Berkeley Talks transcript: Tanner Lectures, day 2: Arthur Ripstein on why it's wrong to target civilians during war

Kinch Hoekstra: All right. Welcome, and welcome back to the 2019 Tanner Lectures on Human Values featuring Arthur Ripstein with commentaries yesterday by Chris Kutz and today by Oona Hathaway and Jeff McMahan.

I want to first briefly lay out the plan for today’s session. This middle day of the Tanner Lectures, the second day, the Wednesday session, is always a bit of a challenge because we’re trying to squeeze in the second of the Tanner lecturer’s lectures plus two commentators plus the lecturer’s response to both of those comments.

So Arthur Ripstein will deliver his second lecture for up to an hour, then after a short break we’ll then hear from Oona Hathaway for about 15 minutes and then from Jeff McMahan again for about 15 minutes, and then finally if all goes according to plan, Arthur will have 15 minutes to reply.

This will all make for a pretty full couple of hours and requires a certain amount of discipline, and I’m going to do my part by keeping the introductions brief, and I’ll point you to their biographies in the program.

I can assure you, and I think my speakers know that this indicates nothing less than complete admiration for our immensely intelligent and accomplished speakers, but it’s indeed a way to allow them to have the floor for as long as possible today.

The other exigency on the second day is that, as usual, we’re probably not going to have much, if any, time for questions from the audience. But this is designed this way in part because tomorrow gives plenty of time for questions and discussion when there will be remarks from all four of our panelists, that combined should total about an hour. So we should have about an hour to discuss these matters tomorrow, and I urge you to come back.

Professor Niko Kolodny did a brilliant job yesterday of introducing our Tanner lecturer, Arthur Ripstein, professor of law and philosophy and university professor at the University of Toronto. So I’m just going to underline what an honor it is to have him here in Berkeley to try to chart a philosophical course through these very fraught issues.

Immanuel Kant is a world historically great philosopher, but he is no more, and philosophically I think the closest we can now get to having him in the room to think with us is to have Arthur Ripstein in the room. Under the general title of Rules for Wrongdoers, Arthur’s second Berkeley Tanner lecture is Combatants and Civilians. Please join me in welcoming professor Ripstein.

Arthur Ripstein: Thanks so much for that generous introduction. I will try to live up to it, and it would be easier to live up to Niko’s because there I merely say I’m incompetent at dealing with urgent matters and here you’re asking me to stand in Immanuel Kant’s shoes. I will do my best to do so for you today.

Let me remind you that yesterday the point of my lecture was to explain how there can be rules governing activities that are already prohibited, and that I illustrated this idea with the example of perfidy. In explaining the example of perfidy, I distinguished between what might be thought of as an empirical round for prohibiting perfidy in that it undermines actual trust between the parties, and what I characterize as a category based characterization of the wrong of perfidy.

And what I want to do today is develop a similar kind of category based characterization of the wrong of targeting civilians. What I want to do then is to explain why it’s wrong to target civilians by introducing a general and formal category, the formal category broadly speaking of those who are not part of the war. And then I want to explain both why this category is absolutely essential to the possibility of a future peace. And further, I want to develop it further and show that this category of those who are not part of the war maps onto the distinction between combatants and civilians. Combatants are those who are part of the war, civilians those who are not. And so that’s the plan.

And one other thing. In the final part of the lecture, I will turn to a remaining challenge that came out partly in Chris’ comments yesterday, but will have occurred to many of you as you listened to yesterday’s lecture.

My example for introducing the distinction between those who are and those who are not part of the war will be prisoners of war, those who have surrendered. One of the features oaf the status of a prisoner of war is that a prisoner of war is no longer an admissible target in war, but once the prisoner of war has surrendered and been taken into custody, the prisoner of war can be detained but not put on trial. And I will use this example of prisoners of war not only to explain the general characterization of the distinction between those who are and those who are not part of the war. But further, I will use this same example in order to explain something that will have occurred to many of you based on yesterday’s lecture.

My examples of the ad vehendum and invehendo rules and the ad parentum and in parente rules were examples of specific wrongs that could be committed by those who were occupying a certain role or engaging in a certain activity, whether they were engaging in it properly or whether they were engaging in it improperly, whether they were, in the case of parents, it’s particularly clear there’s the legitimate parent, and the legitimate parent contrasts with the kidnapper who is a usurper to the role.

Now we might say that those who fight in a war without adequate cause for going to war are usurpers to the role of warrior, usurpers to the role of combatant. And what I will do, returning to the example of prisoners of war, is offer what I will characterize as once again a category based explanation of why it is that a prisoner of war cannot be put on trial. So that’s the plan.

So yesterday in talking about perfidy I used the example, among others, of a false surrender. What I want to do today is not talk about false surrenders but about genuine surrenders. A genuine surrender takes place, of course, when someone who is a combatant raises the white flag or puts up their hand or in other way signals that they are no longer prepared to fight against the other side.

In characterizing the distinctive role of surrender and the distinctive status of prisoners of war, I’m going to begin with the example of a defensive army taking a member of an aggressive army prisoner. But then, as I proceed, I will suggest that once we understand the distinctive status, we see that the distinctiveness of the status does not depend on the fact that one side is aggressor and one side the defender. So that’s the plan.

So what is it about surrendering that makes it so significant? I think that the naïve view of surrender is in fact the correct view of surrender. When someone surrenders in wartime, what they are doing is making a transition, a transition in their status. Prior to surrender the combatant was part of the war. They were fighting for their side, whichever side that happened to be. Having surrendered, they are no longer fighting. They are, that is, making a transition from being part of the war to not being part of the war.

Now notice that we can think of this transition on a number of different scales. For example, we can think of the individual soldier who surrenders. When the individual soldier surrenders, that soldier is no longer a part of the war. But of course, we can think of a battalion surrendering. We can think of an entire army surrendering, and we can think of an entire nation surrendering.

When we think any of those thoughts, what we’re thinking of is some person or a group of persons making a transition from a condition in which they are part of the war to a condition in which they are no longer a part of the war. That is the right way to think about surrender, is as a change in status. And so the right way to think about surrender is in terms of a category.

Now, why does this matter here? Think about it this way. We can think of the possibility of this transition not merely as something that is one of the things that people sometimes do in war time. But it is also, it seems to me, a fundamental feature of the possibility of exiting a condition of war and of limiting a condition of war. Here’s why. The person who has surrendered stops being part of the war. Having stopped being part of the war, that person is now outside of it.

And I want to claim that within war, if war is to be limited in any way, it must always be possible to make that transition, to make the transition between being part of the war to not being part of the war. Another way of putting this point is to say that peace is only possible if there is a distinctive status of not being part of the war. Because to conclude a surrender is part of what it is to conclude a peace. It is to change the status from being part of the war to not being part of the war.

Now in the case of an individual, an individual combatant is not in a position to end the war to enter into a condition of peace. The most the individual combatant can do is change his or her own status from being part of the war to not being part of it. But obviously, if an entire nation surrenders, they’ve changed their status. And if two nations who had been at war conclude a peace agreement, what they’ve done is, although it would be a mistake to say they’ve surrendered to each other, they have simultaneously, jointly exited the condition of war. They have jointly switched from being part of the war to no longer being part of the war.

But that suggests, and this is the suggestion I hope to develop, that the very possibility of a future peace requires that there be a status of not being part of the war. Now, that is not enough to get us to the conclusion that is of interest to me today. Because of course, that there must be such a status if peace is to be possible does not tell us who occupies that status. For that matter, it doesn’t even tell us whether anyone occupies that status within a particular conflict. All that the distinction does, and this of course is just a mark of category based thinking in general, is say there must be a distinction between those who are part of the war and those who are not, not who falls on either side of it or whether anyone falls on the not part side of it.

But so what I want to do now is suggest to you that this distinction between those who are and those who are not part of the war aligns perfectly with the distinction between combatants and noncombatants. In order to show that these two distinctions align, what I propose to do is remind you of something I talked about yesterday, namely that the only ground of going to war is defensive. And then based on that, use that to generate the alignment of the two distinctions.

Now, I should clarify something about this claim that the only ground of going to war is defensive. In saying this, I of course am acknowledging that there will be some wars that are actually started by people who are not fighting defensive wars. But nonetheless, within the broader structure of the argument that I am proposing to develop here, the thought is that even if only one side is a defender, in order to characterize the distinctive wrong that can be committed in the context of fighting a war, we have to focus on the case of those who are entitled to fight. The thought is that if it is wrong to fight in a certain way when you are fighting a just, that is to say a defensive war, then it is also wrong to fight in that way if you are fighting an aggressive war.

Now of course, that depends on what makes it wrong. But if, as I will argue, what makes it wrong is that it is inconsistent of a possibility of a future peace, then I will have established the conceptual priority of defensive war, which will allow me to show that the two distinctions, that between those who are and those who are not part of the war and those between combatants and civilians, are in fact the same distinction.

So in order to bring this out, what I need to do first is actually say a little bit more about the way in which the distinction between those who are and those who are not part of the war plays itself out in the case of other grounds of war. The grounds of war that historically were thought to be the basis of a nation’s entitlement to go to war.

And so what I want to start with is by reminding you that yesterday I talked very briefly about what I called the classical just war tradition from Augustine and running through Aquinas to the Salamanca scholastics. And I want to draw attention to the way in which that way of thinking about the grounds of war drew a distinction between those who are and those who are not part of the war, and I want to suggest to you that their distinction is perfectly consistent with their way of thinking about the grounds of war. And it does not map on to the distinction between combatants and civilians.

So remember, for Augustine and Aquinas, the basic form of going to war, or the basic ground of going to war, was as an enforcement action as a form of punishment of those who have committed some wrong in the past. A secondary ground of war on their view was a kind of remedial war, a war to reclaim something wrongly taken or wrongly detained by another.

Now notice that, although this is not a ground of war, nonetheless, if it were a ground of war, the distinction between those who are and those who are not part of a punitive war would have to be a distinction between those who did and those who did not merit punishment. Now it is, I suppose, important that I concede that for Augustine, Aquinas, and especially Vitoria, the grounds of those who merited punishment was rather broad. They had a broader notion of culpability than thinkers do today. And so for example, they thought that a ruler’s people were themselves liable for wrongs committed by their ruler. And so this is why they thought that at the end of a war it was perfectly all right to kill or enslave the population that had been vanquished.

But I want to in a way put that aside and merely draw your attention to the fact that their distinction between who is and who is not part of the war, those who are guilty and those who are innocent, was a distinction based on culpability. Maybe they had too broad a conception of culpability. But nonetheless, if a punitive war is acceptable, then of course the people who are liable to be attacked in war are those who are liable to be punished. Those who are being punished and resist, that is the armed forces of the nation that is being punished by another nation, they have a special kind of culpability, but it’s not primary culpability. They can be targeted on the traditional just war view just because they’re effectively obstructing justice. They are stopping a just punishment from being carried out and so they are culpable for that. And so force can be used against them because they’re standing in the way of achievement of punishment. But if there’s a punitive war, that has to be the right way to think about it, and that also means that those who are to blame for the injustice, whether they are involved in the war directly or not, are the people who are the appropriate targets of the armed force.

Now what if there were a remedial war? A war, that is, to reclaim something wrongfully taken? We get a slightly different but nonetheless fundamentally similar analysis. That is, that if seeking a remedy, seeking satisfaction were a ground of going to war, then the people who would be part of the war were those who either had wrongfully taken something that properly belonged to another or those who refused to yield up something that they had wrongfully received, even if they received it entirely innocently.

Let me begin, as so many of the classical writers did, with a small scale and domestic analogy. Suppose that I have wrongfully taken some piece of your property. Suppose I now give it to Oona as a gift. You now demand your property back. Now you’re not allowed to engage in self help. You have to call the police and so on. But the structure of the situation is that it can be forcibly taken from Oona even if she had no idea of the wrong. She did nothing culpable. There was no way for her to realize that I was giving her something that had been stolen. You are allowed to reclaim it from there, though Oona would never do this, were she to attempt to resist your attempts to reclaim that, you could of course use force against her or rather a state’s agents could use force against her to stop her from doing it.

But again, if we think that seeking a remedy, reclaiming what was wrongfully taken, corrective justice, is a ground of going to war, then we must also think that those who are part of the war are those who are holding things that they ought to give up, ought to yield up to their true owners.

So too, if, with the later writers in the just wars’ tradition, the Salamanca scholastics, we think that redistribution is a ground of going to war, then those who refuse to give up their excessive shares of whatever scarce resources they’re holding, those are the people who are the proper targets of force.

The same thing is true of any other kind of exercise of a familiar legal power that the state exercises over its citizens, seeing to it that things are well administered, those who stand in the way of that, they’re the ones who would be the appropriate target.

Now as I said, none of these is in fact a ground of going to war. Yesterday I limited the grounds of war to the defense of legal order, and I know in the comments exactly what that means, will get some more attention today, so I won’t say anything more about it now. But notice that as I suggested yesterday, national defense is not a new item to add to the familiar lists of punishment, remedy, or economic redistribution. Nor is it a substitute for the members of the list. It’s a fundamentally different kind of ground for going to war.

What I meant by that is that the thought of using defensive force is not something over and above the upholding of the prohibition on various forms of aggression. And so as such, the defender’s entitlement to use defensive force is just the entitlement to stop the aggression. That’s all that the defender gets to do.

But now, notice that this has an important implication, something that makes it fundamentally different from the other examples that I was considering earlier. It’s very different from the case of punitive force. Again, thinking in the individual case this is familiar. But I believe it’s also true in the case of states. The question of whether someone can use defensive force against another person is not a question about the culpability of the attacker. You are allowed to use defensive force against a non culpable attack. For example, someone who is mistaken about your identity and attacking you, someone who is mistaken about what you are about to do and attacking you, or someone who is for whatever reason not fully culpable.

And so if someone is acting under duress and attacks you, they would not be liable to punishment, but you’re allowed to defend yourself against them if they’re acting under duress. That is because the sole point of defensive force is not to see to it that wrongs are righted. Nor is it to see that people receive the punishment that they deserve in accordance with their culpability. Nor is it to see that people have proper shares of benefits and burdens. There’s a prominent view of self defense according to which the defender is entitled to use defense force so as to shift an unshared burden onto the defender.

I want to suggest that the entitlement to use defensive force is much less complicated than that and has a much more straightforward rationale. The rationale for the entitlement to use defensive force is nothing more than the entitlement to stop the aggression. But if the rationale is just the entitlement to stop the aggression, then you’re allowed to use defensive force independently of the question of whether the person who is attacking you is culpable. You’re entitled to use defensive force only to the point of stopping the ongoing attack. You’re not allowed to use defensive force to see to it that someone gets what they deserve. You’re only entitled to use the amount of defensive force that you require in order to stop the attack.

Now, I want to say a little bit more about the individual case and then I want to shift to the case of defensive war. I apologize in advance if to some of you the idea of drawing analogies between nations and individual human beings seems highly artificial. This is what all of the classical just war writers did. This is what all of the founders of international law did. And in every case they did it, I think, not because they thought that nations were just like people, though some of the founders of international law thought of nations as kind of a private property of their rulers. But rather, because they thought that the relationship between nations, the kind of wrong that one nation could commit against another, was analogous to the kind of wrong that an individual human being could commit against another.

And so following their example, I wanted to begin with the individual case of self defense just to say a little bit more about it, and then come to the obvious and important dis analogies between the individual case and the international collective case.

So first, in the individual case. If I attack you, you are allowed to stop me. Your entitlement to stop me is not because I have somehow forfeited a right that I had in advance that you not use force against me. I don’t actually think I had any such right in advance. I had rather a right to interact with you in a way that is consist with our respect of independence of each other. And so if I exceed my entitlement to act in that way, you can stop me. That’s why if I attack you, your entitlement to stop me isn’t because I’ve lost a right that I had. Because of course I didn’t have a right to attack you. It’s not that I had a right to attack you and I’ve lost it by exercising it, it’s rather that I had no right to attack you, that’s why you can defend yourself against me. You can stop me because I’m an aggressor and for no other reason.

But your entitlement to stop me as an aggressor is not in fact even an entitlement to stop me as an aggressor, it is more precisely described as an entitlement simply to stop my aggression. It’s just that in the case of an individual attacker, stopping the aggressor and stopping the aggression will normally come down to the same thing. The only way that you can stop me from attacking you is by stopping me, that’s why you’re allowed to stop me. How much force can you use? You can use force sufficient to stop me even if that is lethal force.

So we can think of this on the model of a kind of principle of action and reaction. Kinch said that you didn’t have Kant here to talk, and so I’ll use Kant’s metaphor. Kant says that the enforcement of a right is nothing more than the moral analog of the principle of action and reaction. If I am wronging you, you can stop me from wronging you, that’s all. It’s not that I had a right that I’ve lost, it’s that the boundaries of our rights are also the boundaries of our rights to enforce those rights. Your entitlement to enforce is the equivalent of my entitlement, the limit of my entitlement to interact with you and to put you into peril. So if we think of it that way, we then see that the entitlement to use defensive force is just the entitlement to stop the aggression.

Now to carry this example over to the case of not an individual attacker but a collective attack, we need to make some modifications. The modifications are modifications because they don’t change the fundamental structure. As Kant remarks, the point of an analogy is not to say that two things are alike, it is rather to say that relations between one thing and another are like relations between another two things. So too when we say that national defense is analogous to individual self defense, we are not saying nations are like people, we are saying that the relation of aggression is the same and the relation of defense as response to aggression is the same.

But nonetheless, there is a difference. The most important difference, of course, is that nations are artificial persons. In the 18th Century, Samuel Pufendorf called them moral persons, which I think tends to a certain kind of misreading, because it looks like it’s saying something nice about them, but the thought is not that they are moral person as in upstanding persons, but rather they are moral persons in the sense of artificial persons. An artificial person is a person that can act only through another. States can only act through their officials.

And because states can only act through their officials, what we have to do in order to understand what it is for one state to attack another, is to understand it as acting through its officials to attack another. We understand one legal order acting through its officials to attack another. And who is it that it acts through when it does this? And the answer is, of course, it acts through its military. Its military are those who, in the case of aggression, are the way in which the aggressor aggresses against the defender. But if that’s the structure of the situation, then in order to stop the aggression, you don’t stop an individual, you rather stop that through which the state is aggressing. But that through which the state is aggressing is not some set of individuals understood in abstraction from the relations they stand to their nation, but rather those whose role within the state is precisely that of carrying out this act of aggression.

So, think about it this way. When one nation attacks another, it does so. It’s got a chain of command. There are those who make the decisions, those who implement them, and so on. An act of aggression has to be understood as having two elements to it, and these two elements are actually captured nicely in the UN Charters Prohibition, because what it prohibits is both the use of force and the threat of force. A nation’s armed forces are those through which it uses force, and those through which it threatens to use force.

Why are these two things importantly related to each other? Well, it turns out that within any military of which we are humanely aware, there are, in fact, multiple categories of soldiers along any number of distinctions. I’ll just draw attention to a few of these differences among different types of soldiers. Here’s the first one: Some soldiers are willing, and some soldiers are unwilling. Some of them have been drafted. Some of them will only fight because they believe that they will be shot if they desert, and some of them on the other hand, are fighting willingly. Eager to fight. Some believe in their cause. Some just think they’re following orders. Some are skillful at their art, and some are unskillful at their art. Some are well-trained, some are poorly trained. Some, in turns out, have the nerve to do their job, and others, studies have shown, lack the nerve to do their job.

There was a study done in World War II that showed up to 60% of soldiers were unwilling to fire their weapons in combat. These studies have been called into question and that it may be that the numbers are entirely off. And further, the numbers are probably less reliable than they even were then because, unsurprisingly though distressingly, the way in which the nations of the world have responded to the news that many soldiers were unwilling to shoot was, of course, to train them so that they would overcome their inhibitions and be ready to shoot any way. And so, nonetheless, there are some people who will not, in fact, shoot in the heat of battle. Will, in fact, be totally ineffectual. And then there are some who will be ineffectual because they missed the part of how to clean their weapon, and there will be some who will be ineffectual because they were given defective weapons that simply won’t operate.

All of these people are, nonetheless, part of the threat. And if you are entitled to resist aggression, then you’re entitled to resist not only the use of force, but the credible threat of force. And, indeed, one of the ways in which aggressors aggress is the Napoleonic mode. They say if you don’t capitulate, I will have no choice but to invade. You’re allowed to resist before the invasion has started because it is, of course, a threat of force. The credibility of the threat of force, of course, depends on the size of the armed forces. But the armed forces include all of these people, including the ones who, as it turns out, will be ineffective.

Now, notice that within the Geneva Conventions, there’s a special exemption made for military and religious personnel. That’s because they’re not part of the threat. And, of course, if all of the people who were, in practice, unwilling to use their weapons were to wear distinctive emblems to say, “pacifist” or “lacking in nerve” and were to do that in such a way that the country they were about to invade could see that the first 70 divisions here are actually filled with people who will not fire their weapons, of course it would be illicit to use force against those people. Just as it would be illicit if they sent their medical troops, their medical personnel as the front column. But though it would be illicit, those people would also not be part of the threat. By contrast, everyone who is not visibly not part of the threat, who is part of the larger threat, is thereby part of the threat, and so force can be used against them. That force can be used even though some of them, in fact, will be ineffectual.

Now, I want to underscore a point here. And that is that in thinking about this kind of structure, the point is not that in the so-called fog of war, there is uncertainty about who is and who is not, in fact, posing a mortal danger. This is not an epistemic point. There is an epistemic point here, but the epistemic point is secondary to the categorical point. Because the categorical point says those who their nation uses to threaten, those are the people who are part of the threat. And so although some of them may be non-culpable because fighting under duress, are harmless, these are not the equivalent of human shields. These are not people who are being put forward so that the moral qualms of the defender will be engaged so that the defender will hesitate to use force against them. These are people who are being put forward not as protected persons, but rather as aggressors, as part of the threat. And, my claim is that those are among the class of people against whom defensive force can be used.

Now, I’ve been focused, so far, exclusively on the case of a defensive war. And you might wonder how this carries over to the case of an aggressor. Because, after all, an aggressor isn’t really entitled to be doing anything. But notice that the thing that is distinctively wrong with collapsing the distinction between those who are, and those who are not part of the war, is precisely that maintaining that distinction is essential to the possibility of exiting a condition of war, essential to the possibility of a future peace. And so, when the aggressor targets people who are not part of the war, who are not part of the aggressor’s pursuit of military advantage, the people who can have only an indirect effect on the aggressor’s military advantage. When the aggressor does that, the aggressor commits the same distinctive wrong that is the wrong of collapsing this distinction between those who are, and those who are not, part of the war.

And so, although the aggressor’s troops are part of the war as a result of a culpable decision on the part of the aggressor to go to war, and the defender’s troops are part of the war as part of a non-culpable decision on the part of the defender to defend itself against the aggressor’s aggression. Nonetheless, the entitlement to resist … I’m sorry, the distinction between them applies in just the same way. Because the importance of the distinction between those who are, and those who are not part of the war, is precisely the importance of the distinction between a limited war and a war that has not limits in it. So there has to be that kind of distinction if the only way in which war can be justified is as defensive war, then we have no choice but to regard a war of aggression as an importance sense, a defective version of a defensive war, and so subject to all of the restrictions that apply to the case of a defensive war.

And so, I want to suggest the principal of distinction, this principal that draws a line between civilians and combatants, is precisely a distinction that is category-based rather than empirical and the point of the category-based distinction is to preserve the status of those who are not part of the war.

Now, that leaves me with a challenge to which I must now respond. I introduced yesterday’s lecture by talking about Sir Hartley Shawcross’s claim that those who fought in the Nazi armies were no different from a lawless robber band. His apparent claim that there was no difference between their killing of innocent men, women, and children sleeping in their beds, and their killing of Allied soldiers. My response to that turned on drawing a distinction between two kinds of rules. And I illustrated that distinction with a series of examples including the distinction between the ad vohendum and invahendo rules of driving and the ad parentum and in parentae rules of parenting.

But it will not, of course, have escaped your notice, and indeed Chris drew attention to this yesterday in his comments, that actually people who drive without a license get punished for that. Kidnappers, if caught, get punished for that. And so, although there is an independence to these kinds of rules, and we can understand how the possibility of the wrong being committed does not depend on collapsing one side into the other, nonetheless, it seems that I haven’t fully responded to Shawcross’s claim since, after all, we can draw the distinction and we can say, “No, you didn’t commit this wrong, but you committed that wrong.”

What I want to suggest is that what we now need to do is turn to the thought that although what they do is wrongful, nobody is entitled to punish them. I mentioned this very briefly yesterday in responding to Chris’s comments and actually, yesterday evening, Philip Hindman gave me an even better example for making this structural point. Yesterday, what I suggested is that it’s a kind of what Kant calls subreption —to mistake a principal that applies at one level for a principal that applies at another level.

So here is Philip’s wonderful example of which I should have thought, but I will use it nonetheless. Think about a statute of limitation. A statute of limitation says that if enough time has passed after the commission of a private wrong, such as a breach of contract or a tort, the aggrieved party no longer has a cause of action. After enough time has lapsed in the criminal law, the state can no longer prosecute the criminal. It would be a mistake of a familiar sword, but nonetheless a mistake, to think that the right way to draw a conclusion, or the right conclusion to draw from this structure, is that when someone commits a crime, as long as they evade detection for long enough, it isn’t therefor permissible for them to commit that crime. That isn’t the structure at all.

So to, it would be a mistake to think that I am entitled to injure you, provided that you don’t proceed against me within whatever the relevant statutory period is. It’s not that there’s a permission here. There’s a prohibition, but there is a limit on it’s enforcement. So I want to say that there’s a different kind of limit on the enforcement, here. This is not a time-based limit. There’s an interesting question about why every legal order in the world has statutes of limitation and how they operate, but I won’t go into that here. Because this is a different kind of limitation and it’s a kind of limitation that I think we can understand in systematic terms. In order to introduce that, I want to introduce what seems to me the most promising alternative way of explaining that, and then I want to explain why that alternative doesn’t work, and introduce my own alternative account.

So, in a recent book, Adil Haque has proposed that the reason that we don’t punish people who act in conformity with the in bello rules of war, is that what we want to do is have a kind of social arrangement, or international legal arrangement, that will provide the right kinds of incentives. He draws a useful analogy with the case of diplomatic immunity. As I’m sure you all know, a diplomat is allowed to break the law, or rather a diplomat is not allowed to break the law whatsoever, but if a diplomat breaks the law, the diplomat is nonetheless free of criminal prosecution by the host country. There are all of the charming stories about parking tickets by UN diplomats in New York City, as well as the horrible stories about diplomats who quite literally get away with murder.

Nobody things that diplomats have a license to do this. They think, instead, that somehow there is a reason for states not to do this. Not to prosecute. He says that the point of this is just to make diplomacy go more smoothly to prevent people from taking hostages, to prevent diplomatic problems from blowing up and leading to conflict rather than diplomacy. He then says the same thing applies to the case of not prosecuting those even who fight without a just cause for mirror fighting without a just cause, provided that they’re acting in conformity with the in-bello rules. And he says these are just prosaic considerations, factors like providing people with incentives to avoid attacking civilians, because if you’re captured and you would be punished for attacking civilians, then maybe you’ll play it safe, anticipating the possibility of capture, and so you will be less likely to capture civilians. Neither side in a war would want its own soldiers put on trial, and so the nations of the world have agreed that they will not put each other’s soldiers on trial.

Now, yesterday in introducing perfidy and today in talking about prisoners of war, I drew a distinction between what I called empirical and category-based characterizations of principal. What Haque offers us is an empirical categorization. Like any empirical categorization, it’s only as good as the empirical premises on which it rests. I’m in no position to assess the quality of those empirical premises. That is, I don’t know and I suspect nobody knows, and certainly not the drafters of the Geneva Conventions … Nobody knows how likely it is that combatants actually think it is that they will get captured. Nobody knows how they will discount the probability of capture against the likelihood of avoiding capture by killing a civilian, or engaging in perfidy and so on. As a result of which, that’s one kind of problem with this empirical approach. Here’s a second problem with the empirical approach. The second problem is that if we understand it as a kind of conditional empirical agreement, then it has a structure to it that seems to be out of place with respect to the prohibition on putting prisoners of war on trial.

There are a number of different kinds of rules and different attitudes we might have towards those rules. Some of those rules we think apply no matter what. So the prohibition on battery, for example. From attacking another person. You’re prohibited from doing that even if everyone does it. By contrast, think about lining up in an orderly way for the bus. If everyone else lines up in an orderly way for the bus, then you have to line up in an orderly way for the bus. But if no one else does it, then the conventional arrangement simply evaporates and it doesn’t bind you at all.

I think a lot of people think of the rules of property as kind of like the bus case. That is, they think that as long as we have a system of private property, you have to respect other people’s property, but if there’s no system of private property, then you can help yourself to whatever. If you think that the rule prohibiting putting enemy prisoners of war on trial is just a convention, then it would seem to be the case that if the other side violates the convention, they you are free to violate the convention. But I don’t think that’s how we think about it. I don’t think that we think that this is a matter of reciprocity in that sense. It’s not as though we say, “Well, you know what? We made a deal. We had a deal. You broke it, so we’re going to put yours on trial.”

In order to bring this thought out, I’ll return to the topic of diplomatic immunity in a moment. But let me first say one more thing about the difficulty of the empirical account. Yesterday I mentioned that Christian Wolff’s idea of the voluntary law of nations turned on the thought that nations were, with respect to their populations, a little bit like shepherds with respect to sheep. That is, they were entitled to feed some to the wolves in order to spare the rest of the flock. If you really think that this is a wrong that should be punished, it’s not clear that you can make a deal that exempts you from it. If you really have an obligation to punish wrongdoing, then it’s not clear that make this kind of arrangement is within the moral power of states. And so it’s not clear that we can actually have a morally justified legal convention.

In order to sketch my alternative, let me return to the topic of diplomatic immunity. I want to propose a category-based version of the rationale for diplomatic immunity, and I want to phrase in the international lawyer’s somewhat cumbersome vocabulary of the sending country and the receiving country. When one country sends a diplomat to another country, that diplomat has a distinctive kind of status. The diplomat’s status is that of being subject to the sending country’s law while present in the receiving country. The embassy is treated as the sending country’s territory and the receiving country is not allowed to send its officials into the embassy, even to arrest a criminal for whom they have a warrant. Diplomatic immunity is personal and so the diplomat is free of being charged with any kind of crime, while serving in the receiving country.

Now, the structure of this analysis, it seems to me, is not that this makes diplomacy go more smoothly. The structure of the analysis, instead, is not that this is an exception to the principal of territorial jurisdiction. This just is the principal of territorial jurisdiction. If two nations are independent of each other, then when one sends an emissary to the other on official business, the emissary in order to act for that nation, must be not subject to the receiving country’s system of public law, precisely because the sending country is not subject to the receiving country’s system of public law.

But if that’s the structure, then it has the further implication that although nobody actually thinks that parking illegally is part of one nation doing business in another country, nor does anyone think that committing murder is part of one nation conducting its business in another country. The reason that illegal parkers and murderers alike can’t be put on trial for their crimes, both small and large, is that the question of what is and what is not part of conducting another nation’s, the sending nation’s business in the receiving nation, is not a question for the receiving nation. The sending nation is not subject to the receiving nation’s system of public law. As such, the question of what is and is not part of the diplomat’s job is not a question that the receiving nation is legally entitled to answer. But that means you get an immunity that is based on the independence of the two nations.

Now, I want to carry this over by way of analogy, however imperfect, to the case of the prohibition on punishing prisoner’s of war of an aggressor nation. The defender is not allowed to punish the aggressor’s soldiers once they have surrendered, once they are no longer part of the war. As I noted yesterday in responding to Chris. It’s slightly odd that someone can be the target of lethal force and then as soon as they surrender they can’t be put on trial even though being put on trial seems like a better deal than being subject to the use of lethal force. That’s because the very rationale on the prohibition of punitive war, the very rationale that the only ground of war is national defense, is that each nation, conceived as a legal order, is not subject to the jurisdiction of any other nation as a legal order.

As such, that means that the defender nation is not entitled to put the aggressor’s soldiers on trial for the mere fact of participating in an aggressive war, because it’s not entitled to put them under any legal obligations and so it is not entitled to punish them for violating legal obligations under which it was entitled to put them.

Think about it this way. The claim is not that the aggressor’s soldiers are morally entitled to participate in an aggressive war. The aggressor could not give them permission to use lethal force. It can tell them to do that thing. That doesn’t generate a new permission on their part. One way to see that it doesn’t generate a new permission is that precisely because the defender is not subject to the aggressor’s legal order. The defender can’t limit the rights of the defender’s soldiers. And so the aggressor doesn’t give its soldiers a new permission to do something otherwise impermissible. In fact, the aggressor telling its soldiers to fight is doing something that it has no business doing.

However, the question for the authorization of the defender to put those soldiers on trial is a different question. It is a question about whether the defender is entitled to put the soldiers of the aggressor under an obligation to refuse to participate when the aggressor tells them that they are fighting a legitimate war. It’s not that they are under any moral obligation to believe the aggressor’s claim. It’s that the defender can’t put them under a legal obligation to refuse to believe it. But if it can’t put them under a legal obligation, than it can’t punish them for violating that obligation.

Now, this is a very legalistic notion of punishment that I’m working with, I can see. But I, again, can say — this was as close as you would get to Kant — I think that punishment is fundamentally a legal concept. The question of whether someone deserves punishment must always be secondary to the question of whether the party punishing them has the standing to punish them. And my claim is that a defender does not have the standing to punish an aggressor’s soldiers for the mere fact of taking the aggressor at the aggressor’s word.

In a way, this is a little bit like the situation of someone like a prison guard. A prison guard does something wrong by imprisoning a factually innocent person, but the prison guard cannot be put on trial for that because there’s no one who is entitled to put the prison guard on trial for that. I want to say there’s an analogous structure here. Now, of course, what if the prison guard knows this to be a factually innocent person or, to return to our main topic, what if the aggressor soldier knows this as an unjust war? Then, of course, they should disobey. The stakes in doing so may be so high that it is not reasonable to ask them to disobey, but that’s a question for them. That’s not a question for the defender’s legal order.

In my lecture yesterday, I began by responding to Sha Cross by introducing the idea of rules for wrongdoers. What I suggested is that the rules of war have to be understood as limits on the possibility of war. The prohibition of perfidy is the prohibition on conducting a war that makes it impossible for war to end. The principle of discrimination is the principle that prohibits fighting on a war so that its scope is unlimited. In putting this forward, I am taking a Kantian position against what was for centuries the primary way of thinking about war, the way of thinking about war that we find in the writings of Hugo Grotius in particular. Grotius thought that war was by its nature unlimited. He thought that sovereigns went to war against each other when they had a dispute. There was no one superior to either of them before whom they could submit that dispute, and so instead they fought.

Grotius thought it was inevitable that the subject matter of the war would spread. That it would not merely be a war about the initial dispute, but it would ultimately be a war about who was really in charge. As a result, Grotius also thought that it was inevitable in war and acceptable in war, could be done with impunity, that civilians would be killed, including children, lest they grow up seeking to avenge the deaths of their parents. The modern law of war, the Kantian law of war, and the morality of war are just the repudiation of this Grotian picture of war. They’re the repudiation of the idea that war is essentially unlimited. They are the insistence instead that war must always be limited. That it’s never permissible to move from peace to war and in a condition of war, it’s never permissible to fight in a way such that there is not a limit on who can be attacked, a limit on who is part of the war, and a limit on how the war must end. Thank you.

Kinch Hoekstra: All right, let’s begin with our commentaries. Our first commentator today is Oona Hathaway, the Gerard C. and Bernice Latrobe Smith professor of international law at Yale Law School and director there of the Center for Global Legal Challenges, which she founded. Hathaway earned her BA from Harvard and her JD from Yale. She was associate professor at Boston University School of Law, then associate professor at Yale Law School before her career culminated in 2008, when she was, for a year, professor of law here at Berkeley. But, as happens, the pull of New Haven, Connecticut was too strong and she returned to Yale, where she teaches international law, national security law, and the law of war, amongst other things. She also addresses a range of these topics in popular forums and is an active advocate as well.

She has served as counsel of record in a number of DC Circuit and US Supreme Court cases. Among her other many influential publications, one that I can commend to warmly is her 2017 book that she wrote together with Scott Shapiro called The Internationalists: How a Radical Plan to Outlaw War Remade the World. This was referred to a couple of times yesterday and it’s a really extraordinary book. It not only completely upends a received history of a much maligned 1928 peace back, as what was mentioned yesterday, but also provides a bracing case study that challenges a very widespread view that international politics is ultimately only about power and it really insists on the ethicacy of ideas, and in particular on the potential power of international law. It’s therefore a great pleasure to welcome her to talk on these topics today.

Oona Hathaway: Thank you so much. It’s a real pleasure to be here. Thank you so much to the Tanner committee, and thank you to Berkeley Law for having me back. I have to admit, I was walking through Strawberry Canyon this morning whether what possessed me ever to leave this place, so it’s a real pleasure to have the chance to comment on these lectures, which are really brilliant and thought provoking and challenging. As Kinch rightly said, I’m an international lawyer, not a philosopher, and in many ways I’m an outsider to this conversation. This is in some ways a debate that’s been going on for some time and I feel like I’ve sort of walked into the room while the conversation is three quarters of the way done and I’m trying to catch up on the first part of the conversation. It’s also the case that the conversation is somewhat different from the way in which international lawyers would have it. These issues are ones that, strangely enough, international lawyers and national security lawyers don’t tend to talk a great deal about, and actually that’s part of why I find it so refreshing and appealing to engage with it. It actually left me thinking as I read the lectures that we could each learn quite a lot from one another, so I’m excited about the conversation that we’re going to be having.

As you all know, you’ve been here for two days, Arthur is seeking to formulate a Kantian responses to what he calls Sha Cross’s argument. The argument that the killing of combatants in war is justifiable only where war itself is legal, and in doing so he’s seeking to provide a moral justification for what may appear to be an incoherence in the law, an incoherence that my fellow commentators Jeff and Chris have both written about and explored in some depth in their work. Can the law of war and morality of war be reconciled? Now, one problem for the morality of war seems to be a basic inconsistency between these two central bodies of law, the jus in bello, jus ad bellum. As, again, we’ve talked about over the course of the last couple of days, on the one hand the jus ad bellum has very limited possibilities for going to war, limited reasons for permitting war. On the other hand, the in bello says regardless of whether you’re waging a just war, an unjust war, there’s a set of rules that govern your behavior and you get the same privileges and responsibilities regardless of which side of that line you’re on, which seems somewhat puzzling. How could that be the case?

Arthur is, I think, trying to answer this challenge, which in some ways has been raised by my fellow commentator tonight, Jeff, so he’s going to be, I think, tougher on you Arthur than I am because in many ways Arthur is coming to international law’s defense, so an international lawyer is inclined to be quite sympathetic to the argument. I am approaching this really with a sympathetic point of view, but the question is whether the defense is entirely successful, so I’m going to raise what I think are a couple of questions or concerns that I have about the completeness and the success of the defense that’s being mounted there, though again I come from a point of sympathy to the overall project because I certainly do hope that the endeavor that I’ve been engaged in for the last decade is morally grounded, so I appreciate your effort to defend international lawyers who are doing this work. I raise these questions in sympathy and in hopes that we could find some mutually agreeable answers.

Now, I had a couple of points that I had written in advance that I wanted to raise for today, but yesterday’s conversation provoked a third that I wanted to put on the table. It won’t come as a big surprise to Arthur, both because I warned him I was going to do this and because we had a little bit of this conversation at dinner last night. He’s got his work cut out for him in responding to my comments and Jeff’s comments, both of us together 30 minutes and Arthur only has 15 minutes, so he won’t be able to respond to all of this but I thought I’d put it on the table with the thought that we might get to it tomorrow if not today. This concerns the focus on perfidy, which is consistently coming up as the centerpiece of the initial grounding of justification for the reconciliation of these two bodies of law. I will admit that when I first read the lectures, this struck me as a somewhat odd choice because international lawyers today don’t really talk about perfidy. It sort of seems like this antiquated rule. Of course, it’s still valid. In fact, there’s a case in the military commissions right now for perfidy, so it’s not entirely defunct but it’s not really a central rule in the way of the principle distinction, which is our central focus today is.

Initially it struck me as a bit of an odd choice, but I thought okay, I see the purpose it’s playing. Now I’m a little bit worried it creates some problems for the argument so I want to raise them because I want the argument to succeed. First, focusing on perfidy serves this very useful role because while it marks the line between war and peace, I do worry about the fact that perfidy is so central to this argument and yet it is the tertiary role of international law. I worry, why do we have to rely on a rule that is so … It’s on the list of ICRC’s customary international law study. It’s rule number 65. So, wherever you were lying on rule number 65, not one of the central rules to justify the principle to reconcile this apparent contradiction. If there are no other rules of this character, then what does that say about the reconcilability of jus in bello, jus ad bellum? If there are, why don’t we rely on those? Why the centrality of perfidy to the argument?

Second, I am a little concerned that perfidy isn’t generally defined in precisely the way that you have defined it. I think it’s a little closer to the way that Chris described it yesterday. I know, international lawyer, so I have to quote international law. Article 37.1 of the Additional Protocol First of 1977 says acts inviting the confidence of an adversary to lead him to believe that he is entitled to, or is obliged to accord, protection under the rules of international law applicable in our conflict, with the intent to betray that confidence, shall constitute perfidy. Hundreds of military manuals basically repeat this verbatim, so this is the way in which it’s understood, at least today, by international lawyers. To me, when I read this it’s not so much about making peace. Perfidy’s harm is not so much that it makes peace impossible as it is that it turns the other side’s decision to follow the law into a vulnerability. It’s a fact that you are following the rules, that you are fooled into treating me in such a way that that makes you vulnerable to attack. That’s, I think, what the fundamental harm is of perfidy, at least to me.

Now, of course, I’m an international lawyer, so it’s a fact that you’re using the law and the fact that the other side is following the law to undermine the law. The law, of course, understandably, doesn’t care for that and wants to make that difficult, and those who are writing laws want to make that difficult. To me, that’s a better way of understanding perfidy, and I worry that that raises some concerns for your argument so I wanted to mention that to give you a chance to think about it and invite you say something about it either today or tomorrow if you wish. The two key points that I had intended to raise today and that I will say a bit about are two issues focusing in particular on the principle of distinction, which is the focus of the lecture today. I have a third comment that I’m saving for tomorrow, and I’ll say just a brief word about that. The two key things that I want to talk about today are first, I worry that the defense is incomplete. Second, I worry that it may leave civilians vulnerable. Let me say a little bit about each of those.

First, the aim of this lecture is to morally ground the principle of distinction. Again, I’m incredibly sympathetic to that goal. I worry that the defense of this principle is, at best, incomplete. Arthur argues in his first lecture that, “The modern consensus is that the sole just cause is defense, either of a nation’s territorial integrity or, in extreme circumstances, of endangered civilians in another nation. All other uses of force are prohibited.” This definition of just war then is essential to his justification for the principle of distinction between civilians and combatants today. He begins his argument for the principle of distinction by showing there are two different statuses available. Being part of the war, not being part of the war. Combatants, of course, are a part of the war, although they can exit that status. They can be wounded, so they can no longer be in combat, they’re no longer in the war. They can surrender, then they’re no longer in the war. So, they can exit that. If the only just wars are defensive, then civilians too are not part of the war because they are not part of the aggression. As Arthur put it today, you can start that through which the state is aggressing quote, and he focuses the thing through which they’re aggressing is typically the military.

This is where I come to a puzzle. Does the principle of distinction apply to wars fought by aggressor states? There are at least some reasons I think Arthur’s answer is no. Arthur answers that in a nondefensive war, punitive or remedial war, “Those who are not part of the war will not correspond to that between combatants and civilians.” Arthur then explains, “In a remedial war, the purpose of which is to reclaim things wrongfully seized. Those who wrongfully keep things that were unjustly acquired would be appropriate targets of force.” This seems to entail the conclusion that civilians were culpable for the wrong, or appropriate targets, even if they haven’t taken up arms. In fact, at least by one reading, it seems Arthur might be saying that, though I invite him to correct me if I’ve got that wrong. A remedial war, Arthur explains, is one in which, “The aggrieved party engaged in private enforcement or appointed itself as a proper authority, the targets of remedial force would be anyone, combatant or civilian, who resisted the remedy claimed.” If that’s true, then the modern law of war has a really big problem. Although Arthur says that the key just cause for war is defensive, the modern law of war has more reasons than that.

The modern law of war actually lays out three possible just causes for war. One is self defense, under Article 51 of the UN Charter. A second is that states may participate in uses of force authorized by the security counsel under Chapter 7. The third is states may participate in wars where the host state consents. For example, US current war that’s going on in Iraq and Afghanistan are happening with the consent of the host states. Now, for the sake of brevity, I’m going to focus on the second accepted ground for waging war, UN Security Council authorization. Now, curiously, this is utterly absent from Arthur’s description of permissible reasons for going to war. The UN Charter makes very clear that states are permitted to go to war as long as UN Security Council authorizes it. Under this authority, states do regularly wage wars. To take just one example, in August 1990, Iraq invaded Kuwait, you all remember. The Security Council responded with Resolution 678, which specified that if Iraq did not implement the council’s resolutions relating to the occupation of Kuwait by January 15, member states cooperating with Kuwait’s legitimate government were authorized to use “all necessary means” to compel it to do so. By January 16, 1991, Iraq had not complied and the war began. It ended on February 28th, with Iraq forcefully expelled from Kuwait.

In this sense, I suppose this is a defensive war, in that Kuwait was under a forceful occupation. The member states were acting in cooperation with Kuwait to expel an aggressor. But then, it raises the question why did anyone think that a Security Council authorization was necessary? If it really was a purely defensive war, then Article 51 would have sufficed to justify the basis of the war. In fact, in a way one might think that if Chapter 7 only authorizes defensive wars, then it’s kind of redundant with Article 51. You actually don’t really need it because Article 51 would be sufficient because it authorizes states to act in self defense or collective self defense. I think this poses a problem for Arthur’s argument, or if not for his argument then at least for international laws that currently exist. At a minimum, it means that his theory provides a moral foundation for the principle of distinction that is incomplete at best. After all, it would seem to allow states not to abide by the principle of distinction in wars fought by states acting under and within Chapter 7 authority, thus legally under modern international law, though not exclusively defensively.

This connects to a concern that I will elaborate in much greater detail tomorrow, but let me just very briefly preview it. It is where does international law and international institutions created by that law fit into Arthur’s argument? As I say, I’ll focus my remarks entirely on this question tomorrow, so I’m not going to say much about it here, but I do find it puzzling that a project of morally grounding the rules of international law of war is so focused on the independence of states and repeatedly is focusing on, we heard it again today, the necessity of states remaining entirely independent from one another, if not exercising authority over one another, when international law, of course, is entirely about states cooperating and sometimes exercising some degree of authority over one another, at least cooperatively in engaging and creating institutions that then have the capacity to exercise some form of authority over one another and it’s just not clear where that is in the argument. I don’t want to say that it necessarily is denied by the argument, but it seems at times to be at least potentially so. As Arthur said today, each nation, as a legal order, is not subject to any other nation’s legal order. So, where does that leave international law? It’s not clear. But again, that’s tomorrow.

My second concern today is does the moral foundation Arthur provides for the principle of distinction actually protect civilians in war? Arthur argues that members of the military may be subject to attack even if they’re unwilling participants because they are a part of the attack. Civilians however, he says, are different. “They’re not part of the attack. Even if they voted in favor of it, cheered on, and so on.” They may be to blame for it, but, “Force cannot be used against them because they are not part of the attack, and so they cannot be a way of stopping the attack.” Is that really true? What of the civilians that cook for the forces? Drive them? Fuel and service their tanks? Keep the weapons in working order? These are all tasks frequently handled today by civilians, usually private military contractors. For that matter, what about the civilians who work in The Pentagon, approving orders for supplies and transport? Not to mention the lawyers who approve the attack. I worked for a year in The Pentagon as special counsel to the general counsel working on exactly these issues. So, was I an appropriate subject for attack? It seems, at least potentially, so.

They are arguably as much a part of the attack as the soldier driving the tank to the front. Why is attacking them not a way of stopping the attack? Indeed, if one could stop the Army from being fed, fueled, and supplied, and lawyered, one could certainly stop the attack. In short, even in states that do not operate on the model of Sparta, which is mentioned in the written version of the remarks, civilians are almost as integral to the attack. If that’s true, does the principle of distinction hold? Now, just a very brief aside about the implications of this as they relate to non-international armed conflict. Arthur’s not discussing non-international armed conflicts, or what international lawyers call NIACs here, and I will note that this very issue, determining which civilians are so much a part of the war effort as to be targetable is a central debate right now among international lawyers when it comes to NIACs. The United States government currently takes the position that the cook, the driver, and others who support the fighters in an organized armed group are legitimate targets in a non-international armed conflict.

In particular, the US engages in membership based targeting, in which it targets persons based on membership in an organized armed group that also engages in signature strikes, that is striking those who engage in the behavior that appears to signify membership, even if he or she does not directly participate in hostilities on behalf of the group. At least one justification of this approach is that the members who do not themselves engage in military activities are supporting and enabling those who do. Now, the International Committee of the Red Cross, together with many international lawyers, including myself, think this interpretation is dangerously overbroad. This agreement is probably one reason for wildly divergent civilian casualty numbers reported by human rights groups in the US government. I worry a bit that the distinction that Arthur is drawing here is open to that same vulnerability.

In a sense, he’s throwing his hat in with US government here, that if civilians are part of the attack, then they can be stopped. Is that really where we want to go? Does that open civilians up to vulnerabilities that I know are unintentional but I worry maybe the implication of the argument. As I said, tomorrow I’ll say more about the implications of Arthur’s arguments for international law and international organizations, but since those issues go to the broader argument and not simply the second lecture, I will save those points for our final conversation. I look forward to the rest of the conversation today. Thank you.

Kinch Hoekstra: Great, thank you Oona. We turn to our final commentator. Jefferson McMahan hails from the rural south and he earned his first degree from The University of the South in Sewanee, Tennessee. He then took a Rhodes Scholarship to Oxford to do PPE and then his PhD in philosophy at Cambridge, although it was a bit of a mixed degree between Oxford and Cambridge. He then spent 17 years as a professor of philosophy at the University of Illinois, and then 11 years as professor of philosophy at Rutgers University. In 2014, he took up the White’s professorship of moral philosophy at the University of Oxford. Professor McMahan is best known for his wide ranging work in the ethics of war, but he’s also done tremendous work in other areas where theoretical clarity is needed to think straight about some of our most pressing moral dilemmas and our greatest moral disasters. He’s given us penetrating work on the morality of nuclear weapons, population growth, abortion and infanticide, disability, torture, the treatment of animals, assisted dying, and very much else. It’s unflinching work. It’s unflinching in its topics, it’s unflinching in its principles, and it’s also unflinchingly philosophical. Please welcome Jeff McMahan.

Jeff McMahan: Well, thank you. I had lots of expressions of gratitude, praise, and jokes and things like that, but we’re running behind so I’m going to skip all of that. I should say however, that some of what I’m going to be talking about is in Arthur’s lecture text but wasn’t mentioned in the lecture today, so I hope you can follow what I’m going to say. I’ll try to make it clear to you. Let me begin by saying that I agree, or partially agree, with two of Arthur’s main claims. The first is that participation in a war with unjust goals is impermissible. Secondly, that there are in bello prohibitions that apply both to just and unjust combatants, and that unjust combatants who violate these prohibitions commit additional and distinct wrongs over and above the wrong of merely fighting in the unjust war. Now, it’s this last claim that I agree with only partially, and that’s because I think that some of the in bello restrictions take this form that is they can be either respected or violated by unjust combatants and some do not. Some of them cannot be obeyed by unjust combatants. I think, for example, that perfidy is an example of a prohibition that applies equally to just and unjust combatants and can be obeyed by both.

I think that’s true of at least some of the constraints on the treatment of prisoners of war. I don’t have time to go into that, I think that’s rather complicated. I think it’s also true of a rather weak version of the requirement of necessity because even if unjust combatants are fighting for unjust goals, they ought not to cause unnecessary harm in pursuit of those goals. They ought not to cause harm that they don’t need to cause in order to achieve their unjust goals. I do think however, that two of the in bello rules, namely discrimination and proportionality, are problematic in their application to unjust combatants, but only when those requirements or prohibitions are properly understood. What I mean there is that I want to disagree with Arthur’s understanding of those principles, and his understanding of those principles does coincide with the understanding that one finds in the traditional theory of the just war and in the contemporary law of armed conflict, so I’m going to be criticizing traditional just war theory and the moral foundations of the law of armed conflict.

What I think is that, except on very rare and limited occasions, unjust combatants, that is combatants who fight for unjust aims, can’t fight at all without violating the requirements of discrimination and proportionality. I’ll mention proportionality first and then go into discrimination. Arthur follows Geneva Convention Protocol 1 when he says that, “Proportionality is the requirement that any side effects of military action be proportional to their military advantage.” He doesn’t discuss proportionality or necessity in the lectures, but I think proportionality poses a problem for his view. The problem is that military advantage in itself has no value. It’s neither good nor bad. Military advantage is good or bad only instrumentally, and whether it’s good or bad depends on what it is advantageous for. If military advantage is advantageous for the achievement of just aims, then it is impartially good. But, if it is advantageous for the achievement of unjust aims, then it is impartially bad. The important point here is that effects that are impartially bad cannot be coherently weighed against and cannot coherently morally offset side effects that are also bad. For example, the Nazis couldn’t coherently have claimed that their killing innocent civilians as a side effect was morally offset by the military advantage the action gave them in conquering more territory and killing more Jews just makes no sense. That’s not a morally coherent view. So that’s all I want to say about proportionality, I’ll come back to it a little bit at the end in saying what I think the implications are for Arthur’s view. I don’t think they are really in any way fatal to Arthur’s view.

I want to say something now about discrimination. What I want to say here, I think, is more seriously challenging to the view that Arthur defends. Arthur identifies the requirement of discrimination with the principle of civilian immunity; that is what discrimination prohibits is the intentional attacking of civilians. Now I think, by contrast, that mere civilian status doesn’t confer any immunity to attack. One of the differences between Arthur’s view and mine is that I think that moral immunity in war is grounded in what one does, Arthur’s view is a more status based account of the bases of immunity.

So in my view, discrimination comprises both a permission and a prohibition. The permission is not what Arthur refers to as a novel permission, it’s just the familiar permission to harm those people who are morally liable to be harmed. The prohibition is, in my view, a non-absolute prohibition on attacks on people who are not morally liable to be attacked, and in my view the criterion of liability to attack in war is simply moral responsibility for a threat of wrongful harm. This view of discrimination is one I’m not going to try to defend today; the defense would take too long. The important point is that on this understanding of discrimination, the distinction between people who are legitimate targets of attack and people who are not legitimate targets of attack does not align with or coincide with the distinction between combatants and non-combatants. On this view, just combatants are in general not morally liable to attack at all, and some civilians on the unjust side may be morally liable to attack in war.

But if it’s right that soldiers who fight only for just aims by permissible means are not morally liable to be attacked, then it follows from that that unjust combatants cannot engage in fighting against just combatants without violating the requirement of discrimination. This is a requirement that unjust … It’s an in bello requirement that unjust combatants simply can’t satisfy, except in rare instances for example, in which just combatants are engaged in the pursuits of their just aims by impermissible means, then it might be permissible for them to … Then they might be liable to attack.

But of course, on Arthur’s understanding of the requirement of discrimination as civilian immunity … unjust combatants, obviously, can respect the requirement of discrimination. They can simply avoid attacking civilians on the just side, or in neutral countries or whatever. So whether or not unjust combatants can act in conformity with the requirement of discrimination depends on how it’s understood, and how we are to understand it depends on whether I’m right or Arthur’s right, so a lot depends on whether Arthur’s arguments for his understanding of discrimination are good arguments.

Now, one of the problems here is that Arthur didn’t give you those arguments in the lecture today, though they are in the lecture text, so I’m going to review three of them. The first one is that civilian immunity is implied by Arthur’s conception of the grounds of self defense. You did get a bit of that today; he didn’t give you the label, he calls it the protective model of self defense. What he says about it is … I’m quoting from the lecture text now, “You can stop someone who attacks you because you’re entitled to be independent of them. Your right that you not be subordinate to the attacker’s purposes is unconditional. Your right is that you not be subordinated to their choices, including their non-culpable choices.” So in this view, the fact that someone poses a threat seems sufficient to justify defensive force. Nothing depends, as he said in the lecture, on whether the attacker or aggressor forfeits a right not to be harmed. The whole purpose of force and the whole justification for the use of defensive force is to stop the attack, is to stop the harming.

So on this view, the permissibility of self defense is grounded entirely in the victim’s unconditional right not to be subordinated to the will of the attacker, even if the attacker is not culpable; doesn’t depend on the attackers having forfeited a right not to be attacked or harmed. But an account that … grounds self defense in those two features has the highly implausible implication that a culpable aggressor has a right of counterdefense against the defensive action by his innocent victim. Now, Arthur’s not going to think that’s true, but I’m claiming that follows from the claims that I’ve cited. Arthur also says that this protective model of self defense applies not just to individual self defense, but also to national self defense; same principle applies at both levels, therefore … as I’m trying to extract the implications of the view, what it implies is that aggressor states have a right of counterdefense against victim states.

What that means is that while unjust combatants act wrongly when they engage in initial aggression, because an initial aggression they’re not responding to any kind of threat against them … Nevertheless, if the victim state, the just combatants for example, fight back, then the unjust combatants have a right of counterdefense because they have a right to be independent from the choices and the will of those who are attacking them. Nothing hinges on whether they have forfeited a right or not; they have a right coming into things not to be attacked. We all have that right, I take it. They don’t forfeit that right by attacking someone else wrongly.

Now, one more implication of this; Arthur has insisted, and I agree with this very strongly, that people who fight in unjust wars are engaging in a prohibited activity. But what I’ve just pointed out as, I think, an implication of the protective model of self defense is that unjust combatants do have a right to fight defensively, or counterdefensively, against the defensive violence of their victims.

That’s one objection to the protective model; let me give you one other objection which consists of a counterexample. I’ve given this counterexample in an article I’ve published a long time ago, but I doubt that any of you read it so I’ll give it anyway. The idea here is that … Imagine a hypothetical case. There’s one person I’m going to call the initiator; what the initiator does is forcibly implants a chip into someone else’s brain. I’ll call the other person who gets the chip the pursuer, and that’s because the chip is programmed to make this person try to kill me. This person is pursuing me trying to kill me involuntarily, only because of the chip in her brain. We can imagine now the initiator, the guy who did this to her, is bedridden and can’t do anything to stop her.

No time to call the police, I have two options. First option is kill the pursuer, who’s a non-responsible threatener. This is somebody who’s threatening me, but is in no way morally responsible for that; on my view, not liable to be attacked, or I can kill the initiator. That is, I can use the … I can go grab the initiator out of his bed and use him as a shield in a way that will harmlessly incapacitate the pursuer but will sadly kill the initiator. Now, the protective model of self defense says that I’m permitted to kill the pursuer but not the initiator. I think the reverse is true; I think I am permitted to kill the initiator, and given that I am permitted to kill the initiator, that’s what I ought to do. Therefore, I ought not to kill the pursuer. I claim counterexample to the protective model.

Second justification that Arthur gives for civilian immunity is that he claims that targeting civilians presupposes a principle of total war. I think there was a slight suggestion of that in the lecture today, so what he … He also says that the prohibition of targeting civilians is protection against war without limits. I think, in his lecture today, he spoke of … the prohibition on attacking civilians is … intended to ensure that we don’t have a principle that licenses unlimited scope to war. My response to that is simply that … it’s just not true that a principle of discrimination that permits intentional attacks on some civilians but not others implicitly licenses total war. It too imposes limits on the use of force, it’s just that they are different limits, and in fact I think … I’m not going to try to defend this here, that the counterdiscrimination that I sketched for you a short while ago is actually much more restrictive than the principle of civilian immunity. That is, it limits the permissive targets even more than the principle of civilian immunity does, therefore being more restrictive it’s even less likely to … endorse total war.

The third argument that Arthur gives in the lecture did come up both in the talk today and in Oona’s comment, and that is that the requirement of civilian immunity is grounded in this fundamental distinction between being a part of a war and not being a part of a war. What I want to say about that is first of all, I think that that phrase is vague. When I ask myself, “Is somebody or is somebody not a part of a war,” I’m not sure what to say.

Let me give you an example, and I’m going to give this also as a counterexample. Imagine that there’s … a physicist working in Germany during the Second World War, whose aim is to create an atomic bomb for the Nazis to use. Is that person a part or not a part of the war? I would think that Arthur’s view would be no, not a part of the war. At any rate, by any plausible understanding of what a civilian is, a physicist in a university laboratory working to try to create an atomic bomb; not wearing a uniform, not carrying arms openly, not actively engaged in combat, and so on and so forth, is not a part of the war.

Now on my view, however, this person lies on the vulnerable side of the relevant distinction. This person is morally liable to be killed. Ask yourself this: here’s this physicist, he’s just about to be able to … complete the math that’s going to enable him to provide Hitler with an atom bomb, that Hitler will then be able to use to win World War Two. The only way to stop him from handing over the bom to Hitler is to kill him. I think obviously permissible to kill him, has no claim to immunity just because he’s a civilian. He’s morally liable to be prevented from making this enormous contribution to the war, though he is a civilian. I think, in this case, Arthur’s understanding of the requirement of discrimination gives the wrong answer.

There’s one other way in which … Again, you didn’t get this, but the way Arthur … The criterion that Arthur provides for determining whether someone is or is not part of the war is this; you have to determine whether the activity in which this person is engaged is an activity in which people engage in peacetime as well as in wartime, and so … The example he gives is a medic in war — the medic is someone who’s out there with the troops, but Arthur and international law say not part of the war, just there to do medicine, that’s what this person does in peacetime as well. But now imagine … I think that’s actually not true of any medic, but to make the point … vivid, let’s imagine a hypothetical case in which the defending country has weapons that will only wound the aggressors. Won’t kill them, but it’ll incapacitate them for the remainder of the war unless they’re patched up and sent back in by the medics. There are the medics working away; the unjust combatants are being wounded and would otherwise be out of the picture, but the medics are healing them and sending them right back into the war again within a few days. Now, are they part of the war? Yes, because they’re thwarting the military action of the just combatants, and stopping them is part of, and indeed essential to, stopping the attacks.

Now, I’ve got to stop, but let me just say in summary: Arthur and I agree that unjust combatants are engaged in a prohibited activity. We agree that those who engage in perfidy and mistreatment of POWs are engaged in distinct wrongs, violating distinct prohibitions. I think that’s not true of the requirements of proportionality in discrimination. I think that … that it’s an entailment of the ad bellum prohibition that soldiers who fight in an unjust war will violate proportionality in discrimination. So I’m disagreeing here strongly with Michael Walzer, who says that the in bello rules and the ad bellum rules are logically distinct; it’s not true of discrimination in proportionality. But really, that shouldn’t bother Arthur all that much, it’s just … We agree that they’re doing wrong, it’s just are they doing an additional distinct wrong? No, but why should we care about that? Thank you.

Kinch Hoekstra: The good news is that you’ll have plenty of time to respond tomorrow, the bad news is you only have a few minutes to-

Arthur Ripstein: Okay so … Just watch me, as a famous prime minister of ours once said. So I’m very grateful to all three of my commentators; each of them has had a huge impact on my thinking about these topics, so hearing their responses you might think that each of them is of the opinion they’ve had insufficient impact. But let me just say a very few things in a general way, in response to each of them.

The first thing, I think … Perfidy, we will talk about in more detail tomorrow. My general thought about perfidy is it’s analytically basic. I actually believe that in the new types of war that we encounter today, you have to understand the distinction between … acting in a way that is respectful of the possibility of future peace as the fundamental thing that goes wrong on both sides of those wars. I actually believe that both the way in which members of non … in nyacs, as Oona called them, non-state actors move back and forth fluidly between combatant and civilian status and then take advantage of that, is a version of the wrong of perfidy. I also believe that that is in a certain way a kind of secondary wrong, building both on the principle of distinction and the wrongfulness of perfidy, and I think that there is … in addition to that secondary wrong, a tertiary wrong committed by state actors, of taking the fact that it is possible to move back and forth to inculpate anyone who is in any way causally related to the fact of war. I think that that’s a wonderfully important analytical point, but I think that that is in some sense in support of what I’m trying to do.

I also think that Oona’s characterization of the additional grounds of war … It’s really obnoxious in this context to say this is part of a much larger project, and I talk about all of those things there, but here I go, and I think that there is a reason to suppose that coming to the aide of another state is just an instance of defensive force. I have a Kantian inspired view about the role of international legal institutions. Kant was an early … supporter of the idea of a league of nations, something that didn’t quite do its job when something was given that name. But my view is that the basic ad bellum and in bello rules that I have been talking about mark a transition from a state of nature between nations, that is a condition of barbarism, to one that is a condition of mere anarchy.

I think, as Kant did, that you need to exit a state of nature and enter a public rightful condition with international legal institutions, because war is not the way in which states should resolve their differences. I think that once you do that, you get an additional layer of legality that applies to the conduct of states, but I also think that it leads to no further or other duties between states in relation to each other. I think being subordinate to international legal institutions is a great thing, but it is not a way in which one individual state is subordinated to another. If one state comes to dominate those international institutions, that is a defect in the realization of this set of ideas.

Now, as far as the … Most of … Oona also made some comments about discrimination, which I will pick up on in more detail tomorrow, but I want instead to turn to the broader issue about … the issue of discrimination that both Oona and Jeff brought up, and one way of thinking about it is that Oona’s suggestion was that there was a worry about discrimination, because when you have a non-defensive war then they’re fighting some other kind of war. So imagine defender is defending itself against aggressors that tamp to exact a remedy; the question is why doesn’t the rule for remedial war apply to the aggressors’ combatants? My answer to that is that their wrongfulness of remedial war means that the only way we can understand a remedial war is as a defective version of a defensive war, and so they’re asserting a claim of right … They claim to be entitled to defender right; they’re wrong about that, but we have to understand them as operating under the rules that would apply if they were entitled to do it, and so if they were in a defensive war.

I’m delighted to hear that Jeff agrees with me about at least one sentence. I sometimes worry that, as fond as I am of him and as greatly as I admire his work, if we disagreed about one more sentence we’d be incapable of regarding each other as language users. I’m just absolutely delighted by that.

What I want to say about the issue of discrimination, I can refigure by saying something about what Jeff said about proportionality. Jeff’s basic thought about proportionality is that proportionality is to be understood as a kind of balancing of good effects against bad effects; it’s a kind of weighted consequentialism. You’re creating objective value and so you’re allowed to, as a side effect, create objective disvalue. I think of proportionality differently as does international law, as indeed do those legal systems like Canada’s and Germany that make proportionality a fundamental constitutional principle. On my view, proportionality is a restriction on the use of means that would otherwise be permissible, and it is a way in which an act can be defective; it is never something that counts in favor of something.

That’s why in the post World War Two constitutional analysis, the question of proportionality stricto sensu is always the last question asked, it is never the first question asked. The question is always whether you are entitled to do this, is it this, is it that, does it satisfy … Is it for a proper public purpose, and then there’s a question of proportionality. But that means it is a way in which an act can go wrong even if it is defective already in some other way. Disproportional force is objectionable whether for a good cause or a bad cause; there’s an extra defect created by disproportion even if proportionality isn’t a positive principle of justification. I want to say the same kind of thing about discrimination; there’s an extra defect about indiscriminate force, and that is why it is problematic.

I want to close, because Kinche keeps looking at his watch nervously … He even took it out and is turning it into brass knuckles, which seems kind of thematically appropriate to the occasion, but at the same time mildly menacing. So it’s going to work, but here’s the thought about the protective model; I don’t think that it applies to both sides. My claim about the protective model is not protection against harm, it is protection against wrong. You’re entitled to protect yourself against wrong, that is why you’re entitled to resist with right another who aggresses against you, but that doesn’t confer on the aggressor any kind of entitlement to use comparable force.

The reason I reject the vocabulary of forfeiture is that despite sounding very official, the vocabulary of forfeiture has no analytical work to do in this context. If you forfeit something you can only forfeit something you already had, but nobody had a right to be secure against defensive force and indeed, on the view that I want to defend, no one has a right that force not be used against them. Rights are members of a consistent set of rights; everyone is entitled to interact with others on terms consistent with the freedom of everyone. Everyone is therefore entitled to resist with right those who seek to interact with them on terms inconsistent with the freedom of everyone, that is who seek to subordinate their choice to the choice of another. But if that’s why you’re entitled to defend yourself, then the defensive permission is not symmetrical; it applies only against an aggressor.

As far as the initiator and the pursuer goes, I used to worry about these science fiction kind of examples, and indeed when I first read Jeff’s article with this I thought, “Oh, another science fiction example.” But there’s been enough technological change that it’s no longer science fiction, but here is what I would say about it. I would say that in the case of the initiator and the pursuer, the pursuer is not acting. The pursuer is, as it were, the means through which the initiator acts, and as the means through which the initiator acts the initiator is the agent of wrongdoing, and so in the … I find it much easier to imagine the microchip than to imagine it actually working, getting someone out of their sick bed to use as a human shield, but imagining that can happen you can use that person as a projectile and so you can use them as a shield. I don’t think that that is a problem for my model.

I want to end by just saying something very quickly about Jeff’s claim that my distinction between being part and not being part of the war is in some sense vague. I think there’s a sense in which that’s true, but I want to draw your attention to the fact that it’s vagueness, though perhaps vague is not the word I would have preferred for it, but rather the way in which it is imprecise and requires further specification, is itself a fundamental mark of almost all practical concepts. One of the things about the concept of wrongdoing, one of the things about the concept of property, one of the things about the concept of privacy, one of the things about the concept of reasonable care or of agreement in a contract, is that in every case there are highly abstract concepts; concepts we can articulate through some idea of interaction between free and independent human beings.

Then, one of the reasons that we need to, as Kant puts it, exit the state of nature and enter into a legal order, is that we need a common interpretation of them; we need law for those things. I think that morality is fundamentally incomplete without law, and so I think that these abstract categories help provide us with the appropriate form of thought through which we can analyze our interactions with each other, through which we can understand the morality of our situation even though different legal orders might give different interpretations to them.

Then when it comes to international law, which regulates how things stand between different legal orders, we need a common articulation, but the starting point for that common articulation must be the idea that no legal order is subordinate to another, precisely because the part of morality that governs how we interact with each other is structured around these abstract concepts. We need a legal order to impose closure on our interactions with each other, and the imposition of closure must itself be closed; that is, there can only be one closure imposing agency, and so each closure imposing agency, each legal order is entitled to be independent of another. That’s why war has to be defensive, and that’s why we need the in bello rules that we have. Thank you.

Kinch Hoekstra: Thank you very much. Thanks to Arthur, Oona, and Jeff — fantastic session. They’ll each get one more bit at the apple, as will Chris Kutz, tomorrow, and we’ll have plenty of time for questions and further discussion then, so please come back tomorrow. Thanks again to all.