Against the Odds: Defending Defensive Wars
This is a guest post by Professor Gerald Lang (University of Leeds), as part of the Reflections on the Russia-Ukraine War series, organized by Aaron James Wendland. This is an edited version of an article published in Studia Philosophica Estonica. Justice Everywhere will publish edited versions of several of the papers from this special issue over the next few weeks.
Peace is better than war. It takes two to fight. These are truisms: they’re true, but so obvious that they’re not usually worth stating. But they swiftly generate conundrums in the ethics of war in general, and the Ukraine conflict in particular. We can learn something, in my view, from thinking about these conundrums. But we may need to tackle the understandable concern that it’s unhelpfulto explore them at a time when energy and attention levels are flagging in the international community, even though Ukraine remains under attack from Russia and arguably requires all the support, moral and otherwise, that it can get. In some circumstances, indulging in more theoretical speculations—the kind of speculative and hypothetical thinking that forms the daily diet of philosophers of war—may come across as being objectionably detached, or perhaps as just another way of being a useful idiot. These worries deserve careful consideration, not hasty dismissal. If there’s to be a place for serious philosophizing about war, it needs to be reconciled with the more engaged concerns of those who care deeply about the Ukraine war but lack specifically philosophical concerns about it.
I’ll return to the question of philosophy’s role towards the end. In the meantime, there is—for better or worse—some philosophizing to do. There are two general problems in the ethics of war that I want to discuss, and which have a bearing on the defensibility of Ukraine’s resistance to Russian aggression. First, why is an invaded country justified in fighting if it is unlikely to win? Second, and more generally, why is it justified in fighting to retain political independence or sovereignty, regardless of its chances of winning? Let’s look at these in turn.
First, we can assume for the purposes of argument that an unprovoked military invasion is the sort of cause for which a country may fight, at least in certain conditions. Just war theory, the body of moral thought that developed in incremental fashion over several centuries and whose content is to some extent embodied in international law, distinguishes between the conditions under which a country is justified in going to war in the first place (the ‘jus ad bellum’ dimension), and the fundamental moral principles governing the fighting in war (the ‘jus in bello’ dimension). Our interest here lies with jus ad bellum, which contains several further conditions. Perhaps the most important of them is the ‘just cause’ condition: the type of reason a country needs to appeal to in order to possess a justification for fighting. On that front, a defensive war passes muster. Resisting military annexation by a hostile foreign power is the most obvious form of just cause, and it also tends to be treated, in international law, as a sufficient condition for permissible warfare. According to Article 51 of the United Nations Charter, ‘Nothing in the present Charter shall impair the inherent right of individual or collective self-defence if an armed attack occurs against a Member of the United Nations, until the Security Council has taken measures necessary to maintain international peace and security’. When the Russian troops poured over the Ukrainian border in February 2022, Ukraine enjoyed legal permission to resist them by force.
The Reasonable Prospect of Success Condition
Other jus ad bellum conditions complicate the justification test. The one I want to focus on is the ‘reasonable prospect of success’ condition. This condition instructs a country not to wage war—including a defensive war—if it lacks a reasonable prospect of success. The word ‘reasonable’ doesn’t state an exact probability of success. For our purposes, we can take it to mean ‘not very low’; there has to be a fighting chance of fighting one’s way to victory. Now this particular provision of jus ad bellum isn’t encoded in the body of international law governing international conflict. But it still deserves consideration, since the law isn’t the last word on how we should decide to act. Law is permissive rather than dispositive: it tells you what you’re not allowed to do, but it doesn’t ensure you have good reasons for doing what you permissibly do.
The reasonable prospect of success condition may surprise some people, since it suggests that it may be wrong for a state like Ukraine to fight a defensive war against a state like Russia. Again, most of us won’t doubt that it was at least permissible for Ukraine to resist the Russian invaders. But the reasonable prospect of success condition will challenge this conviction if the odds of successful Ukrainian resistance were very low. Let’s focus for the moment on an imaginary case, to make the point vivid and exclude real world complications. Imagine two neighbouring countries. One of them—the ‘aggressor state’, as I’ll refer to it—is hostile, powerful and expansionist. The other—the ‘victim state’—is the aggressor state’s small, peaceful, and relatively powerless neighbour. Now imagine that the aggressor state invades the victim state. The victim state’s resistance to the aggressor state may be conceivably unwise, but why would it be immoral? It seems, if anything, heroic. If we transposed the basic discussion to the ethics of individual self-defence, we wouldn’t morally condemn a relatively defenceless individual from standing up to a more powerful assailant who had cornered him and was determined to give him a beating. We’d be admiring his pluck and cheering him on.
It isn’t just this lack of fit with our intuitive responses that’s a troubling feature of the reasonable prospect of success condition. We should also be concerned with the deeper messaging. The bigger the bully, the less morally acceptable it is to resist him. Is that the message morality is endorsing? If so, then morality seems to be siding with the bully: the bigger the bully, the more morally questionable it is to stand up to him. We surely don’t think that morality should scold underdogs and side with their oppressors. We also don’t want to incentivize aggressors to become so powerful that resistance to them is not just unwise, but morally wrong. These implications seem, perhaps inadvertently, to be in the neighbourhood of a ‘might is right’ doctrine.
These are understandable concerns, but the reasonable prospect of success condition isn’t senseless, and arguably doesn’t involve any unwise concessions to ‘might is right’. The comparison with interpersonal defence, involving moral relationships between separate human individuals, may be misleading in this respect. War isn’t like a boxing match. It isn’t properly modelled on a sporting contest. The reasonable prospect of success condition makes particular sense when we envisage two armies confronting each other. For the soldiers defending the victim state are commanded to fight—they do so at the behest of orders higher up the chain of command. The lives of these soldiers are as precious as anyone else’s in the victim state. Even if the role of these soldiers is to fight in protection of their fellow citizens’ lives and interests, so that they are unavoidably exposed to danger when they are active in their role, the fighting they are commanded to engage in cannot be simply futile. Their lives shouldn’t be needlessly squandered. They are fighting for their fellow citizens, but they’re also trying to defend themselves against their military opponents. If it’s extremely likely that they’ll fail in these endeavours, then they shouldn’t be condemned to sacrifice their lives for no discernible gain to anyone else.
Another relevant consideration is that casualties in war are hardly ever restricted to military personnel, whose role is to absorb the violence and deflect it from the wider civilian population. There will inevitably be casualties among civilians as well. If the defensive forces face impossible odds, then it may be more rather than less likely that there’ll be extensive civilian casualties in the victim state. And if the aggressor state is unscrupulous—it is, after all, being cast here as the aggressor state—then the victim state probably can’t count on it to respect the various provisions of jus in bello and refrain from targeting the civilians of the victim state. (We know these provisions haven’t been respected by Russian soldiers in Ukraine.) If the victim state resists the aggressor state, the victim state’s soldiers won’t be the only victims of fighting. The victim state’s citizens will also pay the price, and the victim state’s infrastructure and cultural heritage may also be destroyed, which will reduce the chances of satisfactorily restoring the integrity of the political community after the hostilities have concluded.
The Problem of Bloodless Invasion
A less obvious and more outlandish problem with defensive wars arises even when worries about the prospects of success are set to one side. This problem is known, in just war theory, as the ‘problem of bloodless invasion’. Imagine, then, that the aggressor state in our example threatens the victim state with bloody invasion unless the victim state cedes its territory and political control of that territory to the aggressor state. Assume further that if the victim state readily acquiesces, then the aggressor state will refrain from violence. The victim state will lose its political independence, yes, and its people will face a future of political subjugation under the aggressor state, but it won’t have to endure the kind of costs that are measured in deaths. The aggressor state, by assumption, is after territory and political control. It isn’t pursuing genocide. It’s prepared to crush military opposition, but the threat is conditional. If there’s resistance, then the threat will be applied, and there will be death and destruction. But in the absence of violent opposition, there’ll be no violence. Remember, it takes two to fight. The lives and general destruction that would inevitably be lost in a bloody conflict won’t be endangered if there is immediate capitulation from the victim state. If the victim state cares about the lives of its soldiers and citizens, then it needs to think carefully about whether it should activate the condition from which the aggressor state’s trail of destruction will ensue.
Is the victim state permitted to offer violent resistance to the aggressor state? Traditional just war theory answers this question affirmatively. The reservation it noted for the victim state under the reasonable prospect of success condition goes missing for this more abstract question. The issue nonetheless deserves a closer look.
Importantly, the problem of bloodless invasion doesn’t pretend there are no moral differences between the victim state and aggressor state. It doesn’t say the differences between victim and aggressor are blurred, or that the roles of victim and aggressor become indistinguishable, simply because it takes two to fight. No one has to deny that it is the aggressor state that has issued the conditional threat against the victim state. This is an obvious, and obviously morally consequential, asymmetry between the two of them. No one should issue such conditional threats, and anyone on the receiving end of them deserves our sympathy and support. The aggressor state can therefore be condemned and deplored in ways from which the victim state will be spared, whatever it decides to do next.
Even so, the fact that the aggressor state has issued a conditional threat makes it plain that the victim state also has a morally consequential decision to make. The victim state shouldn’t assume that, because the aggressor state is blameworthy for issuing a conditional threat, nothing further turns on the victim state’s meeting the condition which will transform the conditional threat into actual violence. What about all the soldiers and civilians who will pay with their lives, or the lives of their loved ones, if the fighting goes ahead? The victim state needs to determine whether that price is acceptable. It shouldn’t assume that morality gives it a green light to fight back simply because, whatever happens, the aggressor state has already earned its place in morality’s bad books.
In the ethics of defence and war, it is common to distinguish between ‘vital’ and ‘non-vital’ interests. Among the vital interests of the victim state are the lives and bodily integrity of its citizens. There are also distinctly political interests: the interest in living in a self-determining political community, free of direct subjugation from a foreign occupying power. Living in a self-determining political community is certainly valuable; it is a distinctive and significant political good. This is something that most of us want for ourselves, and for other societies as well. The question is whether such imperilled political interests deserve to count as vital interests, so that it can be justified fighting to the death to preserve them.
In discussions of the problem of bloodless invasion, we’re often invited to dwell on interpersonal cases. These cases provide relatively uncontentious examples of non-vital interests. I have an interest in my wallet, or my phone, or my sandwich. These are my private property. You have no right to any of these things without my consent. But I may not kill you to hold on to them, if you’re out to snatch them away from me. I may perhaps deploy moderate force to resist your attempted theft of them, but this defensive force cannot rise to lethal levels. If you’ve already taken my wallet, or phone, or sandwich, and I can reclaim them only by killing you—no lesser amount of violence will get them back into my possession—I’ll simply have to stand down. I’ve been wronged, but I lack a specifically lethal remedy for this wronging. You have, in effect, priced me out of the market of a permissible response that can overturn the wrong you’ve inflicted on me. These conclusions reflect the influence of the proportionality condition, which looms large in both the ethics of interpersonal defence and just war theory in both its jus ad bellum and jus in bello dimensions.
These moral restrictions may apply to private property or minor incursions into our physical space, but what about political interests? Don’t they count for more? They may do, but the case for the lethal protection of them is far from clear. Consider my right to vote. This right is one central manifestation of my political status in the democratic and self-determining political community I inhabit. Now what if you’re proposing to use force to stop me from exercising my right to vote? Perhaps I may expend moderate force to extricate myself from your aggression and get to the polling booth; it isn’t a trivial right you’re interfering with, and I’m normally entitled to resist, by force if necessary, such interference with my rights. But is it permissible for me to kill you to break free of this interference? The intuition that I may do so seems much less secure. You’re wronging me, yes, but not every wrong warrants a lethal response. The right to vote may also fall into the extensive territory of non-vital interests. It’s important, but it may not rise to the significance of the central rights to life and limb that would justify a deadly defensive response. Denying me my right to vote isn’t like making an attempt on my life, or attempting to rape or mutilate or enslave me. Avoiding these things is worth fighting to the death for. Is voting? Now it should be pointed out that people have died to win their right to vote. We usually think of these people as heroes, not fools. Expansions in the democratic franchise are rightly regarded as milestones in political history. The frustration of an individual’s right to vote is a serious matter, and not one to be casually dismissed. Still, it’s far from obvious that it would generate a right of lethal resistance. At the very least, this seems to be a borderline case.
There’ll be no shortage of replies to the problem of bloodless invasion.
First, the problem assumes that the threat of violence is merely conditional, and that the relevant condition is entirely under the control of the victim state. Consider, again, an interpersonal case. You say the following to me: ‘I’m going to move into your house: if you resist me, I’ll kill you, but if you don’t, you’re in no danger. We’re just going to be housemates. There will be no violence. I promise you that’. One natural response to this conditional threat is to disbelieve the stated condition. If I let you into my house, I’ve already created the conditions of acute vulnerability to violence and ongoing domination from you. I can’t be expected to trust the word of someone who has issued such a threat. You were prepared to kill me, after all, to get your way. This demonstrates that you have no regard for my interests. As a result, I have little reason to think your promises are credible. Moreover, if I don’t resist you at this stage, I may find it much harder to resist you later on. Given these considerations, there seems to be a firm case for my standing my ground and fighting you, to the death if necessary.
The very same conclusions hold for the aggressor state’s relationship with the victim state. The mere fact that the aggressor state is prepared to slaughter the victim state’s soldiers and citizens to achieve political hegemony should be enough to set off alarm bells. Once the victim state’s territories are occupied and its armies are neutralized, what would stop agents of the aggressor state from then enslaving or killing the population of the victim state? Why should the victim state be prepared to take a chance on the aggressor state’s claim that it will be only conditionally violent? The lesson here is plain: conditional threats aren’t trustworthy, because they’re issued by actors who are untrustworthy, and these actors are untrustworthy simply in virtue of the fact that they’ve issued a conditional threat. The aggressor state is prepared to raze the victim state to the ground if it resists. This can’t be construed as merely a kindly offer of a change in government.
Second, the interpersonal case I provided to help us think about the significance of political rights may be misleading. It’s one thing for you to threaten my right to vote in a particular election. What if you threatened everyone’s right to vote in every future election, by abolishing democratic elections and instituting a dictatorship? Wouldn’t the preservation of our democratic system be worth fighting for, even if you paid for the resistance with your life? If we scale things up, so that a nation of roughly 40 million people are all faced with the prospect of a loss of political independence—which is the situation Ukraine is facing—the stakes appear considerably higher. One answer to this point is that the scaling up cuts both ways. Yes, what is at risk of being lost is of course of much higher value when we add up that loss across everyone who will be subjugated. The death of one saboteur would be justified if the alternative was the permanent abolition of democracy. But the scaling up also includes, on the other side of the coin, the deaths and destruction that will ensue should the aggressor state encounter violent resistance. This particular point isn’t wholly decisive, then. But there is another point to make. Arguably, the value of political independence amounts to more than just the opportunity to vote in an election. This value is the value of participating in a life in which one’s self-respect isn’t permanently jeopardised or hobbled through long-term political domination and oppression. These are grievous losses, which prospective victims are entitled to resist. They may rise to the level of significance which would entitle these prospective victims to use deadly defensive means.
Conditional Threats and Defensive Escalation
It’s difficult, in truth, to determine which interests count as vital and which interests are non-vital. Whatever they may say, philosophers lack any precise formulae for discriminating between them, and are frankly dependent on intuitive verdicts about the more obvious cases lying at different ends of the spectrum. An unwanted and irritating ear flick? Not worth lethal resistance. The theft of a wallet? Not worth lethal resistance. Being killed? Worth lethal resistance. Dismemberment? Worth lethal resistance. Brutal sexual assault? Worth lethal resistance. Permanent political subjugation? It’s hard to say.
There may be a way of cutting through or sidestepping these gaps in our understanding of the exact value of the relevant goods. To see this, notice that conditional threats can actually be decomposed into an unconditional threat and a conditional threat. We will look first at the smaller interpersonal case to illustrate this distinction. When the mugger threatens to kill me unless I hand over my wallet to him, my money is unconditionally threated (he means to make me part with it, one way or another), but my life is only conditionally threatened (I’ll be safe unless I refuse to comply). Some philosophers think that the upper boundary on permissible violent resistance is set by the value of the unconditional threat. Imagine, then, the mugger threatens to kill me unless I hand over my money, which totals £50. Is holding on to £50 worth a life, even the life of the disreputable mugger? It seems not. But we needn’t leave it there. Perhaps the upper boundary on permissible violent resistance is set by the value of the conditional threat instead. My interest in holding on to my property and frustrating the mugger’s wrongful purposes should entitle me to do something: I can push the mugger away, or shove him, or perhaps give him a black eye. Faced with such resistance, the mugger will escalate his aggression, since he’s already told me that he’s prepared to kill me to ensure that he gets his hands on my money. If he escalates, then he’s going to ascend, sooner or later, to a lethal level of aggression. If he’s threatening me with death, then I can defend myself with lethal means. At this point, it’s not the £50 I’m defending. It’s me. I’m defending my life against his attempt to kill me. And if I know what’s in store for me if I resist him to a lesser degree, then I can immediately respond to his initial threat by killing him. I don’t have to wait for the escalated sequence to play out. I can skip the intermediate steps and attempt to kill him pre-emptively, right now, knowing that in due course I’d only be defending my life from his wrongful attempt to take it. We’d have reached that point sooner rather than later, so why not head straight to the climax, especially if I risk lowering my odds of survival by waiting for the full sequence to be realized?
This argument is bound to meet with objections. All of this—someone might reply—could have been avoided if only I’d handed over the £50 to the mugger! The long and short of it is that a life has been lost over something that simply wasn’t worth the loss of anyone’s life. The lethal sequence could have been avoided altogether, and the fact that it wasn’t is due not merely to the mugger’s wrongdoing. If I knew that I was contributing to a sequence of events in which we would arrive at a lethal face-off, I should have surrendered the £50 immediately. Here the ‘it takes two to fight’ thought does encourage a more robust division of blame between the mugger and me. The mugger, and not me, is to blame for initiating the encounter, but we both stayed for the fight, and we’re both to blame for that, even if he remains more blameworthy than me.
This kind of protest is understandable, but I think it can be contested. Why should my right to defend myself against violent robbery be hampered by the fact that the mugger is prepared to kill me to get what he wants? In the case where I have to kill the mugger to retrieve my £50, I will have to abstain. A life for £50? That’s clearly disproportionate. But conditional threats aren’t like that. I’m fighting to defend my life against his attempt to kill me. And I’m entitled to defend my life. True, I could have chosen not to get into that situation. I could have chosen to surrender my wallet to him. But I was also entitled to resist the theft of my wallet with some lesser degree of force. It’s not as though I could have simply retreated with no loss of anything. The fact that the mugger is prepared to kill me to get what he wants shouldn’t remove the non-deadly means of resistance I already enjoy. But if I go ahead and employ those lesser means of resistance, he’ll try to kill me. So, putting these things together, I’m justified in killing him. There’s a definite sense in which he was asking for it. My defensive escalation has in some unobvious way been baked into his aggression against me.
The same lesson can be broadly exported to the case involving the aggressor state and victim state. A full military invasion of the victim state by the aggressor state warrants some forcible degree of resistance. Political independence must count for something. The attempt to undermine it warrants resistance. If the aggressor state has issued a lethal conditional threat, which is what the problem of bloodless invasion assumes, then the victim state can mobilize deadly defence. There’s still, of course, the question of whether the victim state enjoys a decent chance of succeeding, and the political leaders of the victim state must still make agonizing decisions that will lead to bloodshed among both its soldiers and its citizenry. Even if the preservation of political independence is, in theory, worth deadly struggle, what considerations should guide the leaders of the victim state? It seems to me that the victim state shouldn’t make decisions that will predictably result in mass slaughter, unless it suspects that it would be facing a future of enslavement or mass killing in any case. A permissible response might also depend on levels of popular support for lethal resistance, as long as ordinary citizens aren’t deprived of adequate information about what the odds of successful resistance will be.
Philosophy and War
In my view, Ukraine doesn’t lack a principled reason for resisting the Russian invasion. It risks being defeated, yes, but it doesn’t risk being morally defeated at the very outset by the mere fact that it’s fighting against the odds. When expected future casualties and losses become great, however, then Volodymyr Zelenskyy and his government face difficult decisions. The goods of political independence can’t make up for every other sort of loss.
That said, you’ll have noticed that the explorations above of some of the relevant philosophical problems over waging war haven’t had, for the most part, anything very much to do with the particularities of the Russia-Ukraine War. And that’s just one of the many things that can exasperate non-philosophers—and sometimes other philosophers—who are exposed to philosophizing about war. Problems such as the one of bloodless invasion can make you think that the real problem among philosophers is an invasion of bloodlessness. The visceral realities of war, the suffering it causes and the lives it destroys are central to the concerns of journalism and narrative history, but largely go missing in philosophical explorations. Instead we’re offered abstract principles, weird cases, and unlikely what-ifs. The basic apparatus of these discussions can seem offensively detached and uninvolved. Whatever philosophy is trying to do, the suspicion may run, it’s not getting at the full lived reality of war. It has no real place for the reactions of sympathy and horror and grievance and devastation; it turns all these realities into cold calculation, or (at best) feeds all this data into a system of abstract moral geometry. It must therefore fall short—intellectually short, but morally short as well.
How fair are these charges? Can they be parried? I think they can. First and foremost, we shouldn’t expect philosophy to carry the full discursive load when it comes to war. Philosophy isn’t competing with journalism, or history, or political science, or poetry, or narrative fiction, or vigils, or marches. It isn’t meant to match, variously, the commemorative significance or the particularistic human interest or the operational relevance of these other forms of activity. War is complex, and sustains different types of commentary and engagement. Philosophy is primarily concerned with the ethics of war. Should this war be fought? On what terms, and for how long? How, if at all, can we reconcile ourselves to the death and destruction that will ensue from it? To arrive at a more lucid understanding of these issues, we need to abstract from the particular details of the war, and the particular air of emergency that hangs over it, and the particular victims that will be produced by it.
War produces emergencies, and every emergency seems both sui generis and a fait accompli, distinct from and incomparable to every other emergency: it consumes the people who have to deal with it. By contrast, philosophy insists on keeping a cool head, and in consulting more abstract principles which can be imported into an evaluation of any particular emergency. That can sometimes lead to an impression of coldness, or failures of tone or tact. But it can also help to ensure that one precious resource that most warring parties insist upon—a shared conviction in the justifiability of what they do—doesn’t slip out of sight. It’s a price worth paying, since most participants in war, whatever their immediate aims may be, will also wish to be judged kindly by history.
Gerald Lang is a Professor of Philosophy at the University of Leeds, UK, where he has been based since 2005. He has extensive interests in ethics and political philosophy, and is currently writing a monograph on deontology, self-defence, and war. More details of his work can be found at https://philpeople.org/profiles/gerald-lang
