuing them. But this way of reasoning naturally brings intentions into the picture, for we are trying to determine whether we are justified in executing the intentions with which we begin.

Hence, if it was true that the Invaders went to war in Iraq only in order to secure oil, their acts of war were necessarily wrongful, for one cannot justify going to war to secure oil. This is true regardless of whether some other facts about the war, such as the fact that it prevented the Iraqi government killing civilians, would have been sufficient to justify going to war.

A second argument that Kamm offers can be called The Rights Argument. Kamm plausibly suggests that the normal reason why it is wrong to harm and terrorize people is that the rights of those people will be violated. However, Kamm suggests, people normally have rights that certain things be done to them. They do not normally have rights that what is done to them is done for the right reason.

Again, this argument misleads. First, notice that at least some of the people who will be killed in war have rights not to be killed. Even if a person has a right not to be killed, though, it does not follow that it is wrong all things considered to kill the person. Sometimes a person’s rights are insufficiently important to make it wrong for another person to infringe those rights all things considered. If this is true, infringing the person’s rights can be justified. But it will be justified only if the person infringing the rights is motivated by a consideration important enough to provide that justification.

Hence, it is misleading to ask whether a person has a right that others act on certain reasons. The right question to ask is whether a person is justified in infringing the rights of another. The reasons for which she acts will determine whether she is. If she is not, she has wronged the person whose rights she has infringed. This is not because the person has a right that others act on certain reasons. It is rather because those infringing rights need to be able to justify their actions, and whether they can do so depends on the facts that featured in their deliberations about whether to do so.

II) Participating in Wrongly Motivated Wars

Defensive or humanitarian reasons, that could have justified going to war, do not justify the war if these facts did not guide Invader decisions. Going to war may have been justifiable, in that case, but it was not justified. The question to be addressed now concerns the implications of this idea for the acts of soldiers who participate in the war.

Let us assume something that McMahan argues for persuasively: the fact that a soldier is commanded by her government to go to war cannot in itself normally make going to war justifiable (McMahan 2011, chs. 1 &2). If it is wrong to act in a certain way, acting in that way remains wrong even if one has been commanded to act in that way. After all, if it is wrong for me to act in a certain way, it is normally wrong for others to command me to act in that way. It would then be surprising if a command that it was wrong for another person to issue to me could transform what would otherwise be a wrongful act into a permissible act.

Nevertheless, if it was justifiable to go to war, the bad motivations of those declaring war may not make it wrong for soldiers to participate in the war. The reason is that the facts that wrongly motivated the Invader politicians in declaring war need not motivate individual Invader combatants who participate in the war. My government commands me to go to war for oil. If I decide to participate in the war, though, securing oil may not motivate me. If the war was justifiable on other grounds, say humanitarian grounds, I may be able to rely on those grounds in my own decision whether to fight.

For this reason, it is sometimes permissible to participate in a war that is fought for the wrong reasons, and hence to fight in a war that lacks a just cause, at least in one sense of that term. The idea of ‘just cause’, I should say, is somewhat contested. Whilst there was no just cause for the war in the sense that those declaring war did not act for the right reason, there may be a just cause in the sense that there was a cause that could have rendered the declaration of war just. There are people who are liable to be killed as a result of their actions.11 The ‘objective’ condition of a just war may have obtained, but not the ‘subjective’ correlate of it.12

To illustrate the idea that it is sometimes permissible to help a person to do something that they do for bad reasons, consider:

Wrong Reason: Harry launches a lethal attack on James. Debbie hates Harry. She attacks Harry, preventing Harry’s lethal attack being completed. However, she attacks Harry only because she wants to kill him, and not to defend James, whose death she is indifferent to.

Suppose her wrongful intention renders Debbie’s action wrong. However, it would have been permissible for Debbie to attack Harry in the same way to defend James. Now suppose that I can save James only by assisting Debbie. Doing so is permissible. If I assist Debbie, Harry may complain that I will have assisted Debbie in violating his rights. But even if what Harry says is true, his complaint lacks force. Harry is liable to be killed to protect James. It is permissible for me to help Debbie in virtue of this fact even if Debbie is acting for other reasons. I will be motivated by my desire to protect James, and protecting James is what I will achieve. If the motivation of the Invader governments was the only thing that rendered the Iraq war wrong, participating in the war may well not have been wrong.

It is, of course, unlikely that a war conducted for oil will be identical in all respects to a war conducted for defensive or humanitarian reasons. The motivations of Invader political leaders will affect the course of the war—for example, they may prioritize securing the oil fields over sparing civilian lives. The post war actions of the Invaders will also be different than they would have been had they been properly motivated. This may render some conduct, even some well-motivated conduct, in support of the war wrong.

These facts may still not make it wrong for individual soldiers to join the war though. Suppose that the war overall prevented the Iraqi government murdering, torturing and raping Iraqi civilians. I contribute to the prevention of these acts, and that is my aim. In doing so I facilitate other Invader soldiers in securing Iraqi oil fields as a side-effect. I may not have acted wrongly all things considered. I have a reason not to facilitate other Invader soldiers to act wrongly. But facilitating these wrongful acts might be justified all things considered, given the contribution that I will make to defensive and humanitarian goals.13

Now consider the position of soldiers who act in order to secure the oil, but whose actions contribute, as a side-effect, to the humanitarian goal. If I am right that motivations are relevant to permissibility, these soldiers act wrongly. And yet we may not wish to deter them from joining the war. This is so for similar reasons to those outlined in the introduction to the permissibility of encouraging Derek to act wrongly in Racist’s Three Options. These soldiers may act wrongly in joining the war to secure oil, but we would prefer them to join the war in order that they can assist us in pursuing our humanitarian goals. We may prefer this even if they are not acting in order to secure humanitarian goals. There may be no way of achieving our humanitarian goals without also facilitating the wrongful participation of soldiers in the war.

Hence, even though what these soldiers do is worthy of condemnation, given the reasons for their actions, we have strong reasons not to deter them from acting. A law that prohibited them from acting, were it abided by, may hamper the pursuit of humanitarian goals. It is typically more important that such goals are achieved than that wrongdoers are condemned. Although they are liable to condemnation and punishment for what they have done, we have some good reason not to condemn and punish them.

Imposing Democracy

One aim of the Iraq war was ‘regime change’. Many people think that this could not supply a reason to go to war. Some think this on the grounds that it is wrong to impose Western-style democratic institutions on another country. The imposition of democracy, on this view, is another form of imperialism. This view is sometimes defended on the (at best) silly basis that the Iraqi people have a cultural antipathy to democracy, and that democratic institutions may justly be imposed on a people only if they have the appropriate cultural tendencies. There is no reason to believe that Iraqis have a cultural antipathy to democracy, and little reason to think that the justification of setting up democratic institutions depends on people being culturally orientated to those institutions.

Some cultural tendency towards democracy may be required in order to ensure that democratic institutions function. Furthermore, imposition of democratic institutions may itself generate instability because those institutions are seen as serving the interests of outside agents and not the population itself.14 But if a people cannot create democratic institutions for themselves, the reasons against other people establishing them are, I think, quite weak.

In the context of Iraq, the view that we should be sceptical about the imposition of democracy is even more difficult to take seriously. The political institutions that the Iraqi people suffered under prior to the war can only be understood as imposed on them. Pre-war Iraqi political institutions do not represent the legitimate ambitions of the Iraqi people to shape their lives collectively. They certainly did not represent the ambitions of the Kurds and the Shiites, or of many Sunnis. Whilst imposed democracy may be to a degree unstable, and perhaps even illegitimate, because it is seen as serving the interests of outside forces, it is no less stable or legitimate than the tyranny that Iraqis had lived under for many years.

Furthermore, it is not obvious how any set of non-democratic institutions can be seen as representing the ambitions of the population in any reasonably large jurisdiction. We can imagine circumstances where a whole population endorses a non-democratic form of government. But in any reasonably large country there will be disagreement as to the proper shape of political institutions. Any set of political institutions in a country will be imposed on some by others. Suppose that it is important that political institutions represent the will of the people. Only democratic institutions will secure this value, because only democratic institutions can legitimately claim to represent the will of the people. It is difficult to understand what ‘the will of the people’ might be independently of the shaping of that will through political institutions, and it is difficult to see how that will can appropriately be shaped by non-democratic institutions in the real world. When non-democratic institutions govern a country, the vast majority of the population have no role in shaping the political arrangements that they live under.

Even if non-democratic institutions did reflect the will of a majority of Iraqis, we would have no reason to endorse institutions that foster abuse of basic rights. The invaders could justify the imposition of democratic institutions simply on the basis that these institutions would do better than pre-war Iraqi institutions in protecting and promoting basic rights. As the Iraqi people have an enforceable duty to protect and promote these basic rights insofar as they are able to do so, it need not be shown that they wish to protect and promote them. Of course, there is a question whether democratic institutions serve these ends well, but it is difficult to imagine that even defective democratic institutions will have a worse record on this score than pre-war Iraqi institutions in the long term, at least given appropriate resources to provide peace and security.

Perhaps it might be argued, as Michael Walzer does, that the value of a set of political institutions depends on whether the population that is governed by them has secured them themselves, and hence that imposed democracy lacks value. (Walzer 2008, ch .6) But whilst the idea that the value of a set of political institutions depends to some degree on how those political institutions have emerged is reasonably attractive, it is very difficult to believe that this always provides a decisive reason against imposing democracy on other countries. A set of institutions that are imposed on a people may be less valuable than the same set of institutions would be had they been fought for and won by the people themselves. If there is some prospect of a population developing decent institutions for itself, this provides a reason against intervention. Whether this provides a decisive reason against intervention in fact, though, depends on the prospects of the population developing decent institutions of their own. It is, of course, difficult to know whether, and how quickly, such institutions might have been established in Iraq without military intervention.

Let us explore a more sophisticated version of the view that going to war to impose democracy on the Iraqi people was wrong. Although the political institutions that are imposed by the Invaders on the Iraqi people are likely to be preferable to the institutions imposed on them by the Resisters, it might be argued, they are still unjust. The Invaders imposed unjust institutions on the Iraqi people in order to secure their influence in the region, and to help them to have greater access to Iraqi oil.

Let us suppose, as is plausible, that this is true. Even if it would have been permissible to oust Saddam Hussein to create just democratic institutions in Iraq, it might be argued, it was wrong to oust Saddam to impose unjust institutions on the Iraqi people. That is true even if those institutions are preferable to those that existed in Iraq prior to the invasion.

It is true that we have reason to condemn the Invaders for imposing unjust institutions on the Iraqis for their own interests, if they did so. But this fact would not provide a powerful reason against fighting on the Invader side in the war. When considering the question whom to support in a war, we unfortunately typically lack the option of supporting a group with even reasonably just ambitions. In most conflicts, no side fights for anything close to justice. When determining whether to fight and whom to fight for, the right question to ask is not normally which side is acting justly, but rather which side is likely to be the best of a bad bunch.

None of this is to say that imposing democracy could itself counts as a just cause for war. Perhaps if the imposition of democracy would prevent many deaths that would otherwise be caused, fighting for democracy might be justified. But democracy might not in itself be important enough to be worth killing people for. The imposition of democracy on Iraq may not count as a just cause because no one is liable to be killed simply for running a country in a non-democratic way.

Even if it is true that imposing democracy can never be a sufficient just cause for going to war, though, it may nevertheless provide a legitimate reason to go to war, and this may contribute to an overall assessment whether to go to war. For example, suppose that two states are perpetrating human rights abuses that would be sufficient to render it permissible to go to war against either. This could be done at very minimal cost to the invading state. It is feasible that democratic institutions might be set up in one of these states post war but not the other. That factor can determine which state ought to be attacked. Hence, even if democracy cannot provide a just cause for war, it can provide a legitimate motivation of those engaging in war.

Overall, I doubt that the attempt to instil democracy in Iraq was a powerful reason either for the Invaders to fight, or not to do so. The idea that we ought to spread democracy through the world by military intervention is problematic, not because there is something suspect about democracy, but because war is typically a disproportionate and ineffective way to pursue the aim of establishing democratic institutions. But at the same time, the fact that the Invaders claimed to be fighting for democracy does not provide a powerful reason against participating on the Invader side.

The Casualties of War

A better reason to think that the Iraq war was unjust is that it caused far too much death and destruction, both to combatants and especially to non-combatants, to justify the aims the Invaders were pursuing, or even aims that they could have pursued.

It is worth noting at the outset that a large number of the civilian casualties that occurred during the war were caused by the fact that the Resisters, when they were attacked, went to war with the Invaders. When the Resisters defended the territory they controlled, the number of civilians killed may well have become too large to justify the war—both because of deaths caused by Invaders and Resisters, and because of deaths caused by insurgents. Were Saddam Hussein immediately to have surrendered before numerous civilian casualties were caused, as he ought to have done, the war would have been closer to being justified. For, were that to have occurred, the Invaders could have improved the political institutions of Iraq without causing as much loss of life.

Even if the Invaders acted unjustly in going to war, the Resisters also acted unjustly in defending themselves. If the number of civilian casualties caused was the main reason why the war was unjust, the injustice of Invader acts of war was in virtue of the injustice of the Resister acts of war that Invader acts predictably gave rise to.

I) Responsibility for Casualties

Assuming that I am right that it was wrong for the Resisters to defend themselves against the Invaders, it might be tempting to conclude that the main responsibility for the deaths of non-combatants in the war lay with the Resisters. It might be thought to follow that these deaths cannot make it wrong for the Invaders to go to war. This is hard to believe. In deciding whether to go to war, significant weight must be given to lives that will be lost, even if those lives will be lost as a result of other people’s wrongdoing.

A more plausible view treats deaths caused by intervening agents as less significant in the decision whether to go to war than deaths caused directly as a side-effect by combatants on the just side. Even this more plausible view is, I think, false. Sometimes, if I act in a certain way, and that leads other people to act wrongly, my responsibility for the harm caused by the wrongdoing is diminished. For example, suppose that I marry a person of another race. It might be predictable that racists will riot as a result. Even if this is predictable, I am not heavily responsible for the harm caused by the racists. Responsibility for the harm they cause lies with them and not with me.

Even though this is true in the example just offered, I doubt that it is more generally true that we can evade responsibility for deaths that others wrongfully cause as a result of our actions, even in part. If I act in a way that creates a new opportunity for others to act wrongly, and they act wrongly as a result, I bear very significant responsibility for the harm caused by their wrongdoing. For example, if I leave your front door unlocked, and it is predictable that thieves will steal your property as a result, I must take a great deal of responsibility for the theft. I cannot claim, in that case, that as the thieves are primarily responsible for the theft, I am significantly less responsible.

To see this even more clearly, consider:

Which Route: A nuclear reactor is about to explode. If it does, several thousand lives will be lost, including everyone in this example. I can get to the reactor and prevent the explosion by taking either the low road or the high road. If I take the low road, I will dislodge a boulder, which will crush X. If I take the high road, I will dislodge a different boulder, which a villain will then use to murder Y.

I ought to take either the low road or the high road. Which ought I to take? On the view that deaths that result from my actions due to intervening wrongdoers are much less significant in my decisions than deaths I directly cause as a side-effect, I ought to take the high road. X’s death, on this view, is more significant than Y’s in virtue of the fact that I will directly cause X’s death.

Now compare two other views. Some might think it makes no difference which road I take, and I should flip a coin to decide. Others might think that I ought to take the high road, in virtue of the fact that Y will be murdered if I take the high road, whereas X will not be murdered if I take the low road. I think that the least intuitive view of the three is the first view. I am unsure which of the two other views are to be preferred.

For a similar reason, I suspect that the Invaders must bear a significant amount of responsibility both for the deaths they cause, and also for the deaths that result from at least some of the wrongful actions that occurred as a result of the war. For example, deaths caused by insurgents arose because these insurgents were provided with a new opportunity to kill as a result of the war. Overall, civilian casualties that arose during war count powerfully against the permissibility of going to war, even if they were caused by the wrongful acts of others, who bear full responsibility for their wrongdoing.

II) Supporting Decisive Victories

Suppose that the humanitarian and defensive benefits of going to war could not justify the number of deaths that the Invaders caused. I think that this was the main reason to object to the Iraq war. If I am right, how should we assess the contributions that individual combatants made to the war? It might seem that if it was disproportionate for the Invaders to go to war, it was also typically disproportionate for individual Invader combatants to contribute to the war. This does not follow.

After the war began, the effects of the acts of any individual combatant on the war were typically modest. Suppose that I was deciding whether to participate in the war. If I did so I might at most have hastened victory. I could not have prevented the war from occurring, nor could I have prevented an Invader victory, by refusing to participate.

It is quite likely, then, that the main difference that acts of any Invader combatant, taken individually, made to the war was to hasten its end. Each combatant may have helped to ensure that victory was more decisive. Each combatant made no difference to whether or not victory was achieved. Given this, it seems that there are powerful reasons for each Invader combatant to participate in the war once it has commenced. The reason is that we ought to prefer a swift and decisive Invader victory to a long drawn out war. The number of casualties will only increase if the war is long and drawn out.

It may be argued, in response, that I have been presuming that it was certain that the Invaders would win the war. Perhaps at no point in the war was this certain. Even if this is true, it may not provide a decisive reason against participating in the war. First, putting aside considerations of deterrence and related considerations, we should prefer a drawn out Invader victory to a drawn out Invader defeat. Even if too many lives were going to be lost in the war to make it justified, we would prefer that if these lives are lost, they are lost in the course of securing the defensive and humanitarian aims that might have contributed to a justification of the decision to go to war.

Secondly, if it is permissible to act to support an Invader victory where that victory is certain, it is probably also permissible to support it where that victory is highly likely. The issue of risk is complicated, and is beyond the scope of this paper. I doubt that the possibility that the Invaders might have lost was sufficient to render it wrong for Invader soldiers to participate in the Iraq war.

Against this it might be argued that although each combatant’s actions cause the war to end more swiftly and decisively, and that is preferable, we should nevertheless doubt that each combatant is permitted to go to war. A reason for us to doubt this is that each combatant on the Invader side will kill particular individuals whom they are responsible for killing. Consider a particular combatant, who kills some non-combatants as a side-effect of his contribution to the war. The non-combatants that he killed may have survived the war had he not participated in it.

A full analysis of this complicated issue is beyond the scope of this paper. I doubt that it provides a decisive reason against Invader combatants going to war. One reason is that, even though the Invader combatant is responsible for particular deaths that he causes, each Invader combatant may have improved the ex ante prospects of survival of all non-combatants, including those who they actually killed. The Invader combatant can then rightly claim that they have no reason to believe that their actions will make this worse for those who might be killed during the war. Furthermore, the particular deaths that any combatant causes might be a proportionate side-effect of the actions that the combatant performs in the war, given the contribution that the combatant makes to ending the war swiftly. If a particular combatant helps to end the war more swiftly, lives will be saved overall. This may be sufficient to justify the deaths of any non-combatants that come about as a side-effect of his contribution.

Preventing Future Unjust Wars

In the previous section I offered some reasons for us to prefer a decisive, to a drawn out, Invader victory. Let us explore some objections to this view.

First, the war-making capabilities of the Invaders will be reduced if the war is drawn out. This will diminish their ability to engage in other unjust wars. Secondly, a drawn out victory might deter the Invaders from engaging in further unjust wars. As the war is drawn out there are more casualties and that will tend to discourage the Invaders from engaging in future wars, not least because it will be more difficult for political leaders to motivate citizens to accept going to war. It follows, some may argue, that there is some reason to fight against a country that engages in an unjust war even if we know that this country will win the war, and even though a greater number of civilian casualties will be caused as a result.

It is not at all unlikely that unjust aggressors will be more likely to refrain from aggressing in the future if they discover, through the harsh experience of war, that it is difficult to achieve swift and decisive victories. For example, had the Invaders scored a more decisive victory in Iraq, perhaps war with Iran would have been more likely. It would have been easier to persuade the American public that war with Iran would be just. And the US would have had greater resources to prosecute such a war. Yet war with Iran would almost certainly have been unjust.

Of course, the effects of a drawn out war may prevent the Invaders from engaging in just wars as well as unjust wars. There probably have been conflicts which Invader countries ought to have intervened in—the conflict in Rwanda being perhaps the most obvious example. Their experience of the drawn out post-war struggle against insurgents in Iraq will likely deter interventions in such conflicts in the future. I will consider whether humanitarian wars are typically just in a moment. But even if they are not we may have very good reasons to hope that they occur and to advocate for them if they do.

Furthermore, even if it is a good thing that the Invaders do not engage in future wars, because the wars they will engage in will tend to be unjust, it does not follow that it is permissible to refrain from joining the war for this reason. Much depends on why the Invaders will fail to engage in future wars.

There is little objection to a potential combatant failing to intervene because the military capacity of the Invaders will be reduced if the war is drawn out. But if the reason why it will be more difficult for the Invaders to wage war in the future is that people are horrified at the greater number of Iraqi civilian casualties, or Invader casualties, things are different. It seems wrong to allow an increase in the casualties to occur in the Iraq conflict because allowing these casualties to occur will horrify the general public, making them less likely to support future conflicts. To refrain from participating in a war for this reason would amount to acting on an intention that some suffer as a means to save others from the horrors of potential future wars. This would have been wrongfully to use the deaths of those in the Iraq conflict as a means to protect others.

To reinforce this conclusion, compare Warren Quinn’s Guinea Pig case, in which I refrain from treating a person with a serious illness in order that I can learn more about the illness. Suppose that if I do this I can prevent other people from contracting the illness, and hence reduce the number of people who suffer from the illness overall (See Quinn 1993, 177 and Foot 2002, 92). Refraining from treating a person for this reason seems wrong. For a similar reason we cannot justify making or allowing one war to become more horrific in order to prevent horrific wars in the future.15 Were we to do so, we would wrongly exploit the suffering of some people in order to prevent others from suffering the horrors of war.

Humanitarian Wars16.

Recall the possibility that the war in Iraq was justified for humanitarian reasons. Suppose that the number of civilians and soldiers killed during the war was insufficiently large to render the war disproportionate given the number of lives saved. Suppose that this is so for the reason that the Resisters would have killed many people had they retained power in Iraq. I do not claim that this is true. But it is not completely implausible that it is true, especially given not only the tendency of Saddam Hussein’s regime to kill Iraqi civilians, but also its tendency to engage in wars that were unjust and unnecessary, such as the lengthy war with Iran conducted between 1980 and 1988.

It may seem that if the number of lives saved was sufficiently greater than the number of people killed, the Iraq war was just. This does not follow. It is one thing to claim that the number of people killed is proportionate to the number of lives saved by killing them. It is another thing to claim that acting in this way is permitted when we compare the option of going to war with other things that we might have done with the resources that we expend on war.

I. Saving Victims of Injustice?

Evaluating humanitarian wars is a complicated affair. There are many considerations that help to determine whether such wars are justified. One is the cost that soldiers and other citizens of the intervening country will bear for the sake of preventing serious injustices in other countries. Another is whether those suffering from the humanitarian crisis welcome the intervention. Still another concerns whether it is permissible to kill innocent soldiers and civilians of the country to be intervened in to prevent the wrongdoing occurring. None of these considerations decisively rules out humanitarian war in every case.17.

Nevertheless, humanitarian wars may well often be unjust for the following reason. A state must ensure that it fulfils its humanitarian duties generally. Humanitarian wars must be evaluated in that context. When we determine whether it is proportionate to go to war we should consider the claims that others have on the resources that are used in fighting the war—what we might call ‘the war chest’.18. There are many others in dire need of these resources, and the money ought to be spent on them.

Consider the fact that the government of any country must spend a certain amount of money on humanitarian aid of various different kinds. When we identify that amount of money, we take into consideration special responsibilities to avert threats that the country is responsible for creating, for example because it exploited citizens of other countries, used their natural resources, imposed unfair burdens on them through trade agreements and international organisations, or created environmental threats through polluting the atmosphere. This will leave a certain amount of money to spend on humanitarian aid that is not directed to addressing problems that the state under consideration is responsible for creating.

There is a limit to how much this country may spend. Spending more would violate the rights of its citizens to these resources. I make no claims about how large this amount is, though I am sure that it outstrips the foreign aid budgets of most Western states. How should the state spend these resources?

Let’s start with the obvious suggestion – that it should spend its resources to save the greatest number of people that it can from the worst kinds of suffering. This suggestion is wrong. For in an effort to save the greatest number, the state might also kill some people. There is a more stringent prohibition on harming people than on failing to rescue people. Broadly speaking, then, the state should give priority to methods of saving people from harm in ways that do not harm others.

This provides a powerful reason to think that it is typically wrong to engage in humanitarian wars. The state should rather spend its resources preventing and curing illness and disease, for example by providing mosquito nets, clean water or access to essential medicines. Doing these things saves a great number of people from death and serious illness, killing almost no one. Benjamin Valentino estimates that provision of medical aid is around 3000 times as cost effective as military intervention with respect to lives saved per dollar (see Valentino 2011, 60). As we will see, even if this radically overestimates the difference, there are decisive arguments against humanitarian intervention.

The most familiar argument against the view that we ought to spend all of the resources that are devoted to humanitarian causes on humanitarian aid rather than humanitarian intervention is as follows. We have a special obligation to prevent systematic wrongdoing by a state against some group of people. It might be argued that it is much more important to prevent wrongful killing by tyrants than it is to cure disease. Many people feel the force of this complaint when presented with tyrants who systematically attack their own people, as we have seen recently in both Libya and Syria.

There are two responses to this argument. First, it is not very plausible that there are much stronger reasons to prevent harmful wrongdoing than there are to prevent the equivalent non-wrongful harms. Here is an example that helps to demonstrate this. Suppose that a number of children have contracted HIV. An evil person has intentionally infected some of these children. Others have contracted the disease by misfortune. I have a stock of retroviral drugs. It seems abhorrent to push those who have been infected wrongfully to the front of the queue for treatment (see McMahan 2010 and Tadros 2011, chs. 5 & 6) . It is even more abhorrent to treat fewer people who have been infected as a result of injustice when we could treat more people who have been infected without the relevant injustice.

What this example suggests is that people’s rights to an equal chance of being saved from harm are quite robust. Seemingly important differences in how the harm will come about are insufficiently important to justify departing from a practice which gives each person a fair chance of survival.

Secondly, even if we have more powerful reasons to prevent deaths that arise as a result of injustice than to prevent deaths that arise naturally, almost all humanitarian disasters arise from injustice. Almost all of the people who contract serious illnesses and diseases, or who cannot feed themselves, are victims of injustice. Their states, or others states, have perpetrated injustices against them, for example by failing to create just institutions, by violating obligations to provide humanitarian aid, by exploiting the natural resources in their countries, or by propping up dictators.

Perhaps it might be argued that humanitarian wars have long-term advantages over other kinds of humanitarian aid. Through humanitarian wars the belligerents may be able to establish just institutions in currently unjust countries, improving the lives of citizens a great deal for the long term. If this is possible, it provides some reason to conduct humanitarian wars. It is worth noting, though, that the record of establishing just institutions through war is not especially good. It will often be just as likely that just institutions will emerge without anyone engaging in a humanitarian war. If a barbaric regime can be toppled and replaced with a democratic regime at relatively little cost of life in circumstances where this is unlikely otherwise to occur, humanitarian war may be permissible. But these circumstances are rare.

Alternatively, it might be argued that humanitarian wars have a powerful deterrent effect. Dictators will be less likely to perpetrate humanitarian abuses if they realise that doing so will bring with it the risk of invasion. This suggestion would have more force if resources were available to ensure that humanitarian intervention would follow predictably, swiftly and decisively in response to humanitarian abuses. We are, at present, a long way from this situation. Global policing of humanitarian abuses is much less effective than domestic policing of crime, and this is unlikely to change in the short term.

Perhaps it might be argued that even if it is better not to engage in humanitarian wars, a state does not act wrongly if it engages in such wars. Doing so may not be best, but it is not wrong. One reason why this might be thought true is that states have a right to shape their own identities. They can choose to shape these identities by picking the disasters that they should respond to.

I agree that, within some limits, states are permitted to shape their own identities by spending their resources on some projects rather than others. However, it is very difficult to justify killing many people, and saving the lives of far fewer people than could be saved, on this basis. The rights that citizens have to live in a state that can shape its own identity by making choices about which people to save seems insufficiently powerful to justify killing non-liable people, which is what inevitably happens in humanitarian wars.

II. Should Humanitarian Wars be Prohibited?

In the light of this discussion, do we have strong reasons to prevent humanitarian wars from occurring? I doubt it. Although these wars are often unjust, given other things that the belligerents could achieve, we may sometimes have good reason to permit or promote them.

Notice that on the account that I have given, humanitarian wars are typically unjust on what we might call a ‘comparative’ rather than on an ‘internal’ basis. It is wrong to perpetrate them because others have claims on the resources used to pursue them, and not necessarily because the lives that are lost cannot be justified by securing humanitarian goals.

Given this, whether we have reason to prevent countries from engaging in humanitarian wars depends on whether the belligerents will pursue better options if they are prevented from engaging in humanitarian wars. International organisations ought to prevent states from engaging in humanitarian wars if doing so will lead them to spend the same resources on other kinds of humanitarian aid. For in that way, these international organisations could ensure that many more lives are saved. It would be wrong to prohibit internally proportionate humanitarian wars if doing so would not lead to the same resources being spent on humanitarian aid.

It is often easier to motivate people and states to support and engage in humanitarian wars to prevent genocide, or other seriously wrongful acts, than it is to motivate people to provide resources for other kinds of humanitarian aid (see McMahan 2010). Perhaps we ought not to deter humanitarian wars, in that case, even though it is wrong to engage in them. Deterring humanitarian wars would not motivate states to do what they ought to do – to spend the resources that they would have spent on humanitarian wars on other humanitarian projects. Given that, we might reasonably prefer that they engage in some kinds of humanitarian intervention. Hence, perhaps international law ought to permit ‘internally just’ humanitarian wars even if they are ‘comparatively unjust’. It also ought not to deter or condemn those who participate in such wars.


A criminal law that prohibits most serious wrongdoing has relatively few costs if it is abided by. This is because in the domestic context a person’s wrongful acts normally make things worse than they would have been had they not acted. The kinds of dilemma that I have been exploring in the context of war are rarely significant in a domestic context.

In war it is often very difficult to determine who is acting wrongly. Even if we could do this, it is often difficult to determine whether we have reason to prevent them from so acting. Much more stringent, authoritative and well-enforced laws to prevent unjust wars and wrongful actions during war would have great advantages. They would prevent a great deal of unjust killing. But they would also have severe costs. They would hamper the ambition of those with humanitarian aims to achieve their goals. I do not argue that, all things considered, it is wrong to think that we should typically deter unjust action in war. But we should also be aware of the costs in doing so. Sometimes unjust wars are wars worth fighting for, or at least with.

Acknowledgements: I am grateful to the audience at the Ethics, Law and Armed Conflict Conference 2011, where a version of this paper was first presented, and to Helen Frowe and Jeff McMahan for helpful comments.


Buchanan, A ‘Justifying Preventive War’ in H Shue and D Rodin Preemption: Military Action and Moral Justification (Oxford: OUP, 2007).

Bazargan, S ‘The Permissibility of Aiding and Abetting Unjust Wars’ (2011) 8 Journal of Moral Philosophy 513.

Fabre, C Cosmopolitan War (Oxford: OUP, 2012).

Foot, P ‘Morality, Action, and Outcome’ in Moral Dilemmas (Oxford: OUP, 2002).

Kamm, F Ethics for Enemies: Terror, Torture, and War (Oxford: OUP, 2011).

Kamm, F Intricate Ethics: Rights, Responsibilities, and Permissible Harm (Oxford: OUP, 2007).

McMahan, J ‘Just Cause for War’ (2005) 19 Ethics and International Affairs 1.

——— ‘Intention, Permissibility, Terrorism, and War’ (2009) 23 Philosophical Perspectives 345.

———‘Humanitarian Intervention, Consent, and Proportionality’ in N Ann Davis, R Keshen, and J McMahan Ethics and Humanity: Themes from the Philosophy of Jonathan Glover (Oxford: OUP, 2010).

——— Killing in War (Oxford: OUP, 2011).

——— ‘Proportionate Defense’ (2013-14) 23 Journal of Transnational Law and Policy 1.

Neff, S War and the Law of Nations (Cambridge: CUP, 2005).

Parfit, D ‘Future Generations: Further Problems’ (1982) 11 Philosophy and Public Affairs 113.

Quinn, W ‘Actions, Intentions, and Consequences: the Doctrine of Double Effect’ in Morality and Action (Cambridge: CUP, 1993).

Scanlon, TM Moral Dimensions: Meaning, Permissibility, Blame (Cambridge, Mass.: Harvard UP, 2008).

Tadros, V The Ends of Harm: The Moral Foundations of Criminal Law (Oxford: OUP, 2011)

——— ‘Responses’ (2013) 32 Law and Philosophy 241.

——— ‘Punitive War’ in H Frowe and G Lang How We Fight (Oxford: OUP, 2014).

Thomson, JJ ‘Physician-Assisted Suicide: Two Moral Arguments’ (1999) 109 Ethics 497.

Valentino, B ‘The True Costs of Humanitarian Intervention’ (2011) 90 Foreign Affairs 60.

Walzer, M Just and Unjust Wars: A Moral Argument with Historical Illustrations 4th Edition (New York: Basic Books, 2006).

——— ‘The Politics of Rescue’ in Arguing about War (New Haven: Yale University Press, 2004).

1. For a similar analysis of a closely related case, see (Parfit 1982, 131)

2. For a good discussion, see (Buchanan 2007)

3. It is not essential to my argument here that intentions can be attributed to collectives, though I think it plausible that they can be.

4. Jeff McMahan is unsure whether the right intention is required to render a war just. (2005, 5)

5. She draws on the more general discussion in (Kamm 2007, ch. 5)

6. To explain the distinction between what is justified and what is justifiable more clearly: a pro tanto wrongful act is justifiable if a fact supplies a sufficient normative reason for a person to perform that act. It is justified if this fact was a motivating reason for the person who acted.

7. I offer a more complete defence of the idea that intentions are relevant to permissibility in Tadros 2011, ch 6, 7, and ‘Responses’) and (Tadros 2013, 282-91) .

8. See also her discussion of Baby Killer Nation(Kamm 2011, 79-85).

9. As Kamm follows the argument in (Scanlon 2008). This kind of argument was also suggested in (Thomson 1999)

10. This argument leaves open the possibility that it is sufficient for Weden to act on condition that the relevant facts obtain to render the decision permissible. If it acts on such a condition, the relevant facts do play a role in shaping the decision to go to war. I have some doubts about this being sufficient, but I leave the problem aside here.

11. McMahan (2011, 27) emphasises this feature of the requirement of just cause

12. Stephen Neff (s2005, 50-51) suggests that these features of just war were traditionally distinguished. Just cause (justa causa) was the name given to the objective element. This is odd terminology, though, as the fact that some objective criterion was fulfilled could neither cause a war nor be a cause fought for.

13. For incisive discussion of this issue, see (Bazargan 2011, 513)

14. Michael Walzer (2004, 68-69) gives this as a reason against intervention

15. That is not to say that the means principle is without exception. For implications for going to war for punitive reasons, see, further, (Tadros 2014)

16. Some of the arguments in this section are similar to those of Jeff McMahan (2013-14)

17. For an excellent discussion of these factors and others, see (Fabre 2012, ch. 5) . Fabre does not focus on the issue that I explore below.

18. For further analysis to show that this is the right way of understanding comparative proportionality questions, see (Tadros 2011, ch.15)