1 Targeting ‘Moral Fundamentalism’

In The Ethics of War and the Force of Law, Uwe Steinhoff takes on, in his typically energetic, forthright and provocative way, the ‘revisionist’ movement in just war theory, and sketches an alternative moral account of war in which convention plays a more visible role.Footnote 1 My focus in this short piece will be on the main elements of his positive theory of war that are outlined, in somewhat compressed form, in the final sections of his book (EWFL, pp. 214 − 48).

Now Steinhoff rarely if ever mentions ‘revisionism’ without scare quotes, since one of his complaints about this body of writings on war associated with Jeff McMahan, David Rodin, and others, is that it is less revisionist than it thinks: many of the arguments associated with contemporary revisionists can actually be detected in the writings of early modern writers in the just war tradition such as Suárez and Grotius. I shall not engage with questions of historical attribution in this article. My concern is with the truth or plausibility of these arguments, not their source. To understand the broad orientation of Steinhoff’s account, however, we do need a fuller understanding of what he is arguing against.

Steinhoff’s account is opposed to what he calls ‘moral fundamentalism’: the project of uncovering the ‘deep morality’ of war as governed by the same moral principles and rules that govern ordinary life (EWFL, pp. 214 − 15). Moral fundamentalism constitutes the key doctrinal commitment and the methodological underpinning for revisionism. For fundamentalists, there are no distinctive principles or rules of war. The principles applying to war also apply to everyday life whenever violence erupts in it. Revisionists hold that the moral rules of war basically coincide with the moral rules of everyday life.

Steinhoff’s departure from moral fundamentalism is in one way concessive. There is no denial from him that the same basic principles that govern moral dealings in everyday life—the principles of self-defence, the emergency defence justification, the lesser evil justification, and the public authority justification—also apply to war (EWFL, p. 216). We need not enrol any additional sui generis basic principles of war. But these principles apply differently to war, because of differences in the ‘scopes and limits’ or the ‘scale and complexity’ of their application in the circumstances of war (EWFL, pp. 216, 222). There can therefore be different rules in war, due to the descriptive differences between the circumstances of war and the circumstances of everyday life.

Now moral fundamentalists are unlikely to deny that warfare presents us with different descriptive circumstances from those confronting us in everyday life. However, fundamentalists are also likely to downplay the significance of these differences if their aim is to uncover the ‘deep morality’ of war, before we factor in the pragmatic rules and accommodations that are required for the effective management of ongoing conflicts. Fundamentalists’ account of the deep morality of war standardly questions the independence of jus ad bellum from jus in bello: that is, the independence of that dimension of just war theory concerned with the morality of entering into war from the dimension of just war theory concerned with the morality of fighting in war.Footnote 2 If a side has already been condemned by jus ad bellum, fundamentalists will say that there is no longer any prospect, regardless of what happens next, of the unjust side’s successful satisfaction of jus in bello. Now there may be reasons that fundamentalists can acknowledge for operating with a more symmetrical legal regime for appraising combatants. Symmetrical legal regimes avoid unhealthy incentives for the unjust side to avoid prolonging the conflict, and also avoid exercises in victor’s justice, in which the winning side (however morality judges its victory) enjoys the procedural or institutional possibility of prosecuting or punishing the surviving combatants on the losing side. Pragmatically, we should settle for symmetry, but at a deeper level—the level of ‘deep morality’—there remains a stark asymmetry.Footnote 3

This is a division of moral territory to which Steinhoff is notably hostile. For him, there is ‘incoherence’ in McMahan’s suggestion that the ‘oughts’ of (asymmetrical) ‘deep morality’ fail to coincide with the ‘oughts’ of (symmetrical) legal rules (EWFL, pp. 216 − 18, 248-9). If morality endorses the symmetrical legal regime, there can be no genuine rivalry between the ‘deep’ moral ‘oughts’ and the (supposedly less deep) morally approved legal ‘oughts’ that both apply to acts of fighting in war. Morality has already endorsed symmetry, by endorsing legal symmetry, so how can it also be endorsing asymmetry? The metaphors of differential depth cannot disguise the underlying inconsistency.

I think Steinhoff’s challenge to this two-tier model of war is an instructive one. Still, it does leave us with the further challenge of explaining how fighting that was undeniably condemned in advance (under jus ad bellum) can now be affirmed as morally permissible (under jus in bello). How can symmetrical jus in bello regimes emerge from asymmetrical jus ad bellum verdicts? No theory of war can afford to neglect this basic question, which did so much to motivate and focus the revisionist project.

Thankfully, Steinhoff has more to offer. His strategy, to put it roughly, is to station theory and practice more closely together: ‘…the existing laws of war are partly constitutive of the morality of war, and are so on a “deep” level, not only on grounds of “pragmatic” or “epistemic” considerations’ (EWFL, p. 216). Steinhoff’s project, then, is not to ‘to uncover the true, “immutable” contours of “deep morality”’ for war, but to establish the possibility of a ‘social practice… to get onto the same page with regard to one’s practices and standards’ (EWFL, p. 241). How does he do that? I will provide an exposition of the essential moves in Sect. 2, largely abstaining at this stage from critical intervention, and advance some queries and objections in Sect. 3. I do not claim that these objections are enough to sink Steinhoff’s account, which seems to me to contain some promise, but I do think that these objections need to be carefully managed if his account is to be given a clean bill of health.

One final preliminary remark: Steinhoff does not give his account an overarching name, but for ease of reference I will call it the social practice view.

2 Necessity, Proportionality, and Reciprocity

In this section, I will round up some crucial lines of argument in Steinhoff’s articulation of his social practice view, focusing in particular on his treatment of necessity, proportionality, and reciprocity (EWFL, pp. 231 − 41).Footnote 4 I have already indicated that these underlying materials are not unlike those employed by moral fundamentalists: Steinhoff’s theoretical building blocks are drawn from the principles and theories of everyday morality. Admittedly, Steinhoff is determined to do something rather different with them.

Necessity and proportionality are conditions that need to be satisfied in any mature theory of self-defence. The typical understanding of the necessity condition makes defensive violence permissible only if it is unavoidable. If defenders can evade the threats posed to them by aggressors without violence—by retreating, for example—then this is what they should do. Steinhoff describes the necessity condition as involving the requirement that ‘the defender uses the mildest means of self-defense among equally effective and safe means’ (EWFL, p. 236). The proportionality condition insists that the harms inflicted on the aggressor by the defender should not be excessive compared to the harms that the aggressor would inflict on the defender in the absence of a successful defensive response. One of Steinhoff’s important contentions about both these conditions is that they are practice-dependent: how they apply depends on facts about how different agents have been responding to them.

First, necessity. Consider the following case (EWFL, pp. 231-2):Footnote 5

Villagers: I live in a village where, due to some puzzling addition to the water supply, everyone is periodically prone to making unprovoked and psychotic, but also non-lethal, attacks on their fellow villagers. I am no less prone to this violence (both giving and receiving) than anyone else. At first, I restrain my defensive responses according to strict standards of necessity: if I can retreat, I do so, and I inflict only strictly necessary violence on my temporarily deranged attackers. But upon learning that other villagers are operating under no such scruples, and that they also defend themselves in this more violent way against me during episodes in which I am in the grip of this temporary psychosis, I allow my standards to decline to the more relaxed standards of defensive violence adopted by everyone else.

The aim of Villagers is to show how the necessity standard may vary depending on how other agents have been responding to it. At first, other villagers are less scrupulous than me in being restrained by the necessity condition. I am guided by it, but they routinely ignore it. But then my scruples dissolve, like theirs.

One crucial interpretive question raised by Villagers is this: are all the villagers, including me, now ignoring the strict demands of necessity, or is it rather the case that the necessity standard has now shifted?

On the first interpretation: it might be supposed that the necessity standard has endured but that everyone is flouting it. Even so, the conditions for overall moral appraisal may have changed in the meantime. If my fellow villagers have already flouted necessity, then the fact that I have now flouted necessity is unlikely to make me a magnet for blameworthiness, because my fellow villagers will lack the standing to blame me for a moral failure of which they, too, are guilty. Necessity has nonetheless gone unheeded. I will refer to this interpretive option as the weaker interpretation of Villagers. The weaker interpretation makes the necessity standard invariant, before and after individuals’ derelictions from it, but it will concede that an individual’s blameworthiness for flouting necessity will shrink in inverse proportion to the general level of compliance with this condition. By contrast, the stronger interpretation of Villagers holds that the necessity requirement has actually become more lenient, due to these changing facts about actual practice. There can be genuine variations in what necessity requires of us, depending on how various agents are interpreting it.

It is reasonably clear, I think, that Steinhoff is proposing the stronger interpretation rather than the weaker interpretation. This is for two main reasons.

First, Steinhoff places great emphasis on the idea of reciprocity (EWFL, pp. 232 − 33; see also Rodin (2014) and Miller (2012). Let us say that you and I both qualify as rights-holders. The basis for my rights, and the basis for your obligation not to violate them, consists at least in part in my willingness to acknowledge and not violate your rights. The same goes for you: the basis for your rights, and the basis for my obligation not to violate them, consists at least in part in your willingness to acknowledge and not violate my rights. As a more particular illustration of that more general idea, one of the things we can both count on in our moral dealings with each other is mutual awareness that our liability to defensive violence is restrained by the necessity condition. But if you do not heed necessity in your dealings with me, then the basis of your claim that I ought to heed necessity in my dealings with you is immediately weakened (and vice versa, of course). This is because you have offended against the kind of reciprocity that is a sine qua non of your having this particular right against me. Similarly, if you abide by only a weaker or diluted version of necessity in your dealings with me, then there is a much stronger case for my adoption of a similarly weakened or diluted version of necessity in my dealings with you.

The second reason concerns the incidence of strategic advantages and disadvantages (EWFL, p. 232). If I am in a social environment in which everyone fights less scrupulously than me, as I initially am in Villagers, then I am placed at a strategic disadvantage. To put the relevant point in unvarnished form: if others are fighting dirty, then fighting cleanly will place me at a strategic disadvantage, and so there will be nothing wrong with me fighting dirty as well. To refrain from the higher levels of violence that others indulge in would be to reduce my chances of survival. There is no moral reason why I should be forced to suffer these strategic disadvantages, and so I can permissibly switch to a less stringent standard to avoid being enmeshed in them. The net result is that there will be a drift towards convergence in local defensive standards between you and me and the rest of the villagers. At some point or other, morality will permit everyone in Villagers to be on the same page.

I finish this section with a briefer look at Steinhoff’s treatment of proportionality. Proportionality calculations are ubiquitous in these debates. We might accept, for example, the relevance of the lesser-evil justification for killing one to save many others. As a more particular example of lesser-evil reasoning, we might take ourselves to be justified in diverting a threat so that it heads towards one agent rather than five agents, as in standard trolley cases (EWFL, pp. 236-7). The question is: what determines these proportions? Will a five-to-one ratio always satisfy in trolley cases? Should we embrace a more stringent ratio than that, or will something more relaxed suffice? Steinhoff’s line is that there are no determinate mind-independent answers to such questions. Local practices and local settlements are all we can count upon. These then form the basis of moral expectations and the relevant standards for respect and consideration. No one in any given community can complain when these standards rather than others are acted on.

Of course, even if we accept Steinhoff’s suggestion that different moral communities can call on different proportionality standards, we will want to know what happens when sides drawn from different communities go to war with each other. Which common set of proportionality standards should we operate with then? Imagine there are two sides to a conflict: Strict and Lax, which respectively embody the different approaches to proportionality suggested by these names. It might seem that, if proportionality standards were fixed by local practice, Lax would enjoy a strategic advantage over Strict. Since we know from Steinhoff’s discussion of Villagers that the more scrupulous need not put up with this strategic disadvantage, it will follow that Strict can relax its defensive standards in order to match Lax’s more relaxed standards without falling foul of moral demands. Thus there is a natural drift towards a new local equilibrium; Strict and Lax get on to the same page. But a general improvement in mutual scrupulousness is not ruled out, Steinhoff reminds us, since Lax might be inspired by Strict’s practices into accepting a case for more refined and mutually advantageous defensive standards on both sides. A race to the bottom is not inevitable (EWFL, p. 238).

3 Four Challenges

In this section, I outline four challenges to Steinhoff’s social practice view. As I earlier indicated, these may not be fatal to it, and in the space available to me I will sometimes speculate about how Steinhoff might fix these problems. But, as things stand, I do take them to pose obstacles to the social practice view as he actually articulates it.

3.1 Standing Challenge

The first of these is the Standing Challenge. The standing to blame or complain is given a lot of work to do in Steinhoff’s social practice view. In Villagers, my supposedly morally protected drift towards more relaxed defensive standards is secured, in part, by other parties’ lack of standing to blame me if I relax my standards to match theirs. They are powerless to complain about policies I have adopted if they, too, have adopted the same policies.

Is the lack of standing capable of doing this, all by itself? It depends on what explains the lack of standing, and what on what the significance of a lack of standing consists in. The standing to complain is usually held to affect the propriety of various forms of second-person address in morality: I cannot condemn you for wrongdoing if I, too, am guilty of the same form of wrongdoing. There is no denial in this case that we have both acted wrongly, and both of us may be condemned by a bystander with suitably clean hands. We should distinguish between the wrongness of what you do and my standing to condemn you for it.Footnote 6

Now there must be a further story to tell about why, if you have acted wrongly, I cannot condemn you for it. Along with everyone else, I can surely point out that you have behaved wrongly. This is not a colourless or morally inconsequential claim, after all, and I am not debarred from making it. What is the difference supposed to be between saying that you have acted wrongly and condemning you for it? And why, if you have acted wrongly, should my claim that you have acted wrongly, if it meets the relevant accuracy conditions, fall short of its usual negative associations? However these questions are to be answered—Steinhoff himself is not notably detained by them—the moral standards themselves are unaffected by standing. The standing to blame is a feature of blaming practices, not of the underlying moral standards which must ultimately regulate the assignment of blameworthiness. But then it would seem that the situation described by Villagers might be squarely one of moral deterioration, rather than one which involves a change in moral standards. Insofar as standing is concerned, the weaker interpretation looks more defensible than the stronger interpretation.

3.2 Reciprocity Challenge

Faced with the Standing Challenge, Steinhoff might of course switch emphasis from standing to reciprocity. This leads to the Reciprocity Challenge.

Again, we can use Villagers to focus ideas. Steinhoff’s basic point is that, if you are not heeding my defensive rights, then I am under no obligation to heed yours. If you are heeding only a diluted version of my defensive rights, then I am under no obligation to heed anything more than the same diluted version of your defensive rights. Our defensive standards stand or fall together. I want to make two comments about these commitments.

First, it is far from straightforward to disentangle the influence of standing from the influence of reciprocity in Steinhoff’s argument. If I am immune to moral condemnation for adjusting my defensive standards to the more relaxed defensive standards that you have already adopted, then Steinhoff’s working assumption seems to be that I have not acted wrongly. But if I have not acted wrongly, then the standards for wrongness have simply shifted as a result of our changing practice. If blameworthiness is the guide to wrongness, and blameworthiness is not being advanced, then this is enough to indicate that the wrongness standard has altered. Steinhoff is not very explicit about how the conceptual connections between blameworthiness and wrongness hang together, but it is hard to escape the suspicion that they march in close formation in his argument. Yet, as I have already explained, I do not think that the standing to blame considerations are equipped to explain shifts in wrongness and rightness. They are concerned with aspects of our blaming practices, not shifts in the normative valence of the acts themselves.

Second, Steinhoff might have been misled by surface features of non-ideal theory into overestimating the significance of reciprocity. Take the behaviour of defensive liability: however it is grounded, it will usually take the form of the aggressor losing something and the defender gaining something. In virtue of his attack, the aggressor becomes defensively liable: he loses the protection of a right against being harmed that he can ordinarily count upon. Meanwhile, the defender gains normative powers: she acquires the right to harm the aggressor in defence of her life. As I say, I am not going over particular grounding theories at this point: I am only drawing attention to the normative shape of liability ascriptions. Liability ascription appears to involve a tit-for-tat alteration in the respective moral standing of the aggressor and the defender. But this tit-for-tat shape should not necessarily be identified with the grounding theory for liability. That will often consist in some other theoretical story: perhaps a story about rights violation, or a story about the responsibility-sensitive distribution of liability, or something else. Now I do not deny that reciprocity can provide this grounding role. Yet, in truth, Steinhoff does little to substantiate it, except to gesture towards philosophical discussions that deserve to be taken seriously. So, there is work left to be done.

3.3 Necessity Challenge

Now perhaps Steinhoff can make satisfactory progress with the social practice account even if he has unanswered questions arising from the Standing Challenge and the Reciprocity Challenge. Perhaps he can just use the resources of ordinary non-ideal theorizing instead. In Villagers, I am faced with an increasingly violent environment. Acts and policies that I could not have permissibly adopted in ordinary circumstances are now available to me, given the dangers posed to me by my fellow villagers. I will be justified in fighting fire with fire. We do not, in fact, need any special assistance from the standing to blame or reciprocity considerations in order to justify my adoption of more aggressive policies for dealing with them.

I make two comments on this proposal. First, a crucial ingredient in the evolution of defensive standards is that those who fail to align their standards with others are at a strategic disadvantage. But how does this consideration apply to necessity in particular? If my more scrupulous treatment of you genuinely meets necessity, whereas you are less scrupulous, then why am I disadvantaged, and what explains my lower chances of survival? Once I have identified the danger you pose, going on the evidence available to me, then I am still restrained by the requirement that I deal with it with the least violent means at my disposal. That does not mean that I must incur additional costs, exposing myself in the meantime to attacks from others, in order to moderate my defensive violence. Assuming costs are constant in any given case, however, I am still bound by necessity. And so are you. How exactly does the necessity standard shift, and why does the fact that I am still governed by it place me at a greater strategic disadvantage than I was? Perhaps the vaguely Hobbesian thought that the only way of ensuring my safety in the state of nature is to engage in pre-emptive obliteration of my opponents or potential opponents is lurking in the background. But we surely do not want to commit ourselves to this Hobbesian line of thought. Perhaps others lack the standing to blame me if my standards decline, like theirs. But that is consistent with a decline in standards, not just a practice-dependent alteration in what those standards are.

Second, Villagers does not easily identify the problem presented to us by fundamentalists. In Villagers, everyone poses a danger to everyone else. In war, at least one side has already flouted jus ad bellum, while the other side may simply be defending its people and territory against illegitimate aggression. We do not have usable moral materials on both sides of the conflict that might then somehow reach equilibrium within a single social practice. Fundamentalists will reserve the necessity condition for the just side but not the unjust side. Now Steinhoff has had a great deal to say about the moral equality of combatants doctrine, so I am not suggesting that he will be flummoxed by this challenge (EWFL, pp. 156 − 92; see also Steinhoff (2008) and (2012). But the theorizing that bears specifically on the construction of the social practice account seems to be largely inattentive to the asymmetries that are given particular emphasis by fundamentalists. That strikes me as odd, and points to a lacuna in his account.

3.4 Race to the Bottom Dilemma

As we have seen, Steinhoff’s sketch of how different sides in a war can get on to the same page proceeds through his claim that proportionality standards are ineluctably conventional, and on his claims about shifting necessity standards.

I shall not challenge Steinhoff on the claim about proportionality and convention. I find it difficult, in fact, to contest the presence of some conventional elements in proportionality calculations, since these issues are so obviously dependent on an intuitive sense of appropriateness for which, in truth, not much further argument is ever given. The pressing question for Steinhoff’s theory of warfare is how these different communities get on the same page when they fight each other. Different communities may have different conventional defensive standards. When they fight in war, where does the pressure towards convergence come from? How exactly do the different sides get on to the same page?

We know the answer to this question in outline. Again, the lessons are contained in the cases we have already considered. If Strict confronts Lax, then Strict will be strategically disadvantaged unless it commensurates its standards with those of Lax. But how does this protect us from the ‘race to the ‘bottom’? I think Steinhoff faces a dilemma, which I will refer to as the Race to the Bottom Dilemma.

On the first horn of the dilemma, Strict may respond to Lax’s laxity by adopting lax standards itself. On Steinhoff’s view, Strict is released from the sort of moral pressure that would normally apply to it. Upwards pressure is not conceptually impossible, but downwards pressure is substantially more likely, since it seems much more likely that Strict will have to adjust to Lax than that Lax will be inspired by Strict.

On the second horn of the dilemma, there is a distinct sort of moral pressure for Lax and Strict to adopt stricter rather than laxer standards, and this in turn suggests a separate objection to Steinhoff’s social practice view. This is because the practice-dependent approach must struggle to explain what can be objectively interpreted as a lower moral calibre of defensive standards on Lax’s side. If Lax can strategically benefit from the adoption of standards that are championed by Strict, then why shouldn’t Lax have accepted them in the first place? And if there are no such practice-independent moral reasons for accepting these stricter standards, then why should Lax accept them, and how can a race to the bottom be avoided? Genuinely practice-dependent standards are harder to establish than Steinhoff thinks.