How Rights Move: Losing and Acquiring Rights in the International Domain
Carnegie-Uehiro Fellowship Program, Second Annual Lecture, 2009
November 11, 2009
Please go to the bottom of the page to download the lecture handout.
JOEL ROSENTHAL: Good evening. Welcome. I'm Joel Rosenthal, President of the Carnegie Council. It's my privilege to welcome all of you, and especially our friends from Japan and from the United Kingdom, and originally from Australia.
Last year we inaugurated the Carnegie-Uehiro Fellowship Program, with our First Annual Carnegie-Uehiro Lecture. As I said at the time, it's a rare privilege to introduce a first annual program, especially in an organization that is 94 years old. There's a certain boldness that goes along with the label "First Annual." But raised expectations are sometimes a good thing. The events of this past year and our growing research with our fellows and with the Uehiro Foundation give us every reason to think that great expectations are warranted.
As I also mentioned last year, it's precisely opportunities like this fellowship program and this lecture that make the Carnegie Council such an exciting place to be. As an educational organization working in the public interest, we have an ever-present opportunity to renew our mission in new ways, building on our historic legacy and achievements. We work in the spirit of adventure and creativity, knowing that our work is organic, growing, and always unfinished. Our founders expected us to be ambitious and practical, to think big and to achieve specific results. With these ideas in mind, we created this new program.
We work in the spirit of adventure and creativity, knowing that our work is organic, growing, and always unfinished.
Of course, we realize that the first and most important aspect of this work is that it is made possible by the benefactors who believe in the transformative power of the ideas that we discuss and promote. I would like to thank publicly Mr. Tetsuji Uehiro, the current Chairman of the Uehiro Foundation. Mr. Uehiro is not able to be with us this evening, but he sends his personal regards from Tokyo. With Tetsuji Uehiro's support of this program and many others like it, he has affirmed his family's deep commitment to ethics and education.
I have had the privilege of knowing Tetsuji's father, Mr. Eiji Uehiro, and I've heard many stories about his grandfather, Mr. Tetsuhiko Uehiro. It is a privilege for me to have the opportunity to work in this family's tradition. Their work has brought honor to their family, to their country, and to the world. It is a great example to us all.
Noboru Maruyama is the Secretary-General of the Uehiro Foundation. Noboru, thank you for being here.
I've never met a more visionary and practical person. Noboru combines the deep respect for traditional values and a strong belief that we should adapt to new circumstances and embrace new opportunities. We are grateful to him for sharing his knowledge and insight with us. Most of all, I would like to thank you for your friendship.
The purpose of the Carnegie-Uehiro fellowship is simply to promote the study and teaching of ethics in international affairs. Fellows are required to conduct their own research and to work with us in making their findings available to worldwide audiences. The fellowship is founded against a background of concern and the recognition of a great opportunity. The concern is central to our mission: Who are the thought-leaders in asking and answering the most important and challenging problems of ethics on issues of global scope and concern? Can we identify and encourage the very best scholars to generate new ideas, to improve debate in academic and policy communities?
The great opportunity exists in giving voice to these scholars. Given the revolution in communications technology, we can now connect programs like this one this evening to audiences all around the world. The words spoken here this evening will be recorded in digital video and audio formats, print-friendly transcripts will be produced, and people all over the world will watch, listen, or read what is said here. They will have access anytime on demand and they will have the ability to respond, using interactive tools that are becoming familiar to all, especially to younger audiences.
So it is with this rather expansive idea that we begin. Through the work of the Uehiro Foundation, we are able to build on the relationships that they have created between Tokyo, Oxford, and New York. This is a great place to start, especially as we reach out to new friends and old all over the world.
Finally, allow me now to introduce our speaker, David Rodin. David is one of Oxford's most distinguished ethicists, with broad interest in topics related to human rights and justice. David is one of those rare scholars who have the ability to reach out beyond narrow research communities to engage issues of broad public concern. He counts as his areas of specialization the ethics of war and military ethics, as well as business ethics. And this is what really impresses me: He counts among his areas of competence philosophy of law, epistemology, and metaphysics, Kant and Plato.
I can think of no one better to address us on the topic of how rights move. Welcome, David. Welcome you all.
DAVID RODIN:Joel, thank you very much for those very, very kind words and that very warm introduction.
It's a tremendous pleasure and also a tremendous honor and privilege to be here with you today. When I gave the inaugural Carnegie-Uehiro Lecture last year, I felt that I was somewhat on probation. I'm very, very grateful for the opportunity to be with you here today.
It's wonderful to look into the audience and see so many friends and distinguished colleagues here as well. So thank you very much.
At the center of morality stands a set of core negative rights of human beings. The precise specification of these rights is open to debate, but they include at least the following:
- My right to life—that is to say, my right that you not kill me.
- My right against physical assault, against rape, against torture.
- My right not to have my liberty curtailed through detention, imprisonment, arbitrary restrictions on my movements.
- My right to the possession and enjoyment of my property.
These rights are protected in all just legal jurisdictions. But their existence is not dependent upon legal recognition. These rights are pre-legal and moral in form. In an older lexicon, they are natural rights.
I refer to these rights as part of the "moral minimum" because respect for these rights sets fundamental limits on what it is to be a minimally decent human being or a minimally decent society. As such, they stand in contrast to those other parts of ethics which concern how humans may attain fulfillment and happiness, realize virtue, and contribute to the needs and happiness of others. No matter how rounded my character, no matter how good I am as a father, husband, or brother, how patriotic, how charitable or generous, if I violate these core rights of persons, I have not lived well. In fact, I can fairly be described as having lived the life of a moral monster.
But if humans have core rights which possess this central importance, how is it possible morally to inflict upon humans harms that are protected by these rights? It's a crucial question, because many actions, both personal and official, domestic and international, involve inflicting harms that are indeed protected by these core rights.
Punishment involves limiting liberty. Restitution involves depriving people of property. Self-defense, forcible humanitarian intervention, and war, crucially, involve depriving people of life.
The answer to this question is that rights are not simply static, specifying claims and duties at a particular point in time. They also alter over time. In the terms that I will use in this talk, rights can move. Even some of the rights that form part of the moral core—the rights to liberty, to property, to life itself—can be lost, altered, or overridden. Consider these familiar examples.
The first example is organ transplant. Five people lie critically ill in a hospital and will die if they do not receive immediate organ transplants. The only way to get the organs is to kill an innocent person. Clearly, it is not justifiable to murder one person in order to save five others from death. To do so would be to wrong that person in the most profound way.
Consider now the case of self-defense. One innocent person is attacked by five culpable aggressors. The only way that he can save his own life is to kill the five attackers. The victim here is not only permitted to kill in order to save life, he's permitted to kill five persons in order to save his own one life. Moreover, when he kills the aggressors, although he harms them, he does not wrong them. He does not violate their right not to be killed.
Consider now a third case. A hiker is lost in the woods. The only way to save himself from starvation is to break down the door of a mountain hut and to steal food from inside. Clearly, the hiker is permitted to do this, even though it contravenes the property rights of the hut owner. The action, although, in a sense, it wrongs the owner of the hut, is justified by a principle of lesser evil.
So a number of questions naturally arise when contemplating these and other examples. When do rights stand fast against aggregate welfare, as they do in the case of organ transplant and self-defense? When do they give way to aggregate welfare, as they do in the hiker case? Why do core rights against harm sometimes yield in a way that generates a liberty to inflict harm, as they do in the case of self-defense?
These are all dimensions of the problem of how rights move.
In this talk, I want to explore the ways in which these core personal rights move or, alternatively, stand fast in relation to competing considerations. In particular, I will defend three claims.
The first is that rights form a remarkable coherent and self-remedial system which contains internal mechanisms for determining when core rights against harm may be lost or forfeited, generating liberty rights to inflict upon persons harm. Central to the self-remedial system of rights is the mechanism of self-defense or defensive rights more broadly. I will develop a general model of defensive rights explaining how liability to defensive harm is morally grounded. I will show that defensive rights have a general structure which can be understood as a rich form of proportionality relationship between two individual persons in a situation of conflict.
Moreover, and somewhat surprisingly, this deep structure is shared with consequentialist, lesser-evil justifications for inflicting harms. Both defensive rights, a central element of deontological ethics, and the lesser-evil justification, a central element of consequentialist ethics, possess a shared underlying structure as a form of proportionality relationship, though one interpreted under different conditions, crucially.
Third, I will show how a distinction between different forms of harm, those that are fungible and may be compensated for and those that are non-fungible and cannot be fully compensated for, can explain why we allow lesser-evil justifications to override certain kinds of rights, but not others, and in particular, why we are reluctant to allow lesser-evil justifications for killing or inflicting grievous bodily harm.
Finally, I will explore the implications that the system of rights has for our understanding of the ethics of war.
Now, let me first lay out some preliminary distinctions and concepts. As we have seen in the three examples considered above, rights sometimes give way in the face of competing moral considerations. We can divide the mechanisms that cause rights to move in this way into those that are extrinsic to the system of rights and those that are intrinsic.
Extrinsic mechanisms appeal to moral considerations that lie outside the domain of rights. Considerations of consequences or the lesser evil, as we saw in the hiker example, are the most familiar form of extrinsic mechanism, but they are not the only one. Considerations of virtue, of fairness, of equality, of justice may all conflict with personal rights, and on occasion, they may be important enough to override them. In such cases, it is claimed that the extrinsic moral consideration is of such extreme importance that it overrides the right. As in the hiker example, it's characteristic of extrinsic mechanisms that, although the action contravening the rights is, all things considered, justified, the right holder is still wronged. His right is still transgressed. The right, though overridden, still persists, and to mark this, we often say that the right has been infringed rather than violated. As we shall see, infringed rights often reemerge as claims for reparation, compensation, or apology.
Rights may also alter the configuration through mechanisms that are intrinsic to the system of rights. Intrinsic mechanisms have two basic forms. First are consent-based mechanisms. A right holder may transfer or alienate his right through consent and its variants—agreement, contract, sale, lease, waiver, and so forth. Although complex in their operation, consent-based mechanisms are straightforward to understand. One of the most important roles of a system is to protect human agency. It's not surprising, therefore, that their configuration can be voluntarily altered through the consent of the right holder.
More perplexing are non-consent-based mechanisms for altering the configuration of rights. They will form the central concern of my talk this evening. There are three different forms of non-consent-based mechanism:
- First, defense. A person may lose a right by becoming liable to harm inflicted in the course of defensive action. This may be either self-defense, defense of your own rights or interests, or it may be the defense of others, third parties.
- Second is the mechanism of redress or, in the legal context, torts. A person may lose rights by becoming liable to make redress or pay compensation for some prior transgression.
- Third is the mechanism of punishment. A person may lose rights by becoming liable to punishment.
What is distinctive about these three mechanisms is that each consists in a form of liability to harm and each functions as a remedy against the transgression of rights or another form of moral or legal wrong, either by providing a way to prevent the transgression—defense and punishment—or by providing a way to restore the interest protected by the right once it has been transgressed—redress.
By describing these mechanisms as liabilities, I mean to signify four things.
The first is that to describe a person as being liable to harm is to say that the liable party's right against being harmed is no longer extant. A person who is harmed through appropriate defensive action or punishment or by being compelled to undertake redress is not wronged by the infliction of this harm. His right against harm has been transformed into a localized liberty for others to harm him in particular ways. In this sense, they are paradigm cases of how rights move within the system of rights.
Second, each of these three mechanisms is, in a distinctive way, a normative consequence of an agent performing some transgressive action. The fact of performance and agency is crucial here. Liability to harm within the system of rights is assumed, not attributed, and liability can only be assumed through agency, although the precise quality of this agency is hotly debated. Does it require culpable agency, morally responsible agency, causal agency, or merely minimal human agency? There are significant debates about that.
The third thing that I mean to signify by these mechanisms as liabilities is that, unlike the concept of desert, to say that someone is liable to be harmed is to say that the harm has an instrumental purpose. One is liable to be harmed only in relation to the achievement of some further good or goal. As the Rutgers philosopher Jeff McMahan says, the goal is internal to the liability, in the sense that there is no liability except in relation to some good that can be achieved by harming the person. It's not the case that the harm is considered a good in itself.
As we have seen, in the case of defense, redress, and punishment, the goal of liability is to function as a remedy for the transgression of rights. Defensive rights and redress rights in particular are closely linked in the way that they fulfill this goal. Both provide personal remedies to securing rights against transgression, although they do this in different ways. Defensive rights are forward-looking. They seek to avert an unjust harm before it occurs. Redress rights are backwards-looking. They seek to repair or restore an unjust harm after it has occurred.
Punishment is somewhat different from these two. It is both backwards-looking and forward-looking. It looks back to harm that has already occurred and it looks to the deterrence of future transgression and the reformation of character. Punishment also differs from defense and redress in that it is a societal remedy rather than a private remedy. Its primary goal is the securing of rights and welfare on a societal level rather than an individual level.
Finally, each of these three mechanisms is subject to a requirement of proportionality. The idea of proportionality is going to be very, very important in the account that I develop. What proportionality says is that the harm inflicted in each of these cases—defense, redress, or punishment—must be proportional to the liability of the person who will suffer that harm.
What I have done so far is to describe in an intuitive way what I hope will be a reasonably familiar specification of the system of rights as containing self-remedial mechanisms of liability. But what can explain these mechanisms? In particular, how can it be that one can come to forfeit something as important as the core moral rights to life, liberty, property, and freedom from physical attack? What is it that provides a moral explanation of these mechanisms?
I believe that the most basic explanation of the remedial features of rights lies in the conditional nature of personal rights. Conditionality expresses the idea that the possession and enjoyment of certain rights can depend upon the fulfillment of moral duties and requirements. It is a familiar idea, expressed in the old adage that rights entail responsibilities. An example would be the way in which the rights that parents have over their children are conditional upon fulfilling their duties of care towards their children. Or, as teenagers are constantly reminded, the enjoyment of social rights can be conditional on fulfilling domestic responsibilities.
The rights famously function to protect persons against being used simply as a means to the goals of others. But if the possession and enjoyment of rights is conditional upon the observance of relevant moral requirements, and in particular on respecting the rights of others, then it is natural to think that rights can be forfeited in precisely this way. By transgressing a relevant moral requirement, a person can become liable to be harmed as a means to preventing or remedying that very transgression.
The purest and most direct form of conditionality is found in the relationship of moral reciprocity. I believe that some of our most important personal rights are rooted in an interpersonal relationship of moral reciprocity. A case in point is the right to life. I possess the right that you not kill me because and to the extent that I recognize and respect your right that I not kill you. The shape of the right to life and the conditions for its forfeiture and loss are all contained within the subtle, intimate relationship of moral responsibility.
Now, I actually think that this idea of reciprocity is a very, very profound one. I don't think it's a coincidence that when you look around the world, many, many cultures have as a form of greeting a ritualized form of reciprocal interactions, many of which are physical interactions—we shake hands; soldiers salute; in the South of Europe, one kisses on the cheek; in the country where I come from, New Zealand, Maori people greet each other by pushing together their noses and foreheads. It seems to me that what all of these cultural forms signify is, "I am a reciprocator. I will give back in kind." It seems to me that this idea of reciprocation is a very, very deep and fundamental one in our culture.
What we've done so far is lay out what I've claimed is the basic structure of the system of rights, as containing these remedial mechanisms. I've argued that two of those mechanisms in particular, defense and redress, are rooted in notions of conditionality, and in particular on this very intimate interpersonal relationship of moral reciprocity. What I want to do now is to drive that down a level further and show how this can yield a specific analytical model of the way in which rights alter their configuration and move, in the case of defensive rights.
What I'm trying to do is give a perfectly general account of defensive rights. We consider an interaction where we have a relationship between two agents. We call them A and D. In standard cases, of course, A is an aggressor; D is a defender. But I don't want to call them precisely that because we also want to think about cases in which A may offer harm or inflict harm or inflict the risk of harm, even though he may not be an aggressor in the physical sense. He may not even be posing harm at all. We want to think about these cases as well.
So we have an interaction between two persons, A and D. We also have two instances of harm. Call them threatened harm and defensive harm. What makes something a case of defensive action is that A's action will inflict threatened harm—TH—on D, unless D acts in such a way that will inflict defensive harm on A. What we want to know is when it is the case, in situations of that form, that D will possess a liberty right to inflict defensive harm on A.
What I want to claim is that there are a large number of variables that will contribute to making that assessment. Those are the 15 variables that you find on your handout. Let me just run through a couple of those to make it a little bit more clear how some of these work.
The first variable, of course, one has to consider is, what is the magnitude of the threat? One obviously has the right to inflict greater defensive harm when the aggressive or threatened harm is higher. When the threat is to my life, I'm entitled to inflict lethal harm. I'm not entitled to inflict that harm for a more trivial form of threat. One also needs to think about the probability of that harm actually coming to pass. If one is threatening rights to my property but the probability of that occurring is very low, 10 percent or 5 percent, it may be that I have a lesser liberty to respond than in cases where the probability is 100 percent.
The liberty of defensive rights will also critically be dependent upon whether A has a justification for harming me. Why might A be justified in harming me? It may be that the harm is one that I have no right against. It may be that it's one that is not prohibited by any norm at all. Many externality harms are of this case—the harm of congestion or pollution that somebody inflicts upon me by driving a car around. I simply don't have a right against that. It may be that I have waived or consented away my right against harm in certain ways. In that case, again D does not have the right to defend. It may be that D is liable to that harm. That may be because that harm is justified defensive harm. Obviously, if an attacker is now being threatened by the legitimate defensive harm of his victim, the attacker does not suddenly receive a right of self-defense. So if one is liable to the harm, that does not generate defensive rights or, for example, if the harm is justified as a form of redress or punishment.
Things are less clear when we come to harm inflicted justifiably in the case of a lesser evil. Imagine again our case of the hiker. The person who owns the hut in the mountain is having his rights infringed by the hiker who breaks in and steals the food. But breaking in and stealing the food is, all things considered, justified. Is it permissible for the owner of the hut to defend himself against that harm? Many people think not. I actually think that he has. I think that a liberty of defense is generated simply by the fact that a person will transgress your rights, whether that transgression be a full violation or merely a justified infringement.
Also crucial will be the question of whether A is responsible for the harm that he offers towards you. How might an aggressor fail to be responsible for harming you? It may be that he's not an aggressor at all; he's simply an innocent bystander. He is situated in such a way that I can't save my life from the attack of an aggressor except by killing an innocent bystander. It may well be that the only way I can escape from an attacker is to drive my car across a bridge, running over an innocent person that's standing there. We would not think that the innocent bystander is liable to defensive harm. Why? Because he has no responsibility for the harm.
On the other hand, it may be that there are excuses, that A has an excuse for inflicting harm upon you. Excuses, again, can come in a number of different forms. Many forms of excuse are what I call agency-degrading excuses. They demonstrate that an agent has diminished responsibility for a certain action or harm. They can be of many different kinds—provocation, insanity, infancy, mistake, ignorance. In these cases, most people believe that although harmful action may be fully excused, so that the actor is non-culpable and would not be liable to be punished, it can still generate liability to defensive harm. A clear case here may be a psychotic aggressor, somebody who attacks you who is criminally insane, and therefore not responsible, in this rich sense of culpable. Nonetheless, most people believe that it is permissible, fully permissible, to inflict lethal defensive force upon a psychotic aggressor.
I think that that is the case. Although an excused actor, like a psychotic aggressor or a mistaken aggressor, has diminished responsibility, they still fulfill the basic elements of agency responsibility. They are still acting, although in a very minimal sense, as a human agent.
But there are other forms of excuse where that is not the case. Think, for example, about the case of physical compulsion, where a person seizes your body and moves it against your will. There's a famous—some would say notorious—example that was proposed by Robert Nozick, where he said, imagine you are at the bottom of a well and somebody pushes a fat man down the well. The fat man would crush you to death if he landed on you, but he would survive because you would cushion his fall. You have a ray-gun and you could disintegrate the fat man before he falls on you. Would you be entitled to do that?
Many people argue yes. I argue no. I argue that the falling fat man is no different from the innocent bystander. He has no agential engagement with the threat that he is posing to you at all, and that that agential engagement is crucial. That's why it's permissible to use force against a mistaken or infantile or ignorant aggressor, but not against somebody suffering from physical compulsion.
There will be a question about whether A brings about the harm through action or simply through an omission. There is a greater right to defensive harm against action than there is from, for example, somebody who omits from protecting or saving you from harm.
There are also very important and difficult questions about the proximity between A, the aggressor, and the threatened harm. These run along two dimensions. We can think about the temporal dimension. On classical accounts of self-defense, defensive rights were limited to imminent harm. It had to be immediately prior to the harmful action, and it could not be after the fact. There are also questions about the causal proximity of the agent's act to the harm. A key question here is whether these two conditions are relevant simply because they contribute towards responsibility or whether they have an independent significance. Here's an example to suggest that they may have a significance independently of responsibility.
Imagine that somebody is the boss of an organization responsible for the transplant system in my state, and he embezzles all the money from the transplant organization, criminally, knowing that there is a risk that the organs will not be there when they ought to be there and that some people will be at risk of dying. Let's say now that I am one of those people who are at risk of dying through his criminal negligence. Would I be permitted to kill him and harvest his organs? Most people think not. Yet a criminally negligent person who has embezzled money foreseeing that this will put people's lives at risk is surely as responsible for that harm as a psychotic aggressor or an infantile aggressor or a mistaken aggressor.
So there's an example to suggest at least that proximity considerations may play a role independently of the contribution to responsibility.
There will be questions about the intention of the actor for bringing about the harm. Was the harm brought about as a direct intention? Was it intended either as an end in itself or as a means to some other end, or was it merely the foreseen but unintended consequence of action?
There will also be questions of the duty of care of A towards the victim. This really acts as an exacerbating feature. If the aggressor had duties of care or special duties towards the person he's harming, this may generate particular liabilities. I think this may be what underlies the intuition that many people have that so-called battered wives may be justified in inflicting a greater level of harm than other people would be who were not situated in a situation of particular care.
What I think is really important and really revealing about the reciprocity way of thinking about defensive rights that I'm suggesting is that it forces us to look at both sides of the equation, both at the situation of A, the aggressor, and his relationship to the harm and also the situation of the defender and his relationship, both to the defensive harm and to the threatened harm.
On the other side of the equation, of course, we need to think about the magnitude of the defensive harm. The balancing has to include both the threatened harm and the defensive harm. We also clearly need to think about the probability that the defensive harm will actually succeed in averting the threatened harm. If you can defend yourself against a harm by inflicting harm on A, but the probability of it succeeding is infinitesimally small, that may reduce the right to inflict that harm.
We also need to think crucially about the defender's responsibility for the threatened harm. Think about a case of provocation. If I want to find an opportunity to kill my nemesis, who, let's say, is mentally immature, and I provoke him to the extent that I know that he will, through that provocation, engage in an unjust attack upon me simply so that I can then kill him and claim a justification by reason of self-defense, clearly we would not regard that action as permissible. Why? Because the defender himself is responsible for the threat.
So we have to look at questions of provocation. We have to look at questions of mutual escalation. We also have to look at whether D brings about his defensive action through something that he has done or something that he has allowed to happen. Again, there will be questions of proximity, both causal and temporal. And just as we had to look at the intention of A, the aggressor, in bringing about harm, we have to look equally at the other side, at the intention of D in bringing about the defensive harm. Again, there will be a distinction here between whether the defender directly intended the harm or merely foresaw it as a side effect of other action.
Direct intention is an interesting one. As I said, what's really distinctive about defensive rights is that they are specific permission to directly and intentionally harm another person as a means to bringing about some further end. That's part of what defensive rights do. The greater problem is where the good achieved by the defensive action was not the end in itself. Here's an example where that might be the case.
Let's say that Al, who is a notorious gangster, walks into a bar in order to kill a rival gangster. Just as he's about to take aim at this rival gangster, the rival is in the process of carrying out a hit upon a third person within the bar with a concealed weapon under his table. If Al the gangster had known about the lethal threat that his target was posing to the third party and if he had acted with the intention of thwarting that threat, his acts would have been justifiable defense. But the fact that he didn't know about that, the fact that protecting the life of this third party was simply an unforeseen side effect of his action, that entails that he does not possess the liberty of self-defense.
I think this is important, because many theorists will say it's all about the responsibility of the aggressor, that the liberty of defensive right turns upon the responsibility of the aggressor. No. It turns also upon the responsibility and intention of the defender.
We also need to look about duties of care on the part of the defender. I have gone through that quickly. We can come back to those in Q&A if you like.
So my hypothesis is that the general structure of defensive rights is the one that you find on the back of your handout. A is liable to defensive harm by D in relation to threatened harm, TH, if and only if the threatened harm can't be averted without inflicting the defensive harm—the defensive harm is necessary—and there is this very rich proportionality relationship. It's interesting. Traditionally, the familiar way of thinking about proportionality in defensive cases is that you say the proportionality is between the aggressive harm and the defensive harm. That's basically how proportionality has been understood. No. It's a much more complex and rich relationship than that. It's a relationship between the aggressor's relationship to the threatened harm and the defender's relationship to both the threatened harm and the defensive harm, in a very rich normative way.
So there has to be this appropriate proportionality relationship. And the same is true for any additional aggressors—A-1, A-2, A-3, A-4, A 5, A-6. Now, crucially, as we're going to see in a minute, those are considered individually and not aggregated.
So that's my hypothesis about the way that defensive rights, in their most general form, work, if one approaches them from this perspective of conditionality and reciprocity.
Now, what I think is interesting about this model is that when you think about lesser-evil justifications for harming, it seems also to be true that each of these 15 variables is potentially relevant to a lesser-evil justification for harm, with appropriate modifications for how these variables are thought to operate.
Which of these variables are considered relevant within a lesser-evil justification will, of course, depend upon one's underlying theory of value. For utilitarians, it may be the case that only the magnitude of the threatened harm and the defensive harm is relevant. You just look at the two harms, and whichever of those is the higher is the one that you ought to do. But richer accounts of value, more commensurate with common-sense morality, may also allow that other variables are relevant. For example, it may be considered that inflicting harm on an innocent bystander ought to be considered as a greater evil than inflicting harm on a person responsible for an unjustified threat. I think that is the way that we think about commensurate harms. We ought to, within a lesser-evil account.
Similarly, it may be that one takes into account intention, the distinction between doing and allowing, all of the other variables that I've indicated here.
This is very interesting. This suggests that both the liability justification for defensive rights and the lesser-evil justification of defensive action have this similar underlying structure. They are linked by both being, in essence, complex proportionality relationships, made under differing conditions. You have in your handout there the assessment of how I think the lesser-evil justification can work.
I said they operate, but they operate under different conditions. What are the different conditions? There are a number.
First of all, crucially, within a rights-based liability account of justification, one considers the proportionality relationship, the question of liability, individually rather than aggregately. So we ask, is A liable, is A-2 liable, is A-3 liable, is A-4 liable? We ask that in an individual sense. That's why in the specification it says "and it's true of consecutive other aggressors." That's not the way we do it within lesser-evil justifications. Within lesser-evil justifications, we aggregate the harm over all agents.
This is very important because it's this that explains the intuition that we saw right at the beginning, that an innocent person is justified in killing a greater number of aggressors in order to save his life. Killing five to save one is not disproportionate, because each of the five is individually liable. What's more, it doesn't seem that there is any limit to this. It would seem that a defender has a liberty to kill literally any number of aggressors, provided that each is individually liable to be killed.
The lesser-evil justification of defense of harm, on the other hand, aggregates the harm inflicted at all aggressors. So even though it may consider that harming a culpable aggressor is a lesser evil than harming a non-culpable aggressor, it still may be the case that the aggregate evil is sufficient to make it impermissible.
We often think the consequentialist reasoning is more permissive than rights-based reasoning. But this is an interesting example in which rights turn out to be substantially more permissive, in the sense that they actually enable a much more permissive use of force.
Another key difference will concern which of these variables are necessary—which ones have to be present, which ones have to be non-zero—in order for the justification to go through. Clearly, on a lesser-evil account of what's going on here, the only variable that needs to be non-zero is the harm. It has to be the case that one is preventing some kind of threatened harm. If not, and if the defensive action is harmful, then ipso facto there cannot be a lesser-evil justification.
That's not true of the liability account. On the liability account, as I presented it, what one is justified in defending against is the violation of a right, and a right can be violated even though the right holder is in no way harmed. Think, for example, about a person who instructs his financial agent not to perform a certain trade. The agent performs that trade and realizes a substantial gain for his client. But, still, he has wronged the client. He has done something he oughtn't to do. It seems to me that there would be some—some—permission to use harmful means to prevent the violation of that right.
For the rights-based liability account, the two crucial elements are, clearly, justification and responsibility. One cannot be liable, on the rights-based account to defensive harm, if the threatened harm is justifiable or if there is no responsibility whatsoever for the threatened harm.
Just to recap a little bit, we have looked a little bit about the general structure of a system of rights that I've claimed has this coherent self-remedial nature, and this remedial aspect operates, crucially, through the idea of forfeiture of rights that generates liabilities and correlative liberty rights. We have seen how the lesser-evil justification for inflicting harms reflects, in some ways, a very similar form of relationship between these different moral variables, although interpreted, crucially, under different considerations.
What I would now like to do is reflect upon the question of when it is that we permit a lesser-evil justification for the infringement of rights. It seems to be true that we are, in general, much more ready to allow lesser-evil justifications for intentionally infringing property rights of innocent persons than we are for infringing their rights to life, to bodily integrity. So paradigm cases of lesser-evil justification all involve to property or commensurate values. Think about the classic examples: A farmer burns a neighbor's field in order to stop a wildfire from engulfing a town. A person breaks down the door of a house in order to save a child from noxious fumes. A captain throws his cargo into the sea in order to save his vessel from a ferocious storm. Those are the uncontroversial cases of lesser-evil justifications.
Philosophers, of course, are very good at coming up with fictional cases, in which we're very tempted to countenance killing or torturing or other central attacks on the integrity of the person in order to prevent some massive calamity. But in the real world, we almost never permit right to life and bodily integrity to be infringed in order to avert a greater evil. In the most celebrated legal case to consider the lesser-evil justification for homicide, the famous case of R v. Dudley and Stephens, two shipwrecked sailors were denied a lesser-evil justification for killing and eating the cabin boy, who was also trapped with them on their raft, even though it was reasonable to suppose that they otherwise would have starved if they hadn't have killed and eaten him.
Why should this be the case? Why should it be that we are much more inclined to allow lesser-evil justifications for infringing property rights than we are for infringing rights to life and against serious bodily harm?
What I want to suggest is that the answer to this question has to do with the distinction between different forms of harm, and in particular between harms that can be fully compensated and harms that cannot be. Different harms are amenable to compensation in different ways. The most perfectly compensatable harms are those concerning fungible goods, such as money. Money can literally be perfectly compensated.
At the very other extreme are harms that cannot be compensated for, even in principle. The paradigm of an uncompensatable harm is, of course, death, for the very simple reason that the wronged party no longer exists. Compensation and apology may still be owed to the estate, but the primary bearer of the claimed right to redress cannot, even in principle, receive compensation.
Of course, between perfectly fungible harms and harms that cannot, even in principle, receive compensation is a wide range of partially compensatable harms. Psychic harms, such as a victim's feelings of fear and insecurity after an attack, can be partially compensated through payment of money. But clearly it's a very imperfect form of compensation.
The same is true for physical harms that cause pain or injury. Compensation can be partial, but it's far from perfect.
Some injuries are so severe that they come close to the status of death in their inability to be compensated—blindness, paralysis, loss of a major limb, disfigurement, all injuries of this kind. I think that rape and torture also have a very special status because of the extreme nature of the violations that they inflict.
So we have a distinction between the way that different forms of harm are amenable to compensation or redress. But why should this distinction between compensatable and non-compensatable harms explain why we allow lesser-evil justifications in some cases, but are reluctant to allow them in others?
I think again that this has to do with these basic mechanisms of how rights move and the way in which the system of rights operates as a self-remedial system. When a right is justifiably infringed as a lesser evil, it does not, as we saw, simply disappear. Although it is justifiably overridden as a means to averting a greater evil, the infringed right persists, generating in turn a claim for compensation or redress. Infringing a right, even with the intention of fully compensating the right bearer, still requires significant justification—hence, the operator within the model of lesser evil, the "significantly greater than" operator. The reason for this is that the infringement of the right, even with compensation after the fact, imposes significant costs and risks on the right holder, costs and risks he has a right not to be forced to bear.
Nonetheless, in the case of fungible goods where the possibility of compensation exists, there is a way to comprehend and manage the justified infringement of a right within the system of rights itself. The infringed right does not simply disappear. It is transmuted into a right to compensation. But in the case of non-compensatable harms—death, grievous bodily injury, torture, and rape—this accommodation is not in any way possible. Here the inflicted harm departs the system of rights altogether.
The right has not simply been infringed, overridden to reappear in another context; it has disappeared altogether. It is not surprising, therefore, that if we take the idea of rights seriously, we will never or almost never allow the transgression of rights against such harm, even to avert significantly greater harms.
I would like to allow time for discussion, so I think I'll stop there. Thank you very much.
Questions and Answers
QUESTION: Can you comment on the condemnation of private property for another private person, ostensibly for the public good? For example, condemning property for a basketball stadium and other personal property developments for personal gain.
DAVID RODIN: That seems to me to be a classic case of the infringement of a right for some greater public good or utility. It seems to me that what we need to say about that is that we need to look at the competing goods on both sides. We can consider all of the variables that are laid out in the model. If we believe that there genuinely is an overriding public value or good or utility, then we may say that that infringement of rights is justified. But it will, of course, require payment of compensation, of redress. That's the distinctive nature of the system of rights, that we sometimes allow them to give way to public goods, aggregate welfare, but we require that those rights be recognized through the payment of compensation.
But the assessment will depend clearly on the specifics of the case, whether that claim can be made good. But it will fit within that model.
QUESTION: I thought the paper was really interesting. I have one basic question of clarification, but, before that, kind of irritation.
You began by saying that the core of morality is negative rights. You motivated that by saying that compliance with these ensures you meet a minimum standard of decency. Falling below them, you become a moral monster.
DAVID RODIN: I don't know that it ensures that you meet a standard. But if you don't comply with those rights, you haven't met the standard. It's a necessary, if not sufficient, condition.
QUESTIONER: Good, because it just seems to me that there are obvious cases where you're in a position—your hiker discovers a baby on the verge of starvation—
DAVID RODIN: I don't want to deny the existence of very, very, and equally strong positive rights. I don't want to deny that at all. But my focus was really just on the negative rights.
QUESTIONER: The question is really about this thesis about the conditional nature of rights. There are two ways of reading the "conditional," a wide scope and a narrow scope. One thought might be, conditional on having the right in the first place, or is it a right whose content is conditional? It seems to me you were talking about the first, because this whole notion of forfeiture means that you won't have the right anymore. What's problematic about that is that we tend to think of human rights, at least, as rights we have, not on the basis of any of the conditions you talked about, but simply by virtue of being human, and that they are inalienable.
I just wondered what you thought about two cases. One of the conditions you talked about was reciprocity. I wonder what that meant for you, about the rights of children, the rights of the mentally disabled, the rights of people with Alzheimer's disease. They're not in a position to reciprocate, because they can't grasp that you've got rights, nor a position to threaten your rights.
Does it follow that whatever morality applies to our treatment of them is not rights-based?
The other case is punishment. Someone is subject to a perfectly just punishment. What are we saying in that case? That their right to liberty is forfeited, no longer exists? Or their right to life; let's say the punishment is the death penalty—then one of their fellow prisoners could kill them without violating their right, because they have forfeited their right. The alternative is to say they still have a right to liberty, but their right to liberty was subject to the condition that if they went around killing people, they would have to bear a just punishment. That just punishment is an exception to their right. So their right already has in its content the fact that you can't get liberty if that's incompatible with retributive justice being served.
DAVID RODIN: Good, good. I don't actually think that there is a deep distinction between those two ways of thinking about conditionality. It seems to me you can catch out these ideas in either term.
In the case of forfeiture, if one forfeited a right, doesn't that mean that anybody is at liberty to treat you in a way that contravenes the interest that had been protected by that right, for any reason whatsoever? Well, no, it doesn't mean that. One of the ideas or intuitions that I mean to get at by describing certain of these negative rights as being based in this very intimate relation of conditional reciprocity is the fact that forfeiture is a very, very specific and limited fact about a person. I forfeit a right to a particular person for a specific purpose. That's also entailed by the idea of liability as being instrumental, having this instrumental nature.
So the fact that I have forfeited my right to life to you for the purposes of defense entails nothing about whether anybody else has a liberty to kill me or attack me in any way at all. We saw that in the Al-and-the-gangster case. The mere fact that my enemy was about to kill another person unbeknownst to me doesn't mean that anybody is at liberty to kill that person at any time and for any reason. What it means is that somebody is at liberty to kill him for the purposes of defense of the innocent person's life.
This, I think, is a real virtue of thinking about these rights as having a very, very interpersonal basis.
Then you raise the question of what this means about infants or the incompetent. I think that this reciprocity does apply to them as well. Think again about defending oneself against the psychotic aggressor. Here is a person who may not have full moral agency. He may, perhaps, through his mental illness, not be fully capable of reciprocating, respecting the full panoply of rights. Does that mean that he can't become liable to defensive force? No, I don't think that it can, provided there is at least some minimal agency there, so that one can understand the response as being the response towards an agent rather than a response towards a human being who is being treated as a mere object, as you find in the Nozick case of the falling fat man. As long as there is some element of basic minimal agency, then I think it's possible to talk about rights and to talk about forfeiture of rights.
I'm sorry, you had one further question.
QUESTIONER: It was the Nozick example and the distinction you were making of that with the person who is, let's say, completely mentally incompetent. I'm not sure that I really see how, where somebody is completely mentally incompetent so that they have no control or rational decision on their action, you can make a distinction and say that in that case there is agency, but in the case of the fat man being thrown down the well, agency doesn't exist.
I'm just wondering if you could explain the basis for your saying agency exists in the one case and not in the other.
DAVID RODIN: That's a very important question. I was talking rather glibly about insanity as an excuse. Of course, insanity covers a very, very wide range of different conditions. I think to answer that question properly, you need to look very, very carefully at the way that mental illness disturbs responsibility and disturbs agency. It seems to me that different cases will go different ways.
If you think at two extremes, in one type of case there is a physical disturbance that occurs within the brain of a person. Diabetes might be one example, where the operation of the brain is suddenly disturbed by some kind of extraneous chemical imbalance. In that kind of a case, it seems to me that you are much, much closer to the description of the falling fat man. This kind of affliction has just come upon you as something entirely external to your agency. It's very, very difficult to attribute the actions to anything that we could recognize as some kind of human agency.
On the other extreme of the spectrum is the case of a psychopath, somebody who, over a long period of mental illness, has simply developed the character of an entirely ruthless and amoral killer. Now, of course, the fact that he has developed that character is not something for which he is responsible. It may have all kinds of underlying causes. But the fact is that the harmful action, the threat that he poses to others, almost paradigmatically derives from him, a moral agent. It's true that his agency has been corrupted, but it seems to me that it's still possible in those cases to regard the harm as being linked to his agency in a way that makes it meaningful to talk about sufficient responsibility for forfeiting a right.
I think it's a very, very subtle thing. I think what you really need to do is go through case by case the different ways in which mental illness degrades and afflicts agency.
QUESTION: I read the title, "How Rights Move," and I came with the construct in my mind that you were going to be dealing with the degree in which a society is open or closed—for example, what happened with apartheid in South Africa, how in the Middle East a woman who is raped can be perceived as the aggressor and not the defender, and punished for that. That is the construct I had.
Can you state the assumptions you're using in your framework?
DAVID RODIN: We are actually beginning tomorrow a program of conference presentations on themes around more political and social rights, including rights to democracy. Those are obviously extremely important questions, but they are different questions than the ones that I was trying to think through here today.
The assumptions or the way that I'm approaching this: Rights are often thought of as having a paradigmatic political nature, that they are claims against states, in the first instance, and other institutions. It's clearly true that many, many rights have that as their primary function. But that's not the only kind of rights that we have. There are also very, very central and important rights that are not social, not political, not even legal. They are aspects of the fundamental moral relationship between individual agents.
It seems to me that one of the things that's very, very important about them is that they have this very, very determinate structure that enables us to draw normative inferences about them and draw normative conclusions from them. They tell us things about how we may or may not treat other human beings. They tell us that we cannot harm another human being, even to prevent harm to others, if that person does not have responsibility for it or if that harm was not, itself, justified. That does have very, very significant consequences.
To give you one example, one of the areas that I do a lot of work on is the ethics of war and conflict. One of the very significant developments that I and a number of other scholars have been teasing out over the last number of years is the question of why we should consider it to be the case that a soldier acts permissibly when he fights in a war that is not itself justified. Most people think that a soldier doesn't do wrong simply by fighting in a war, even if that war is an unjust one. Many people think that the war in Iraq was unjust, but they don't think that the soldiers who fought there did something wrong by doing that.
But when you think about it from the normative structure of interpersonal rights that I've been suggesting today, it's immediately obvious that there's something profoundly problematic and difficult about that idea. If a soldier is fighting in a war that is not justified, he's using lethal violence against other persons—let's say persons that are simply defending their state from an act of aggression. Those defending soldiers are presumably not doing anything wrong. They're offering harm to you, but they're offering harm that is justified, in precisely the same way that a victim engaging in self-defense is using justified harm. So how can it be the case that a soldier using force in this case is using force that is permissible?
I think this raises a really, really profound question and problem for the way that we think about the rights of soldiers, the way that we configure our militaries, the way that we think about the opportunities for conscientious objection, for example. So it's a different question, but a nevertheless important one.
QUESTION: I would like you to talk a little bit more about this proportionality relationship. At least as you have it on the handout, it seems to me highly implausible, and it seems to me that something like a threshold relationship might obtain. But let me give you just one example.
Imagine, after leaving this lecture and going back to my hotel to pick up some luggage, I'm kidnapped, and my kidnapper says, "I'm not going to keep you for very long. I just want to have a little bit of fun, so I'm going to play a game of Russian roulette. I have a gun here with one bullet in it." Unbeknownst to him, I have a gun as well, but my gun has six bullets. So I could certainly kill him. Do you think I should remove five bullets so that there's some proportionality?
DAVID RODIN: The short answer is, no, you certainly shouldn't. But it's a very, very interesting question.
I should just clarify one thing. To say that I conceive of this as a proportionality relationship is not to say that I think that each of these variables should be perfectly trade-off-able against each other. When I first started thinking about these kinds of issues, I thought, "This is great. If we've got this model and it's the case that when you increase the value of one variable on this side by 50 percent, you can decrease another variable on the same side by 50 or increase the value of a variable on the other side by 50 percent, and it will all match up."
I call that the continuous operator model. If that were the case, it would be terrific. You would have this incredibly elegant model where it was perfectly continuous in the operations.
Now, it seems to me that it does work roughly like that, but only for certain kinds of threats. Again, interestingly, there is also here a distinction here between threats to property and other fungible goods and threats to life or grievous bodily harm.
If instead of your example we're thinking about a person who doesn't have a revolver with one bullet in a chamber, if we think about a petty thief and he is about to snatch your wallet, but—we would have to think about some kind of mechanism here—he had only a one-in-six chance of actually getting your wallet. You could protect your wallet through a range of different defensive actions, one of which would impose a moderate but reasonably painful harm on the pickpocket and would be certain to succeed, the other of which would inflict a much lesser form of harm and would, say, have a one-in-six chance of succeeding. There it doesn't seem to me unreasonable to think about these as actually operating in a continuous way.
But when you think about threats to life or grievous bodily harm—you are absolutely right—it would be absurd to think about it in that way. In cases of threat to life and grievous bodily harm, we don't think about these as continuous operators. We think about them as threshold operators.
Now, why might that be? I think, again, it's part of this extraordinary way in which the system of rights works. If you're thinking about threats to fungible goods—your property paradigmatically, your wallet—if you abstain from using the harmful but guaranteed-to-succeed defensive mechanisms and you give up on opportunities of defense and allow your property to be unlawfully taken, through the normative structure of rights, you will obtain some sort of a right of redress. A right to compensation is not as good as having your property, because it's risky and all of those things. But it doesn't seem unreasonable to ask you to surrender certain defensive remedies. The reason for that is because your interest in your good hasn't simply disappeared. What you get in lieu of that is some kind of claim right to redress or compensation.
Of course, as I claim, in the case of threats to your life, that's not the case. If you are killed, you cannot, even in principle, be compensated. That's why I think the picture is not as nice as this elegant continuous operator model. I think it does operate more or less continuously for fungible goods. I think it operates much more on a threshold model for threats to life and grievous harm.
QUESTION: In your response to an earlier question, I thought you were suggesting two different roles for minimal agency. The first, which was the one that was illustrated in your diagram, was as a criterion for liability within the structure of rights, so the psychotic aggressor was liable, but the man falling down a well wasn't. But in response to the question about mental competence, I thought you were saying that minimal agency also played a role as a criterion for membership in a system of rights.
I'm not criticizing that. I was just wondering if you saw some relationship between those two roles.
DAVID RODIN: Let's just take the last question there, and then I'll pick it up.
QUESTION: Just very quickly, I gather from what you said that you don't find lesser-evil justifications of torture persuasive. I would assume by extension that maybe you don't find such lesser-evil justifications of preemptive war persuasive either. But I would like to hear what you have to say about both of those—the ticking-time-bomb example and the smoking-gun-as-mushroom-cloud example—with respect to the liability framework.
DAVID RODIN: Sorry, mushroom cloud is?
DAVID RODIN: On mental incompetence and position of the status of rights, I think there's a very deep and difficult question about what it is that grounds the possession of rights. All of the classic theories of rights have difficulties with the cases of mental incompetence.
You're right that there is a tension in the way that I was thinking about that. It may well be that what I want to say is that the minimal agency is part of the criteria for the forfeiture of rights, but that it's not necessarily the basis for the possession of the rights. It may be that there are some independent grounds for possession, but that the liberty of forfeiture actually turns on this. But I think that's something that probably requires some further thought and analysis.
To the two questions on torture and on preventive war, I think you would be right on my assessments of both of these counts.
I'm not, in general, in favor of preventive war. The reason for it is exactly for these liability reasons. In this liability reciprocity view of rights, one only loses rights against harm by actually engaging in some kind of significant unjustified harm or threat of harm. I think it's very, very difficult to see how you can persuasively make that case out, in the context of action that is highly preventive.
But, of course, on the model that I'm suggesting, the possibility can't be ruled out, if the harm is really great enough, provided that there is some minimum agency on the part of the persons that you are actually harming. I think there are real questions there. If the people that you're harming are the soldiers, those are the people that are not engaged in the plans or the conspiracy to generate these atomic weapons. Those are ordinary, regular soldiers going about their business, training. What have they done? What's their involvement?
Again, on the lesser-evil justification, that can be commensurate, because responsibility is not a criterion. It's simply an additive feature. But on the liability approach, it's very difficult to see how they could be liable.
Your second question was on torture. It just happens that torture was the topic of my lecture last year. It was called "Explaining the Absolute Prohibition on Torture."
So I do think that torture is something that is prohibited by an absolute, exception-less prohibition.
But the real question for someone like me, who has this liability account of rights, is why I should believe that. If it's the case that rights are based on reciprocity, there are some really, really evil people out there who have done things as bad as torture or even worse. Why shouldn't it be the case that they should on occasion, perhaps, lose their right against being tortured? The answer to that is that that is indeed possible. I do think it is possible to lose the right against being tortured, by which I mean that by torturing that person, you would not be wronging them. But that's not the same thing as saying that it would be permissible to do it.
It's sometimes impermissible to do something to another person, even though that person may not have any right that that thing be done.
Why might that be the case? I think that that's the case sometimes because we have moral obligations, not simply arising out of the duties that other people possess, but out of our own conception of the kinds of persons that we want to be and, crucially, the kind of society that we want to live in. This actually comes back to the question raised by the lady over here. It seems to me that the totemic idea of possessing and living within a society that rules out torture in all contexts is something of extraordinary importance to us. Because of that, it's possible to understand morally how we ought sometimes to bear the risks of that, even where the risks of that might be being harmed by a ticking-time-bomb situation.
So the short answer is, I do think the prohibition is absolute. But the grounds of that prohibition is not necessarily the fact that the right is absolute.
JOEL ROSENTHAL: David, thank you very much.