1 Introduction

The Ethics of War and the Force of Law: A Modern Just War Theory is the culmination of Uwe Steinhoff’s (2020) work on just war theory. Like much of his previous work, it is original and provocative. Steinhoff shows a refreshing willingness to challenge many of the central philosophical shibboleths that are taken for granted by just war theorists. He exposes and clarifies significant problems in existing accounts, and leads us back out of many blind alleys that others have taken just war theory down.

In this contribution, I focus on Steinhoff’s account of jus ad bellum. Steinhoff is generally sceptical of the standard list of jus ad bellum principles, presenting only three principles: right intention, just cause, and proportionality. He claims that the other criteria – last resort, reasonable prospects of success, and legitimate authority – are redundant since they can be reduced to proportionality assessments. First, I argue that we should go further than Steinhoff and hold that in the ideal morality of war all of the principles of jus ad bellum, including right intention and just cause, can be reduced to proportionality. Second, and notwithstanding, I argue that in the non-ideal morality of war we should retain the standard jus ad bellum criteria, including legitimate authority and last resort (which Steinhoff rejects), as separate principles.Footnote 1

2 Jus ad Bellum in the Ideal Morality of War

Steinhoff defends a stripped-down version of jus ad bellum. It amounts to the following: agents need to (1) know that they (2) have just cause and (3) that their war will be proportionate (p. 101). Consequently, he includes just cause, right intention, and proportionality, but excludes legitimate authority and last resort. In this section, I argue that, fundamentally, i.e. in the ideal morality of war, just war theory should be stripped down even more. Proportionality is the only criterion. We do not need separate criteria of just cause and right intention; they are redundant.

Before beginning, I should make two brief clarifications. First, the ideal morality of war is akin to what Jeff McMahan calls the ‘deep’ morality of war, where nonideal issues, such as abuse and other problems that come with non-compliance, are largely overlooked.Footnote 2 The ideal morality of war contrasts with what I call the ‘nonideal’ morality of war, which I consider in the second half of this essay (where I explain the distinction between the ideal and nonideal morality of war further). Second, proportionality is not simply a consequentialist assessment. Rather, it is a matter of weighing the goods and bad of launching war (and other options). These goods and bads comprise both agent-neutral considerations (such as how many lives will be saved by waging war) and agent-relative considerations (such as the comparative badness of doing harm compared to allowing it and whether the harms are intended). For instance, if a war will involve a significant amount of doing harm by the belligerent, such as in bombing raids, this will be weighed more heavily.

Let us start with the redundancy of just cause. According to Steinhoff, an agent has a just cause for waging war if there is an injustice, an emergency, or an agreement to wage war, and under the circumstances a proportionate war as possible (p. 60). However, this understanding of just cause could be more simply reduced to proportionality. On the view that I propose, agents do not need to have just cause to go to war; their war simply needs to be proportionate, that is, expected to bring about more (morally weighted) goods than (morally weighted) bads. Steinhoff’s potential conditions for just cause – an injustice, emergency, or an agreement to wage war – add nothing substantively to this because we already know that wars are permissible if they are proportionate. They merely provide an indicator of whether a war is likely to be proportionate. (To be sure, as I later argue, indicating when a war is likely to be proportionate is an important role played by the just cause criterion in the nonideal morality of war. But here I am focused on the ideal morality of war, where we already know that wars are permissible if they are proportionate.)

Some might hold that just cause depends on whether the aims of the belligerent are just. This potentially adds another consideration, akin to right intention, where those fighting need to intend to achieve certain goals. This might not be so easily reducible to proportionality (although, as an aside, I think it could be). Steinhoff (rightly) criticises the understanding of just cause as an aim, showing that this approach to just cause has several problems. Instead of being an aim, he argues, just cause is a state of affairs. However, when just cause is viewed as a state of affairs, it is more straightforwardly not doing any additional moral work itself. It simply indicates the state of affairs when war is likely to be proportionate.

It might be replied that wars cannot be proportionate in the absence of just cause, even according to Steinhoff’s extremely broad conception of just cause. This would be mistaken. Suppose that there is not an injustice, emergency, or an agreement to wage war. State A goes to war against State B to annex a remote, unpopulated island that State A (mistakenly) views as historically part of its territory. State A is concerned about facing civilian backlash of its war with State B, so it tries to mitigate some of the bad feeling by giving civilians in state B massive amounts of money, technology, and medicines, which has the result of hugely decreasing the number of deaths annually in State B. The war would bring about vastly more good than harm – it would be proportionate – but lack just cause.Footnote 3

To summarise, then, my argument is that Steinhoff’s account of just cause (as requiring an injustice, emergency, or agreement) adds unnecessary detail in the ideal morality of war. We can simply say that whether a war is permissible depends on whether it is likely to be proportionate; we do not need indicators at the ideal level of the morality of war. To see this more clearly, consider that there could be lots of potential indicators of the likely proportionality of a war. For instance, suppose that larger armies tend to be more likely to fight wars proportionately because they are more likely to be able to effectively defend the basic rights of those under attack. We could then add a ‘principle of large armies’ as part of jus ad bellum. This would help to indicate when wars are likely to be proportionate. But it would not be an independent condition and, like just cause, would be redundant in the ideal morality of war.

Having seen that just cause is redundant in the ideal morality of war, let us now turn to the redundancy of right intention. For Steinhoff, right intention is not, in fact, about intentions or motives. It requires the knowledge that the criteria will be met. This could also be reduced to proportionality. That is, right intention can be included in a mixed account of proportionality, where proportionality has both a subjective and objective element. On this view, agents should have a reasonable expectation that their action will be proportionate and their action should in fact be proportionate. To be sure, such a move might cohere with what Steinhoff argues, since he claims that right intention is “the subjective element of a justification for war” (p. 90) and that right intention is “not an additional criterion, but one that constrains the interpretation of the other criteria” (p. 294). But we can go further than Steinhoff and explicitly state that right intention is part of proportionality and so, like just cause, is redundant as a separate criterion.Footnote 4

Thus, we should see the principles of jus ad bellum as ultimately reducible to proportionality in the ideal morality of war.Footnote 5

3 The Nonideal Morality of War

There is much more to the ethics of war than the ideal morality of war. On its own, the ideal morality of war is too parsimonious and too abstract to be of much use. Saying that we need to simply look to proportionality to judge wars does not help much to assess whether wars are in fact likely to be proportionate and so permissible (or even obligatory). Suppose that you become an advisor to the South African President about whether South Africa should participate in an intervention to protect the Tigrayan people in Ethiopia. It would not be very helpful if, when providing your advice, you say merely that the President needs to look to whether a war would be likely to be proportionate. We need a more practical theory, with more detail, to help us get a better picture of the morality of actual wars. And this is where the nonideal morality of war comes in. This includes several nonideal principles to guide decision-makers. These are the jus ad bellum criteria, which provide strong indicators of when war is likely to be proportionate. Unless one adopts a simplistic consequentialist understanding, this is not simply about whether wars might save more lives or not. It is also about indicating how well wars are likely to fare according to the various agent-neutral and agent-relative considerations that go into proportionality calculations. The criteria can help to indicate whether wars in particular circumstances are, for instance, likely to involve large amounts of doing harm.

Accordingly, my aim in this section is to argue that we should generally maintain the standard jus ad bellum principles, including last resort and legitimate authority (which Steinhoff rejects). In addition, I argue that, again for practical reasons, we should have a more demanding notion of just cause than Steinhoff endorses.Footnote 6 To be clear, I do not claim that these principles should be viewed as absolute; they are instead useful guidelines.

Before progressing, I should say more about the nonideal morality of war and how it contrasts to the deep or ideal morality of war.Footnote 7 As John Rawls (1999) famously argues, nonideal theory is concerned with noncompliance with ideal moral principles and unfavourable circumstances. Technically, all of just war theory (including the deep morality of war) is an exercise in nonideal theory since there is some noncompliance (such as aggression) that merits going to war. Notwithstanding, there can still different degrees of ideality in just war theory, which reflects how much non-compliance and unfavourable circumstances are reflected in the approach. For the ideal morality of war, some, limited noncompliance is assumed, but many unfavourable circumstances are overlooked, such as lacking the means to accurately determine the liability of combatants or to assess accurately the proportionality of wars. It may also overlook the likely noncompliance with the idealised moral principles, such as whether a revisionist account of noncombatant immunity would be followed closely by those waging wars. By contrast, the nonideal morality of war reflects much more non-compliance and unfavourable current circumstances. It is concerned with the practical problems of achieving the idealised principles, such as the risks of abuse and epistemic and practical difficulties. It therefore offers a version of the morality of war that significantly reflects the nonideal circumstances that belligerents face.Footnote 8

In doing so, the nonideal morality of war aims to be more directly action guiding in the sense that it can be far more easily used by policymakers (and by those, such as NGOs, trying to influence policymakers) to assess the justifiability of potential wars. This is because the principles offered in the nonideal morality of war reflect the circumstances that they face, including their epistemic and practical limitations.Footnote 9 Just war theorists, especially certain revisionists, sometimes present their account of the ideal morality of war as the ethics of war that can be used to judge the justifiability of currently ongoing wars. The risks of this are that the principles in the ideal morality of war are far too thin and fail to provide sufficient guidance to leaders. There are also risks of abuse by actors, who may simply claim that their wars are likely to be proportionate; we need some way of being able to assess such claims, to hold them to account, with likely indicators of proportionality. This is where the standard principles of jus ad bellum come in.Footnote 10

Let us return to just cause. Steinhoff defends a notion of just cause that is potentially extremely permissive compared to other, more practical accounts. For Steinhoff, recall, an agent needs only to be responding to an injustice, emergency, or agreement. But if we want just war theory to be a useful guide for policymakers and for those who want to hold policymakers to account, we need something far closer to the standard version of just cause in jus ad bellum. The standard, more demanding conception of just cause, which requires that states (and maybe other actors) wage war only in cases of self-defence and in response to significant violations of basic human rights, is a far more useful guide to make practical judgments about real wars than Steinhoff’s conception of it as responding to an injustice, emergency, or agreement (or the rights violation view Steinhoff criticises). His account would seem to provide states with just cause to go to war in a very wide range of circumstances.

At one point, Steinhoff appears to concur. He argues that “in political discourse, we have good reasons to steer clear of a just war theory that is extremely quick in conceding just causes for war… Such a theory would be misleading and might have a poisonous and counter-productive effect on public and political discourse” (p. 89). Likewise, he argues that it is implausible that every rights violation is a just cause for war because this could reduce the restraining function of “the pursuit of war and would be misplaced in just war theory (which does purport to restrain war)” (p. 53). So, why not adopt a narrow view of just cause that limits it much more, such as to cases of serious aggression and serious rights violations, such as the mass violation of basic human rights?

Steinhoff seems sceptical of providing a list of just causes linked to serious rights violations for several reasons. However, most of these objections respond to accounts that hold that just cause is an aim and do not seem to be applicable if we agree with him that just cause is a state of affairs. Notwithstanding, there is a more clearly relevant objection that he makes. This is that such accounts make two proportionality judgements – one about just cause and another about proportionality – and so are problematic (p. 57). I do not see the problem. What these accounts do is provide a heuristic for when a war might potentially be proportionate. There are two steps here. If a war does not meet the just cause criterion, then it is very unlikely to be proportionate, and so we can probably rule it out. For instance, if a war is not in response to circumstances of serious external aggression or mass violation of basic human rights, it is very unlikely to be proportionate. But if a war does meet the just cause criterion, it has a chance of meeting proportionality. We then go onto the proportionality assessment proper. There are not, therefore, two proportionality judgments, but a filtering process provided by just cause. It filters out the cases very unlikely to be proportionate.

Let us now turn to legitimate authority. Steinhoff argues that legitimate authority is redundant because the necessary work is done by other just war criteria (p. 33). However, pace Steinhoff, it is still useful to have legitimate authority as a separate principle to help to constrain those considering going to war, to reduce the unjust resort to war by states and other agents.

Steinhoff rejects this. He first worries, in his discussion of what he calls the ‘universalisation principle’ for legitimate authority, that the legitimate authority criterion might rule out permissible private war (p. 43). I agree with this concern, if legitimate authority is understood to be an absolute principle. However, there is a more moderate, more plausible, version of legitimate authority, whereby it does not deny private action in every case. In certain, potentially exceptional circumstances, it permits exceptions to the general rule. But this does not mean that the principle of legitimate authority is not justified overall. So, we can say that, in general, there needs to be legitimate authority to go to war, but in exceptional circumstances this might be overruled. Indeed, this is the predominant view at least about humanitarian intervention: UN Security Council authorisation is morally required for legitimate authority, but on occasion unilateral action might be morally permissible.Footnote 11

The second objection to legitimate authority that Steinhoff presents concerns what he calls the ‘bad example’ argument (p. 45). This is the argument that legitimate authority should be always maintained because otherwise there could be a bad precedent set, even if good nonstate actors go to war. Against this, he argues that private war will not always set a bad precedent that tips the proportionality scales against nonstate violence. He also notes that the same objection is applicable to state violence. But, again, we do not need to adopt such a categorical position in defence of legitimate authority. We can simply say that the bad precedent effects may often tip the proportionality scales against nonstate violence, which is enough to generally rule out nonstate violence.

Steinhoff further argues that those wanting to reject private violence and to defend legitimate authority need to provide the evidence (p. 46) – and is sceptical that there is such evidence. But we do have the evidence from International Relations (IR) that allowing many more private actors to launch wars would lead to more instability. Take Wagner, the Russian private military company that has caused instability in sub-Saharan Africa and Ukraine. It is implausible to hold that the Russian secret services would have been able to do what Wagner has done to anywhere near the same extent – the plausible deniability has been crucial to its goals (until its use became much more publicly known in 2023).Footnote 12

Lastly, let us turn to last resort. Steinhoff argues that last resort is not an independent, necessary condition. But he does also note, intriguingly, that it is “a label for considerations regarding the available alternatives to war” (p. 98). But, if we grant this, then cannot we also grant that it is a useful ‘label’, that is, a useful principle? Indeed, he says that last resort “need to be taken into account” (p. 106, n. 72). We can see the principle of last resort as helping to maintain to policy makers that they should pursue other options.Footnote 13 We can use this language to be able to criticise, say, the US and the UK for not giving Hans Blix more time in Iraq. Seeing just war theory solely in terms of proportionality makes it too easy for those wanting to go to war to simply claim that their war is proportionate.

4 Conclusion

I will finish by linking my argument to the overarching aims of The Ethics of War and the Force of Law. In reply to me in a workshop on his book, Steinhoff said that he sees his objective as to advise philosophers. As such, we can see The Ethics of War and the Force of Law as trying to offer the best ideal morality of war. But if this is the aim, then we should go even further than Steinhoff in being parsimonious about the just war criteria in the ideal morality of war: just cause and right intention should be presented as clearly reducible to proportionality.

Yet much of Steinhoff’s discussion of jus ad bellum gives the impression that he is trying to do more than that. For instance, the opening line of the book says that it aims “to offer a modern just war theory that can give practical action-guidance” (p. 1). As we have seen, he rejects some of the criteria of jus ad bellum in the nonideal morality of war. As I have been arguing, if the nonideal morality of war is the focus, the understanding of jus ad bellum should differ from Steinhoff’s account. The standard jus ad bellum criteria should be maintained to indicate to policy makers and other actors whether a war is likely to be proportionate.