Uruguayan Air Force Flight 571, also known as the Andes flight disaster, and in South America as Miracle in the Andes (El Milagro de los Andes) was a chartered flight carrying 45 people, including a rugby team and their friends and family and associates that crashed in the Andes on October 13, 1972. More than a quarter of the passengers died in the crash, and several more quickly succumbed to cold and injury. Of the twenty-nine who were alive a few days after the accident, another eight were killed by an avalanche that swept over their shelter in the wreckage. The last of the 16 survivors were rescued on December 23, 1972.
The survivors had little food and no source of heat in the harsh conditions, at over 3,600 metres (11,800 ft) altitude. Faced with starvation and radio news reports that the search for them had been abandoned, the survivors fed on the dead passengers who had been preserved in the snow. (Wikipedia)
In my survey, as is shown above, I asked the Chinese, the Russians and the Americans about a rather drastic example, where people in a planned and regulated fashion kill and eat each other, in order to survive. My idea was to question them about something that could in principle have happened, but only in extraordinary circumstances, where we need not bother with side effects of any established practice. I was inspired by the real but less radical event, also reported above from Wikipedia, the air crash in the Andes in 1972. However, reality is not as drastic (at least not if we judge from the reports about it) as I wanted it to be, and hence I had to revise it. It is crucial that people in the kind of survival lottery I want to discuss kill one another in order to survive. However, a problem with my example, inspired by the real event in the Andes, is that it involves cannibalism. I realise now, in particular when I see how people have reacted to my example, that much of their reluctance to accept it may have something to do with a dislike, not for a survival lottery as such, but for cannibalism. There seems to exist a widespread taboo against cannibalism that may have distorted the reactions to my example. Judging from the reactions, this taboo is much stronger in China than in the USA.
1.1 JOHN HARRIS’ ORGAN LOTTERY
The idea of a survival lottery is not new, of course. In particular, John Harris is well-known for his suggestion that one may have a lottery as the basis of the distribution of organs for transplant purposes.136 The idea is that those who draw the winning ticket in the lottery are killed and their organs distributed among people in vital need of organs. Since one donor of organs can serve many recipients, such a lottery, if agreed upon, could be to the advantage of most people. In particular, the expected utility for those who enter the lottery may be higher than the expected utility for those who don’t, at least if we judge the matter exclusively from the point of view of the survival value of the lottery. It might be a problem that those who need organs will be older than the average donor. But this problem could easily be solved, Harris notes. One could for example arrange the lottery in a manner that sees to it that the age of recipients and donors match. I will discuss this kind of lottery as well. It is superior to the cannibalistic lottery in that it avoids all sorts of food taboos, but it has another drawback as compared to the cannibalism example; it presupposes a new and never heard of social institution. The lottery is taken to be, not exceptional, but as part and parcel of how we live together in a society. Here the best move is perhaps to try to abstract from the problem we have to imagine us arranging such a lottery, as Harris does. We are invited to say what we think of such a system, assuming that it is in place somewhere else in space and has been established on a voluntary basis by those who take part in it. This is also how Harris describes this possibility:
Suppose that inter-planetary travel revealed a world of people like ourselves, but who organized their society according to this scheme ... In such a world a man who attempted to escape when his number was up or who resisted on the grounds that no one had a right to take his life, might well be regarded as a murderer. We might or might not prefer to live in such a world, but the morality of its inhabitants would surely be one that we could respect.137 Under such circumstances, what is wrong with the lottery, if anything? Is it worthy of our ‘respect’? It may be difficult to imagine this, however. We are likely to think that, while we could adjust to the fact that one of our children dies because of heart failure (and a lack of a suitable heart for transplant purposes), we could never adjust to the fact that one of our children draws the winning ticket in the survival lottery and is killed. In order to assess the lottery in a fair manner, we must accept the assumption that this is the way we behave, however. Perhaps it is easier to do this if we think of Harris’ extraterrestrial case. Note also that the lottery may be voluntary. Some may have chosen to stay out of it. They receive no organ if one of their own fails, of course, but they do not risk having to sacrifice their lives to save the lives of others. Under this assumption, what would be wrong with the lottery, we should query.
1.2 JOHN TAUREK’S RESCUE LOTTERY
Let me also note that in the philosophy literature there exists another and related survival lottery, the one described and (weakly) recommended by John Taurek.138 This lottery is different. It concerns the saving of lives, rather than killing. And yet, it deserves a place in a book on killing since, even if it is framed in terms of how many lives we save, it could equally be framed in terms of how many lives we sacrifice. Moreover, how the theories under scrutiny in this book treat it, will through some further light on them. Here the idea is, not that we arrange a lottery where those who draw the winning ticket are killed. On the contrary, here we arrange a lottery and those who draw the winning ticket are saved (and those who don’t are allowed to die). The idea is that for practical reasons we can only save one of two competing groups of people, either a small group (one person in the limiting case) or a larger group (typically something like three to five persons). In Taurek’s favoured example six people need a certain medicine to survive. Five of them could share it between them, and yet survive, since they need only one fifth of the available amount. The sixth person can only survive if he gets all of it. According to Taurek, if all things are equal, and no special obligations exist, the right thing to do would be to construct a survival lottery of a very simple nature: we should flip a coin, and give the medication to the five on one outcome (heads, say), and to the single individual on the other outcome (tails, say), giving an equal chance to survive to each individual. Is he right about this? Does he get any support for this idea from any of the theories under scrutiny?
Let us here focus on a somewhat more detailed and elaborated description of a survival lottery of the kind about which I asked the Chinese, the Russians, and the Americans. A broadcasting company is casting a reality TV show. The intention is to transport sixteen people, together with the host, of the show to a small island in the Pacific Ocean. The members of the expedition will gather for successive tribal councils and cast their votes. One by one the members of the expedition will be voted off the island. Whoever is eventually left at the end of the show with the host is the ‘Survivor’ and wins a fortune. On their way to the island, however, the plane they are travelling in develops a technical problem. After an emergency landing on the water in the middle of nowhere the aeroplane soon sinks, taking the crew with it into the deep water. The members of the expedition, together with the host of the show, however, succeed in swimming to a nearby island. Here they find themselves on wasteland. They possess one sharp knife and a functioning lighter, they find a well which provides them with fresh water, they can make up a fire, using driftwood which they collect on the shore, but there is nothing for them to eat: no fish, no game, no roots or vegetables. They wait for help but none arrives. After two weeks they realise that they will all probably starve to death. They gather for their first tribal council and agree to arrange a survival lottery. The ‘winner’ of the lottery will be held down by the rest, killed with the sharp knife, roasted over the fire and then consumed. One person a week will be killed in accordance with the rules of the lottery. Two members of the expedition declare that they are Kantians. They are not willing to take part in the lottery, and this is accepted by the rest. After several weeks five persons have been killed and eaten (among them the host of the show, who was the third to draw the killing ticket in the lottery). The two Kantians have meanwhile died from starvation and have been buried them according to their wishes. Then a ship arrives and the nine survivors are rescued at last.139
2.1 THE AIR-CRASH EXAMPLE
I will discuss this version of the example in this and the following sections. It might seem that it should be more palatable to adherents of deontology than the short version. After all, those who do not want to take part in it are allowed to stay out of it. However, from the point of view of the doctrine, this makes no difference. It is wrong to arrange the lottery, it is wrong to take part in it. Those who do are no better than murderers. It is clear that intentionally they kill innocent human beings. The fact that this killing is voluntary and life-saving doesn’t make it right. As a matter of fact, if they stick to the Kantian, retributivist version of the doctrine, discussed in the chapter above on capital punishment, those who survive deserve to die for what they have done.
This is indeed what Kant should say about cases such as this one, given his retributivist view on capital punishment. He may have hesitated, however, when it comes to the actual execution of the death penalty in cases such as these. In a comment on a similar case, where one person kills another one on a life raft, in order to save his own life, Kant defends ‘law professors’ and claim that they are:
quite consistent in making legal allowance for such emergency acts. For the authorities can’t attach any punishment to this injunction, because that punishment would have to be death, and it would be an absurd law that threatened death to one who refuses to die voluntarily in a dangerous situation.140 Yet, for all this lenience shown by Kant, it is not doubt about the fact that the survivors did the wrong thing. The right thing would have been to accept to die in the situation.
2.2 JOHN HARRIS’ ORGAN LOTTERY
John Harris survival lottery is no better from the point of view of deontology. Here innocent (rational) people are intentionally killed. Or, should we rely here on the Sanctity-of-Life Doctrine and the principle of double effect and say that there is really no intent here to kill them? When we take the organs from a person we ‘ ... merely wish to use a couple of his organs, and if he cannot live without them ... tant pis!’ — as Harris puts it.141 This seems fair enough on a very liberal reading of the principle of double effect, but since it is obvious in the other examples that in those examples there is an intent to kill the victim (before we eat him) and since it is also very plausible to assume that, as a matter of fact, the intent is there even in Harris’ example, I leave this possibility to the side in my discussion.
Let us suppose that people have deliberately and freely entered the lottery, in order to enhance their survival chances. Suppose they willingly give up their lives, when they have drawn their ticket. Those who stay out of the lottery receive no organs when they suffer from organ failure. They are allowed to die. Is there then any moral problem with what they do? From the point of view of deontology, in both its Kantian and its Thomistic version (the Sanctity-of-Life Doctrine) there is certainly still a problem. Those who are killed are indeed killed intentionally and they are indeed innocent, so this is wrong. It would be of no avail if they were to kill themselves. Suicide is no better than murder. It is not right to kill an innocent human being even to save many lives.
Harris’ example is also different from the air-crash example in that when people arrange with the lottery, they are not in a situation of any emergency. They do it just to enhance slightly their expected survival chances. Most people who take part in the lottery will neither need any organs themselves or have to contribute and organs to others. This means, I suppose, that in this case Kant would not have hesitated. Those who killed the ‘winners’ in the lottery are not only murderers; they deserve the death penalty for what they have done.
2.3 TAUREK’S EXAMPLE
Let us finally consider Taurek’s example. The example concerns the saving of lives, not killing. This means that, irrespective of whether I save the many or the few, or flip a coin, I will not violate any perfect duties specified by the doctrine. I do not kill intentionally, I do not treat any one as mere means to the rescue of another, and so forth.
However, we can also see the lottery as deciding who I should allow to die. And now the principle of double effect kicks in. If I save the sole person I can claim that the death of the other five is merely a sad foreseen side-effect of what I do. So I violate not obligations by saving the sole person. But this is not the end of the matter. We must also query whether the requirement of proportionality is satisfied. The criterion is a bit vague, of course, but it strikes me as obvious that it is not satisfied if I save one life and foresee the loss of five. So, what should I do, according to the Sanctity-of-Life Doctrine is to save the many. From a Kantian point of view we reach the same conclusion, it seems to me. The perfect obligation not to kill is not violated regardless of what I do. However, I also has an imperfect obligation to make the world a better place, to the extent that I can do so without violating any perfect duty. But then I should save the many rather than the sole person.
Or, is that assessment rash? After all, according to deontology, every human (rational) being has the same value; it is because of this inherent value or dignity or sanctity that no innocent human (rational) being may be intentionally killed. It also reasonably implies that every innocent human being should be treated with equal respect. Does that imply that, in Taurek’s example, we should flip a coin?
Only if there is no other and better way of showing everyone equal respect. As we will see in the next section, such a better way exists.
3. THE MORAL RIGHTS THEORY
In the more refined description of air-crash example, where the two Kantians are allowed to stay out of the lottery, where it is clear, then, that everyone who enters it does so deliberately and freely, there is no moral problem with the lottery, according to the moral rights theory. We own ourselves and we are free to enter lotteries such as these if we wish. Moreover, it seems rational to do so. The same is true of Harris’ lottery, of course, should it ever be adopted. The problem with Harris’ lottery, however, is that it is hard to see that ordinary people would enter it. This has to do with the fact that we would find it terrible to be killed in the manner described by Harris. In particular, it would be terrible to know that one of ones near and dear would be killed like that, even if her or his death would save several lives. But this is not an objection to Harris case. If we treat it as an extraterrestrial phenomenon or mere thought example we can conclude that those who enter it do nothing wrong. They are perfectly rational, moreover, when they enhance their survival chances by entering (given that they are capable of looking upon the prospect of being killed as no worse than the similar prospect of dying from an organ failure).
On the moral rights theory we are not allowed to force people into a lottery such as this one. They have to enter it on a voluntary basis; otherwise it violates their rights. But what if they are not capable of giving assent? How should we think of a lottery where embryos, rather than grown up people are involved? Julian Savulescu has famously argued that it would be permissible to enter an embryo into such a lottery, even if it would be a person. His argument is quite general. Here I will consider it in relation to the moral rights theory exclusively, where it seems to work.
Most advocates of the moral rights theory argue that embryos are not persons. Hence embryos have no rights. But at least one advocate of the theory has thought differently about this. Anna-Karin Andersson has argued from potentiality that embryos are persons.142 And for the sake of the argument, Savulescu assumes that embryos are persons.143 Does this mean that embryos can be put in a survival lottery? Yes, it seems so. And it seems so even if we assess the matter from the point of the moral rights theory.
Embryos are not capable of giving consent, of course, but it is reasonable to assume that, the person who has created them is for this very reason responsible for them.144 So this person should act on their behalf and in their best interests. And to act in the best interest of an embryo may be to enter it into a survival lottery, where some embryos are sacrificed and used for therapeutic purposes, while others are allowed to develop into grown up individuals, with an enhanced chance to survive (because of the cures that exist as a result of the use of the sacrificed embryos).
You could think of this, as Savulescu does, as a choice between sending an embryo to a planet where few, because of natural disasters, survive, and sending it to a planet where some are used for medical purposes and where the survival chances for each embryo are therefore much higher. A responsible guardian of an embryo should send it to the planet where the survival lottery exists. And those who administer the lottery should not hesitate to sacrifice the embryos that draw the ‘winning’ ticket and use them for therapeutic purposes. Consent has indeed been given on behalf of these embryos.
3.1 TAUREK’S EXAMPLE
What is a moral rights theorist to say about Taurek’s example? As a matter of fact, there is much in Taurek’s own presentation of it that presupposes that he himself adopts a moral rights theory.
His discussion concerns six individuals who need a certain medication to survive. As we remember, one of them needs all the medicine to survive, the other five can survive on one fifth each of the available stock. This means that if the medication is given to those five, they will survive, and the one who needs all the medication will die. If instead the medication is given to the one who needs all of it, he will survive, but the rest will die.
Taurek claims first of all that if the person who needs all the medication owns it, he has a right to use it. He need not share it with the five. This is what follows from the moral rights theory, of course.
Moreover, Taurek claims, if I own the medication, I may give it either to the five or to the single individual. If I happen to know the single individual, who needs all the medication, I am free to give it to him. Again, this is what the moral rights theory seems to imply in those circumstances.
Need I give it to any one at all, if I own it? I suppose not, on a strict understanding of the moral rights theory. If I own it, I may destroy it and let all six die. Taurek does not address this possibility.
What if I have not contracted to give the medication to any one in particular, what if, furthermore, I have no special relation to any one of them, but I still want to save lives. Whose life should I save? Should I save the single person who needs all the medication or should I save the five who needs each only one fifth of it?
So what do I think one should do in ... a situation in the absence of any special concern for any of the parties involved?
First, let me suggest what I would do in many such cases. Here are six human beings. I can empathize with each of them. I would not like to see any of them die. But I cannot save everyone. Why not give each person an equal chance to survive? Perhaps I could flip a coin. Heads, I give my drug to these five. Tails, I give it to this one. In this way I give each of the six persons a fifty-fifty chance of surviving. Where such an option is open to me it would seem to best express my equal concern and respect for each person. Who among them could complain that I have done wrong? And on what grounds?145
This is a strange claim. It has no rationale in the moral rights theory. According to the moral rights theory I can do as I see fit. But I suppose Taurek may here think of some kind of addition to the hard (negative) core of the theory. He seems to acknowledge at least a weak positive right to life. Each one among the six people facing me holds this right. How should I, given that I am prepared to do anything about their plight in the first place, best satisfy their demands on me? Taurek’s answer is: flip a coin.
This answer strikes me as the right one in a choice between two people (all other things being equal). Perhaps some if this credibility spreads over to cases where the choice is between one person and only a few other persons. But Taurek’s point is more radical:
Yet I can imagine it will still be said, despite everything, “But surely the numbers must count for something.” I can hear the incredulous tones: “Would you flip a coin were it a question of saving fifty persons or saving one?” Surely in situations where the numbers are this disproportionate you must admit that one ought to save the many rather than the few or the one.”
I would flip a coin even in such a case, special considerations apart.146
Since this is just a description of Taurek’s own disposition to act, it may very well be true. However, if understood as a recommendation, it doesn’t strike me as at all plausible. And it is indeed less plausible, the greater the difference is in numbers. Moreover, if we take into account the probability that our rescuing mission will succeed, even more problematic implications follow, as Katharina Rasmussen-Berndt has shown.147 Suppose I am on a mission where I can save lives. I can either save five people (and here success is certain), or try to save a solitary person. I cannot do both. Suppose the chance that I will succeed, if I attempt to save the solitary person, is 1 %. Now, if I want to give equal chances of survival to each individual, I am not allowed to flip a coin any more. This would give a 50 % chance of survival to the five, but only a 0.5 % chance of survival for the solitary person. So instead I must arrange with a lottery with 101 tickets; if either of the tickets 1 ... 100 is randomly drawn, then I go for the solitary person, and only if the ticket 101 is randomly drawn do I go for the five. This means that all have an equal chance to survive, slightly less then 1 % (1/101). It also means that, almost certainly, I will embark on a mission, which almost certainly will fail.
If we find this conclusion ‘awkward’ (which Berndt-Rasmussen does), then we should instead distance ourselves from the actual case at hand, and adopt as a general method of decision-making to save the many rather than the few, when the chances that we succeed are the same, and to save the largest expected number of lives, when we have access to probabilities that we can take into account. If we all abide by this method of decision-making, then this means that we all stand a better chance on the whole of being saved, if we happen to need to be saved. For statistical reasons, it is more likely that we end up among the many than among the few!
As a matter of fact, Taurek concedes this, and concedes that, if people for prudential reasons agree about a practice where the many are saved rather than the few, then they have a right to abide by it.
This is pure moral rights talk, of course. But it means that we may abide by any principle we may come up with, if only we agree to do so.
However, the case here discussed by Taurek seems to be different. He seems to discuss a positive right held by the people I can save — even if a very weak one. This weak positive right exists irrespective of any agreement. But then it is not a good idea to give an equal chance to everyone in the situation to survive. A better idea is that, quite generally, everyone should be given an equal maximum chance of getting saved. And this could be thought of as a completely general rule.
According to this rule we ought to save the many rather than the few. And I suppose that, if we have an obligation to save lives at all, that is not grounded in the core of the moral rights theory (we have agreed about it, contracted to do so, and so forth), but which comes as an addition to the theory in the form of an idea of a positive right to life held by everyone, it would best be spelled out as a positive right to an equal maximum chance of being saved. And this means that we should be saved if we are among the many rather than among the few.
The idea that we have a positive right to an equal maximal chance of being saved has credibility in its own right. It makes sense, moreover, to abide by it as a utilitarian rule of thumb. Were we to think like rule-utilitarians (which we had better not do, if my argument in chapter 2 is correct), we could also say that it has a rule-utilitarian rationale.
Does this rule mean that we ought to arrange survival lotteries of the kind where some are killed in order to save the lives of others? This does not follow, of course. The principle can be added to either deontology or a strict libertarian moral rights theory. On both theories we are forbidden to kill innocent people unless they agree about this. On the former we are forbidden to do so even if they allow it.
Utilitarianism has no problem with the survival lotteries here described, either in the air-crash version, or in Harris’ version, where people deliberately and freely enter the lotteries in order to survive (the air-crash example) or to enhance their survival chances (Harris’ lottery). However, utilitarianism is very liberal in its verdict on the air-crash and organ lottery examples. This may be seen as a weakness or as strength. In the modified version of the air-crash example there are two Kantians who stay outside the lottery. They are allowed to do so and when they die they are buried, according to their wishes. This is good news if it is assessed from the moral rights point of view. In a situation as special as this one, where we are free to assume that there are no bad side effects of our actions (we may even keep it secret, if necessary — perhaps this is what happened in the Andes!) it is reasonable to assume, however, that an implication of utilitarianism is that it was wrong to bury the Kantians. Perhaps one more life could have been saved, if their bodies had been consumed as well. They should have been told that they should not be eaten, but, even so, after their death, they should have been eaten. Very likely, it was wrong according to utilitarianism, to bury them.
Does utilitarianism imply that Harris’ survival lottery should be made obligatory? It does not. The problem with the lottery is that we can only have it on an institutional basis, and because of our reluctance to accept to be killed, even in order to save lives, the establishment of the system would create terror among us. However, if it could (and this is a mere thought experiment) be established secretly, with no bad side effects, then, according to utilitarianism, it ought to be established. After all, with the system more lives would be saved than if we did not have it.
4.1 HARMAN’S MORE REALISTIC EXAMPLE
It is hard to imagine how Harris’ lottery could be established systematically and secretly. But think instead of an example given by Gilbert Harman:
You have five patients in the hospital who are dying, each in need of a separate organ. One needs a kidney, another a lung, a third a heart, and so forth. You can save all five if you take a single healthy person and remove his heart, lungs, kidneys, and so forth, to distribute to these five patients. Just such a healthy person is in room 306. He is in hospital for routine tests. Having seen his test results, you know that he is perfectly healthy and of the right tissue compatibility. If you do nothing he will survive without incident; the other patients will die, however. The other five patients can be saved only if the person in Room 306 is cut up and his organs distributed. In that case, there would be one dead but five saved.148 Again it is hard to believe that the killing of one in order to save the five could take place secretly. But let us assume that, in some rare situation, it would be possible for the physician to be absolutely and correctly sure, that if he would do so, no one would ever come to know about it. Here it is possible to think that, if he kills the one and save the five, this could happen without any bad side effects. In particular, the public support of the health care system would not be jeopardized. We must also assume that no one will miss the killed patient. If this was so and, if he did, and if there were indeed no bad side effects, what he did would the right thing, according to utilitarianism. It would have been wrong for him not to kill the healthy patient in Room 306 in order to save the needy ones. In this verdict the Sanctity-of-Life Doctrine, combined with the principle of double effect may concur. When the doctor cuts up the patient in Room 306, the intention need not be to kill him. The idea is just to use his organs to save lives. If she could take the organs without killing the patient, then she would do so. From a Kantian perspective, however, what she does is wrong. She uses the patient as a mere means.
4.2 UTILITARIANISM ON TAUREK’S EXAMPLE
Finally, what is a utilitarian to say about Taurek’s example? Taurek himself has articulated the answer. He assumes that David, who needs all the medication, also owns the medication and that the utilitarian should provide David with a reason to give away what he owns. David asks why he should give it away:
I think David’s question deserves an answer. What could there be about these strangers that might induce David to think it worth giving up his life so that they might continue to live theirs? The usual sort of utilitarian reasoning would be comical if it were not so outrageous. Imagine any one of these five entreating David, “Look here David. Here I am but one person. If you give me one-fifth of your drug I will continue to live. I am confident that I will garner over the long haul a net balance of pleasure over pain, happiness over misery. Admittedly, if this were all that would be realized by your death I should not expect that you would give up your life for it. I mean, it may not be unreasonable to think that you yourself, were you to continue to live, might succeed in realizing at least as favourable a balance of happiness. But here, don’t you see, is a second person. If he continues to live he too will accumulate a nice balance of pleasure over pain. And here is yet a third, a fourth, and finally a fifth person. Now, we should not ask you to die to make possible the net happiness realized in the life of any one of us five. For you might well suppose that you could realize as much in your own lifetime. But it would be most unreasonable for you to think that you could realize in your one lifetime anything like as much happiness as we get when we add together our five distinct favorable balances.149 This captures nicely the utilitarian reasoning. It should be noted in passing that, according to utilitarianism, the numbers do not count, at least not directly. It is the sum total of happiness produced that counts. In the quoted passage, in contradistinction from other passages of his paper, Taurek caters for this. Taurek himself does not find this argument convincing. He even finds it outrageous. But what better argument could be put forward to David? David may retort, of course, that the sum total of happiness realised if the five are given the medication is not felt by anyone. This is a standard objection to utilitarianism, of course. But a standard answer to it exists as well. The standard answer is that, even if the sum is not felt or experienced by any one, it is a sum of felt or experienced happiness. Note that it is true also that the net balance of pleasure over pain within a single life — referred to by the person confronting David in the quotation above — is not anything that this person or anyone else ever experiences. And yet, we find that it has the utmost importance to us. Again, it is a net balance of experienced pleasure over experienced pain.
The survival lottery I presented to the Chinese, the Russians, and the Americans, was not very popular. This is true in particular of the Chinese and the Russians, but even among the Americans there was a strong majority condemning it. This may be seen as good news for deontology; it is the only theory discussed in this chapter that concurs in the majority verdict. However, my own intuitions are quite different. Furthermore, I tend to believe that an important reason why people dislike the survival lottery I have presented to them has to do with aesthetics rather than ethics. These people have a strong inhibition against cannibalism. Their gut feelings tell them that it is wrong to consume human flesh. They may be right about this, as a general rule, but their gut feelings lead them astray in the case presented to them. Or, so I would be prepared to argue, anyway.
Would I have obtained a different result if instead I had presented John Harris’ survival lottery to them? I doubt it. As I have noted, since in his example killing is done on a regular basis I believe most people would chose not to join in. The lottery would enhance our chance to survive slightly, but it would not maximise happiness (because of our irrationality). So the right way of assessing Harris’ example is to see it as a mere thought experiment. Again, it might be difficult to get people in a survey such as this one to accept this assumption. Anyway, surveys such as these are extremely costly, so it is not possible for a philosopher to raise funding for any experimenting with them. My survey was indeed a single shot.
Yet, for all that, there seems to be a good idea to accept, in principle, the idea that we arrange survival lotteries, in particular if this does not only enhance our chances to survive, but also maximises the sum total of happiness among us. This is what happens in the example I presented in my survey and this is what would happen if rational people were to organise Harris’ kind of lottery. You may even imagine here that people who draw the winning ticket in the lottery suffer less than a person who has to accept that he will die because there is no organ available for him. Those who draw the winning ticket in the lottery, if perfectly rational, can comfort themselves with the thought that their death is not in vain. When they accept to be killed they know that they save many lives.
If I am correct in this assessment, then this is bad news for deontology. The implications of the deontology are at variance with the idea that it is defensible to save lives in the manner described in the examples. But I find it hard to see anything wrong, at least in principle, with such lotteries.
However, both utilitarianism and the moral rights theory are compatible with this judgement. They both imply that it is defensible to save one’s life in the manner described in the example. You could say that they both explain this observation. Which theory gives the best explanation of it?
Some may find that it counts against utilitarianism that, not only does it sanction the lottery as it has here been described, in the air-crash example, it also sanctions that those who stay outside the lottery are consumed for food, once they are dead. And it may be thought that it counts even heavier against utilitarianism that it would sanction the physician in Harman’s example who secretly kills a patient and uses the patients organs to save several other patients — and gets away with it. This is certainly Harman’s own opinion. The moral rights theory seems, on the other hand, to draw the line where it should be drawn. The lottery is acceptable if and only if those who voluntarily join into it in fact accept it.
However, we have seen in relation to Taurek’s example that the moral rights theory has an implication that many may find equally objectionable. On the moral rights theory, we need not save lives, even if we can do so easily. We need not share our medicine with others, provided we own it, even if this means that they die, and we need not make any efforts at all to save people in need; after all, we own ourselves, and we can do as we see fit with our physical and mental capacities, so far as we do not violate any negative rights.
I leave the final verdict of the theories to the concluding chapter of this book.
Chapter 9. Killing In War
If our country were offered guarantees for its safety and sovereignty by the United Nations, now equipped with military forces of its own, would it then be in order if our government decided to disarm completely only offering some restricted military resources to the United Nations, to keep the peace in the world?
China Russia USA
Yes 22 % 15 % 31 %
No 68 % 79 % 63 %
Don’t know 10 % 6 % 6 %
After each one of the twentieth century’s two terrible world wars attempts were made to obviate war altogether, through international organisations, first the League of Nations and then the United Nations. These attempts have not yet come to fruition. This does not mean that this kind of approach to the problem must be a dead end. Some people still strive to establish a global democracy where there is no more war (or global injustices, or environmental disaster).150 And one out of five Chinese citizens, and almost one out of three Americans, are prepared to give up the national defence, if only the UN provides guarantees. I find that extremely hopeful, considering that this is an idea that has hardly been discussed among people at large. However, we still live in a world with wars, even if we happen to live at a comparatively extremely peaceful time.151 And yet, many bloody wars are in progress while I write this. And so long as we do not have a world government, with a monopoly on the use of armed force, the situation could easily worsen (again). Hence, another approach to war than just abolishing them has also been tried out. Rather than to obviate wars, this approach intends to render them less prominent and less brutal. During the last century a lot of rules, with their roots in an old moral (philosophical) tradition about just and unjust wars, have been established in international law, codified in the Laws of Armed Conflict (LOAC).152 They specify both when war is legitimate, jus ad bellum, and how a war should be conducted, irrespective of whether it is just or unjust, jus in bello. The focus in this chapter is on this attempt and, in particular, on the legal idea of jus in bello.
War is hell, but yet, for all that, war is regulated by the LOAC. In today’s world, there is a widely shared view that there are just and unjust wars. The ones that are unjust are the ‘aggressive’ ones. The ones that are just are those fought in ‘self-defence’. Perhaps there are also just wars fought on humanitarian grounds.153 Here the LOAC are less clear. It is also part of the LOAC that wars, irrespective of whether they are just or unjust, can be fought in a just or unjust manner. In international, positive, law these two distinctions are thought of as independent of one another. Jus ad bellum is one thing, jus in bello quite another thing. What will interest me here are both, but my main focus will be on jus in bello. Can we make moral sense of roughly the legal rules that exist? In particular, can we make sense of the idea that jus in bello can be understood as independent of jus ad bellum?
I will not go into any detail in the description of the rules themselves. It should suffice to note here that among them are rules to the effect that it is illegal to target civilian bystanders, that even if military actions directed at the enemy solders may ‘spill over’ as direct harm done to civilian bystanders, this should never be the aim of the attack. All sorts of measures should be taken to minimize harm to civilian bystanders and, if there is no reasonable proportionality between the aim of harming the enemy and the damage done to civilian bystanders, then a military action is criminal. There are also rules about how prisoners of war should be treated. It is prohibited to punish them, once they have surrendered and so long as they accept their status as prisoners. To kill them is no better than to commit murder.
The idea is that international courts should punish violations of rules such as these, when possible.
Does a rationale behind rules such as these exist? In order to find this out it is crucial to look at very basic moral ideas and try to apply them to the problem at hand. This is what I will attempt to do in the present chapter. This approach, in this application of it, is controversial, however.
2.1 MORAL METHODOLOGY
In the preface to his seminal work on the ethics of war, which inaugurated the recent philosophical interest in the subject, Michael Walzer claims that it is pointless to move bottom up from the most basic moral ideas to the more practical precepts we know in our ordinary moral discussions about war as well as in positive international law. If we were to start our discussion, as I do in this book, and as I plan to do in this chapter, with the basic moral principles, we would ‘ .... probably never get beyond them’, he writes.154 In this chapter I will try to show him wrong on this count. And it should be noted that, in his rejection of such an approach, he is not quite consistent. After all, he claims that he is basing his argument on a moral rights theory. ‘The morality I shall expound is in its philosophical form a doctrine of human rights’.155
I suppose what Walzer means to say is that he doesn’t stick to any very precise theory of rights. Moreover, he is not trying to apply it. Rather he adheres to a more intuitive method, where the theory is moulded while his argument is developed. My idea is, as we have seen in this book, that we can do better if we apply different theories to the problem and see our enterprise as a kind of crucial test. Which one of them can best explain our data? And the datum to be explained in this chapter, more exactly, is the intuition that existing rules of law are roughly as they should be. This is my firm belief. Theories that cannot explain this datum are in trouble. The theory that best explains it gains support — through an inference to the best explanation.
2.2 THREE DIFFERENT APPROACHES
I will look at the three basic moral views I have discussed thus far in this book also in this chapter, in order to investigate their implications for the problem at hand: a deontological theory, both in its Kantian retributivist the form and in form of the Sanctity-of-Life Doctrine where the principle of double effect is part of it, a libertarian moral rights theory, and utilitarianism. I will try out these basic moral approaches to the problem and try to find whether they can provide us with a rationale behind existing positive international law. To the extent that they cannot, I will hold this against them, not against positive existing international law.
I should caution the reader that, even if these three traditions, in my description of them, are clear and distinct, this does not mean that thinkers fall neatly into them. It is very common that actual historical thinkers incorporate elements from more than one of these traditions in their thinking. This is true in particular in the discussion about just and unjust wars. I consider this bad intellectual habit, but it is a fact.
As I understand deontology here, it is taken to include a strict prohibition against the killing of innocent human (rational) beings. According to the theory there is also an imperfect (utilitarian) duty to make the world a better place. It is not quite clear what it means to say that this is an imperfect duty, but this is of no importance in the present context. Here it suffices to note that if much is at stake, if important values are threatened, there is a need to do something about this. However, it is never permissible to violate any perfect duty in the attempt to fulfil an imperfect one. It is not permissible to kill even in order to save lives. I do not assume that an adherent of deontology must think that all immoral actions should be prohibited. For example, it is natural to think that an adherent of the theory is prepared to live with a legal system not forbidding lying, at least not forbidding lying in all sorts of circumstances. It does seem though, as if most adherents of the theory would argue that serious kinds of wrongdoing, such as murder, must be kept illegal. In particular, Kantianism typically goes together with a retributivist theory of crime and punishment. It is important that those who intentionally perform immoral actions, such as murder, should be punished. But then it is crucial to have such actions criminalised. Now, since this will turn out to be a problem with the theory in the present context, I will also discuss whether the theory can be relaxed in this respect. More about that later.
Who is a human being? Who is a rational creature? We have been deeply into this problem in the chapter on abortion. How we understand deontology in this regard is of importance if we discuss problems such as abortion, but it is of no importance in the present context, so for now I just set it to one side. The victims of war typically qualify on all counts.
3.1 DEONTOLOGY AND PACIFISM
If life is sacred, should one ever kill? On one plausible reading, pacifism is the logical conclusion that follows from deontology. But that might be a bit simplistic. Could not a war in defence of important values such as national independence, democracy, and welfare, render war legitimate? I think there are two promising lines of argument to this effect available for the adherent of the doctrine.
First of all, it is possible to argue that even if one aims at the attacking combatants in a defensive war, the intention is not really to kill them. At least this is so if war is fought without anger, and only as a means to saving important values. You rely then on the principle of double effect which forms part of the version of deontology I have called the Sanctity-of-Life Doctrine.
It is true that weapons kill but it is possible to fight a defensive war with the following intention. I throw my deadly weapons at the enemy in order to incapacitate him, foreseeing his death, but not intending it; had it been possible to incapacitate him without killing him (putting him to sleep for a week, say), I would have done so. If this is the case, on the traditional understanding of the principle of double effect, the death of the enemy combatants is merely a foreseen, and not an intended effect; the intention was to defend important values such as national independence, democracy, and a happy life for all. The Augustinian view that wars should be fought with the right attitude seems to reflect this kind of thought:
The passion for inflicting harm, the cruel thirst for vengeance, an implacable and relentless spirit, the fever of revolt, the lust of power, and such like things, all these are rightly condemned in war.156 It is true that, on the principle of double effect, a requirement of proportionality (between the good you intend and the bad you merely foresee) should also be satisfied. Is it satisfied in defensive war? It is at least arguable that sometimes it is. If important values are at stake and no killing performed by the just combatants is intentional, then this may well be true. It is true if there is reasonable hope of success in the defensive attempt.
The other line of argument is different. You then refer to the necessity of capital punishment insisted on by Kant. Many adherents of deontology have also thought that there exists a strict obligation to execute murderers. They adhere to a retributivist theory of crime and punishment. Here is, as we remember from the chapter on capital punishment, what Immanuel Kant famously had to say about it:
Even if a civil society were to dissolve itself by common agreement of all its members (for example, if the people inhabiting an island decided to separate and disperse themselves around the world), the last murderer remaining in prison must first be executed, so that everyone will duly receive what his actions are worth and so that the bloodguilt thereof will not be fixed on the people because they failed to insist on carrying out the punishment; for if they fail to do so, they may be regarded as accomplices in this public violation of legal justice. (Metaphysical Elements of Justice, p. 140)
If those who attack you are no better than murderers (they are murderers), they deserve to die. This means that you are allowed to intend their death. On this reading, only losses among civilians on the aggressive side of the war count morally against your actions. But then it is quite plausible that the requirement of proportionality can sometimes be satisfied.
The idea that war could be seen as just (deserved) punishment was often alluded to in the medieval writings about just and unjust wars. Those ‘ ... who are attacked should be attacked because they deserve it on account of some fault, as Aquinas wrote.157 And those who fight on the wrong side are ‘ ... guilty of strife and commit sin’.158
One may argue, of course, that many who fight on the wrong side never kill any one. How, then, can they all be legitimate targets? I suppose the answer to this question is that you are complicit in the killing that goes on by just being a member of the army responsible for it.
3.2 JUST WARFARE
All this means that, on the doctrine, if interpreted in a charitable manner, there is room for defensive war. It makes room also for a requirement of restraint when a defensive war is fought, roughly in the manner it is defined in international law. Those who fight on the right side should respect the immunity of innocent bystanders on the aggressive side. And even if they foresee in some circumstances that their actions targeted on the enemy combatants spill over on innocent bystanders, their action may be morally permissible. The crucial thing is that they do not to aim at the bystanders. If they do, they act as murderers. The requirement of proportionality should furthermore be respected. Finally, of course, they shall do what they can to minimise the harm done to innocent bystanders, when such harm is inevitable.
Michael Walzer is critical of the principle of double effect and proportionality. It is too permissive he thinks. It is not enough to see to it that the requirement of proportionality is met and to see to it that you do not intend the death of civilians, you must also be prepared to pay a price yourself in order to lower to costs among civilians. At least you need to take some ‘extra’ risks:
... if saving civilian lives means risking soldier’s [sic] lives, the risk must be accepted. (Just and Unjust Wars, p. 156)
This seems to imply that you should be prepared to jeopardise victory on the right side in order to keep the costs among civilians on the bad side low. If this is how he should be understood, then I do not share his intuition here. And I note that the Sanctity-of-Life Doctrine here seems to be in agreement with actual international legislation. It can provide a moral rationale behind it, then.
Perhaps Walzer just wants to claim that risking soldiers’ life is only obligatory when it has no expected effect on the outcome of the war. But then his observation has little bite. Few military commanders would feel that they have more soldiers than they need.
Literally understood the principle of double effect urges a military commander to attack only combatants, even if they know that by doing so they will kill more innocent civilians than if they actually target only a few of these very same civilians. This is a congenial understanding of the principle, I submit, even if this has been questioned. It has been contested by Frances Kamm who argues that:
Being constrained by someone’s right not to be treated in a certain way (e.g. terror killed rather than collaterally killed) may not have the same moral significance when it will not make a difference to whether he is harmed or terrorized as it has when it makes such a difference. This is an instance of what I call the Principle of Secondary Permissibility (PSP).159
I do not think a deontologist should accept the Principle of Secondary Permissibility. To do so would be to compromise the rationale behind deontological thinking. And, as we will see, the principle of double effect plays no role at all in a moral rights theory (it is not quite clear whether Kamm relies on deontology, as defined in this book, or on a moral rights theory).
In a similar vein we should reject the idea that a morally bad bombardier, discussed by Kamm,160 who secretly targets civilians when he drops the bombs on a munitions plant, does nothing wrong according to the principle of double effect. It is true that his action is in conformance with international law, provided there is a reasonable proportion between the good achieved by destroying the munitions plant and the deaths among civilians, but if his real motive is to destroy the civilians, not the munitions plant, then he acts wrongly according to deontological thinking.161
If deontology can provide a rationale behind the international law controlling those who fight on the right side it fails completely when it comes to those who fight on the wrong side, however. They are no better than murderers. They deserve to die, according to the retributivist version of the doctrine. And even setting retributivism to one side, they act wrongly. Even if murder need not be retributed in kind, it is still plain wrong.
This has the funny implication that prisoners of war, taken from the bad side in the war, should be punished (executed). After the war, not only those who started the war, but all those who took part in it, should be punished. This flies in the face of international law. And this is a serious problem for the doctrine.
3.3 MORAL SYMMETRY
I have claimed that, typically, those who defend the theory are not prepared to allow that a serious crime such as murder goes unpunished, let alone that it is not rendered illegal. That’s why the adherent of the doctrine needs to show that there is some moral symmetry between just and unjust combatants, otherwise there exists no rationale behind existing legislation. And they have gone to great length in order to show this. It seems to be a hopeless task, however.
One could try to argue that those who fight the unjust war (the unjust combatants) have often not chosen to do so. This is often not true. But even if in some cases it is true, this is of little avail from the point of view of this doctrine. For the prohibition against killing is strict. Not even if threatened at gunpoint are you allowed to kill an innocent person just in order to save your own neck.
It has also been noted in the recent discussion, for example by Jeff McMahan, that on this kind of theory, there is no way of meeting the proportionality requirement, when innocent bystanders are killed by unjust combatants. They do the wrong thing when they kill the just combatants. And they do even further wrongs when their actions spill over to the bystanders. So there is no moral room for jus in bello, when you fight an unjust war, not even if the matter is assessed from the point of view of the Sanctity-of-Life-Doctrine with its stress on the principle of double effect. The idea behind the requirement of proportionality is that the bad effects you merely foresee should be proportionate to the good effects you intend. But to intent to kill the just combatants is evil, not good.
... in order for the killing of innocent civilians not to be excessive in relation to the expected military advantage of an attack, military advantage must be good. Yet military advantage for those who are in the wrong and are fighting for the defeat of a just cause is, except perhaps in highly anomalous conditions, bad, impartially considered.162 It is no coincidence that Immanuel Kant had his plan for perpetual peace, rather than a philosophy of just war, and that the Pope tends to pacifism. From a deontological point of view, no support can be gained for existing international rules of just warfare. Deontology cannot sustain the idea of a legal symmetry between just and unjust combatants since it cannot acknowledge any moral symmetry between them.
Could the adherent of deontology give up on the notion that murder should (always) be illegal? Could the adherent of the theory accept, on utilitarian grounds, that many murders (committed by unjust combatants) should be legal, in the way they are legal in international law? This seems to be a high price to pay. I will discuss this possibility at the end of the chapter, however.
4. THE MORAL RIGHTS THEORY
If deontology, not even when interpreted as the Sanctity-of-Life Doctrine, with its reliance on the principle of double effect, cannot provide a rationale behind existing international legislation it is tempting to follow Michael Walzer and turn to a moral rights theory. He has famously argued that from a moral rights theory it is possible to provide a rationale behind existing international law on jus in bello by showing that combatants on both sides are morally speaking equal. Here, as in the rest of this book, I will discuss the moral rights theory in its strict libertarian version rendered famous by Robert Nozick in his Anarchy, State, and Utopia. On this version of the theory we all own ourselves. This means that we can do as we see fit with ourselves, so long as we violate no (negative) rights of anyone else. It is a violation of my rights if someone actively intervenes to stop me from doing what I want to do with my own body, my own talents, and so forth. The worst kind of violation of my rights takes place if, against my will, I am actively killed by someone. I have the right, on the other hand, to kill myself or to hire some one else to do the job for me. We have met with these implications of the theory in the chapters on suicide and euthanasia.
There is no room for any idea of just (retributive) punishment in this theory, as we saw in the chapter on capital punishment. Nor is there any room for any ‘positive’ rights. No one is allowed actively to kill me, unless I consent to it, but I have no right to active assistance when my life just happens to be in danger. On this moral theory, you are allowed to live high and let die. This may be considered pretty mingy of you, but not any example of wrongdoing.
One important aspect of the moral rights theory is the idea that we have a right to self-defence. When my life is threatened, I have the right to kill in order to save my life. A third party has also a right to intervene and help me to defend my life against an aggressor. However, if he doesn’t want to do this, he is at licence to abstain.
If the moral rights theory makes room for war at all, it is with reference to the idea of self-defence. A just war is a war fought in self-defence. This defence is not without complication, however. Often, when my state is attacked, this does not mean that I am attacked. Can a libertarian moral rights theory allow that I see the defence of my state as self-defence?
This is not as straightforward as when I defend myself against someone attacking me. However, if the state is a minimal state, I suppose I have a right to defend it. And even if it is not a minimal state — in real life it never is — it might be possible to argue that to the extent that it contains a minimal state (in the form of some basic legal structures, a police force, a military defence, and so forth), I may also be allowed to defend it. So let us grant the moral rights theorist the claim that defensive war is legitimate.
Can the moral rights theory defend the idea that innocents should be spared in a defensive war? It can, and very much so. As a matter of fact, this spells problem for the idea. Michael Walzer complained that the Sanctity-of-Life Doctrine, with its reference to double effect, was too permissive on this count. If he wants to be true to the moral rights theory this is certainly so. The stress in this theory is against active killing (derived from the right of a self-owning individual not to be actively killed). There is no room in the theory for active killing even if death is merely foreseen. The principle of double effect is not part and parcel of this theory. So it seems that the only kind of defensive war acceptable to the moral rights theory is a war fought in the desert or at sea (or in outer space).
This aspect of the theory is rendered clear in a recent contribution by Judith Jarvis Thomson, to be discussed more carefully in chapter 11. She is a moral rights theorist, although not of Nozick’s extreme variety (she leaves room for some positive rights as well, but this is of no importance in the present context). In a comment on the discussion, engendered by Philippa Foot and Thomson herself, on the so-called trolley cases, she makes a rather remarkable observation. It concerns the first, simple Switch case, where you can save five innocent persons down the track if you flip a switch and redirects a trolley onto another track where one innocent person is placed. On the Sanctity-of-Life Doctrine this is what you ought to do. You merely foresee the death of the single person when you act in order to save five lives. And the requirement of proportionality is fulfilled. It is permissible to foresee the death of one person if the alternative is the death of five persons. According to the principle of double effect, it is not permissible not to flip the switch, however; if you do not flip the switch, it is true that you merely foresee the death of five persons, but their death is out of proportion when instead your could have acted in a manner that would have meant that only one life would have been sacrificed. On the moral rights theory, however, you are not even allowed flip the switch. You may sacrifice your own life in order to save the five but you may not actively kill the one innocent bystander (by flipping the switch) in order to save the five. This seems congenial to the moral rights theory. And it shows that this theory cannot cater for the lenience in international law when it comes to the active killing of civilian bystanders.
If this observation is correct, then, according to the moral rights theory, it is impossible even for the solders on the right side to fight an ordinary war, where there are civilian losses, in a morally permissible way. At least this is so unless these civilians can correctly be seen as complicit in the unjust warfare.
But is the observation correct? In the example I actively kill one in order to save five, but no violations of rights takes place if I do not act. Suppose instead that the trolley has been put in motion by a murderer who wants to kill the five. Does this mean that I have a right, actively to kill the innocent person on the side-track, by diverting the trolley onto him? I would think no, if we adhere of the core of the moral rights theory. I am not allowed to violate rights in order to minimise the number of rights violations. The adherents often take a pride in pointing this out (even if, as we saw in chapter 2, they have a hard time explaining why this is so).
Suppose finally that the trolley is running towards me. It has been put in motion by a murderer. Am I allowed to divert it onto a track where an innocent bystander will be killed? Perhaps I am allowed to do so according to some kind of common sense morality, to be discussed in Chapter 12, but I am not allowed to do so according to a strict understanding of the moral rights theory. I am not allowed actively to kill innocents even to save my own neck. To bend the theory to accept this kind of killing would be to compromise the rationale behind it.
4.2 MORAL SYMMETRY
The theory has a problem also in defending the idea that, as a combatant, you have a right to serve on the wrong side. Michael Walzer thinks that those who serve on the wrong side have a right to do so and, hence, he takes it that they too kill in self-defence. After all, the just combatants threaten the lives of unjust combatants. Only the political leaders, who initiate an unjust war, are to be seen as criminals.
But how can it be that those who fight on the wrong side have a right to do so? How can it be that they have a right to kill the combatants who fight for a just cause? This is the kind of explanation Walzer provides:
In our judgments of the fighting, we abstract from all consideration of the justice of the cause. We do this because the moral status of individual soldiers on both sides is very much the same: they are led to fight by their loyalty to their own states and by their lawful obedience. They are most likely to believe that their wars are just, and while the basis of that belief is not necessarily rational inquiry but, more often, a kind of unquestioning acceptance of the official propaganda, nevertheless they are not criminals; they face one another as moral equals.163 This is a strange statement to come for someone who believes in a moral rights theory. Robert Nozick thinks differently. He has insisted that ‘ ... it is a soldier’ responsibility to determine if his side’s cause is just ... ‘ (Anarchy, Sates, and Utopia, p. 100). The implication is, of course, that if it is not just, it is just plain wrong to participate. Walzer quotes him and objects that those who don’t refuse to take part are not blameworthy.164 That is certainly so, on the moral rights theory, which does not deal in blame and retribution. However, if they don’t refuse, they risk becoming aggressors. And aggression is wrong, regardless of whether it is ‘blameworthy’ or not.
Frances Kamm has recently defended the symmetry with a move similar to Walzer’s. It is often reasonable to believe that those who fight on the wrong side believe that they fight on the right side. This is of no importance, according to Nozick. It is of the very last importance, according to Kamm, following Walzer. She stresses the point with the following analogy:
... suppose two people are debating an important issue to which no one yet knows the correct answer; they take opposite positions and one and only one of the positions is correct though no one yet knows who is correct. Are these debaters moral equals entitled to do the same things in the debate? We know that one of them must be defending an untruth (though he is not lying), and he may influence other people to believe in an untruth on an important matter. We would ordinarily think these debaters are moral equals and that it is permissible for each to do what the other does to defend his view and defeat his opponent in the debate.165 Is this not an utterly strange analogy? Do we not grant equal rights to those who debate since we believe that this is helpful, if we want to find out the truth about the matter at hand? And being defeated in a debate is not like being killed in a war, or is it?166
Could one find a better rationale behind the right on the part of those who fight on the wrong side, to kill the person who fights on the right side, founded on the fact that combatants pose a threat to one another? This seems to be the standard rationale behind the idea of moral equality between combatants on the just and the unjust side of the war, but it does not stand up to reason, at least not if founded on a moral rights theory. Jeff McMahan has famously argued against this notion of moral symmetry with reference to the fact that a murderer has no right to kill a policeman in self-defence, and he makes references to both Pierino Belli and John Locke (who belong, of course, to a moral rights theory), who long ago made the same observation.
One would expect a defence for moral symmetry, in this theory, with a reference to consent. Could one argue that combatants, by becoming combatants, consent to being killed, regardless of whether they are on the right or the wrong side? One could compare here with participants to a boxing contest.
Walzer alludes to this and the idea has been defended by Thomas Hurka.167 It is hard to believe, however, that it is true that those who engage in a modern defensive war consent, in the relevant sense, to be killed.
Walzer agrees about this and then resorts to the idea that combatants on both sides are like gladiators, who have to face one another in a deadly battle.
But is this true of those who enrol themselves only in order to defend their country when it is submitted to an unjust attack? This is hard to believe. They may well volunteer and they may well do so for a good reason. And very often the unjust combatants have joined the armed forces voluntarily too, but for a bad reason. Moreover, on the moral rights theory, you are not allowed to attack innocent people even if you are coerced to do so. There are kinds of coercion you are supposed to resist, even at high risk on your own part. You are not allowed to kill a guard and rob a bank, only because some criminal told you that, unless you complied, he would punish you. Killing the guard and robbing the bank is still a violation of rights.
Could one say instead that, when we, ordinary citizens, consent to living in a world of nation states, which is much like a Hobbesian state of nature, we also accept war between states? After all, we could establish a world government and do away with war altogether on a global scale, in the manner we have done within each nation. When we don’t, mainly because we do not want to do so,168 we must also accept the fact that our state is liable to attack from other states, whenever it proves weak and vulnerable? War is a regular thing, to be expected to happen in the kind of world in which we have chosen to live.
The argument is shaky. First of all, it doesn’t show that everyone has consented to being killed by combatants from an aggressive army. Some of us do whatever we can to establish global democracy. This is crucial from a moral rights perspective. Secondly, this would still not suffice as a defence of existing legislation about just warfare. In particular, this would mean that we would have to give up on the idea of jus ad bellum, and restrict ourselves to ideas about jus in bello. Now there would be no unjust wars. So it would not be possible to hold political leaders responsible for having initiated them, let alone to punish them. But international legislations allows us to do so.
In one respect, the moral rights theory does provide a rationale behind the LOAC. Since it does not rely on any retributive ideas, it caters for the idea that prisoners of war should be treated in accordance with international conventions. After all, when they have surrendered, they pose no threat any more, so it would be wrong to kill them.
This rejection of retributive thinking may be seen as an intellectual resource in the present context. It is true that on the moral rights theory, you are doing the wrong thing, when you fight on the wrong side. But there is no point in punishing you, when war is over. Then you pose no threat any more to any one. So your moral crimes should go unpunished.
Does the idea leave room for punishment directed at those who started the unjust war? This is what is required by the LOAC. But if the moral rights theory has no room for retribution, how is it possible to defend such punishment? The obvious answer must be that people who have succeeded in starting unjust wars may do so again. They still pose a threat to the world. So we had better incapacitate them, either by executing them or by sending them off to something like a St. Helena, in our own defence.
The theory succeeds in providing a rationale behind existing legislation on some counts, but not all of them. It is very (much too) restrictive in its defence of civilian bystanders. And it cannot cater for the idea of equal treatment of combatants on both sides.
The reason moral rights theorists try to establish moral symmetry between just and unjust combatants is that they, like deontologists adhering to the Sanctity-of-Life Doctrine, want to keep serious wrongdoing (murder) illegal. I have hinted at the possibility that deontologist give up this ambition. Perhaps moral rights theorist can do so as well? In one way it may be easier for them, than for deontologists, to do so. They do not adhere to any retributivist theory of crime and punishment. I will return to this possibility below.
According to utilitarianism we ought to do what maximises the sum total of happiness in the universe. This criterion applies to all actions, also to the adoption of international laws. We ought to adopt the laws that are optimal in the sense that, having them, maximises the sum total of happiness in the universe. It is likely that, for the time being, existing legal rules of just and unjust wars, the LOAC, are close to the optimal solution. But then we ought to stick to them. It is as simple as that.
This is not to say that all acts condoned by these laws are right. Many of them are wrong. This is true in particular of most actions performed by unjust combatants. Yet, we should have them legally permitted since having them legally permitted have the best consequences. There are also actions that are right, even obligatory, but forbidden by the LOAC. There are bound to exist at least some situations where by violating the rules on jus in bello it is possible to hasten peace and hence of making the world a better place. In such situations, according to utilitarianism, these actions are right (even obligatory). And yet, if the agent is found out, she should be punished. A utilitarian, who has access to the notions of blameless wrongdoing and blameful rightdoing, has no problem in acknowledging this.169
Note that utilitarianism is flexible. The first explicit rule of warfare I know of is the one instituted by Solon in Athens to the effect that, after a civil war, no one should be punished for having fought on the wrong side. People who had not taken part in the war, however, should be punished. I suppose the idea was that this was a way of making civil wars short and reducing resentment post bellum. If it worked, it could be supported from a utilitarian point of view. Perhaps there was a time when it was impossible to render war less terrible through legislation. In those times, there existed a utilitarian rationale for not having such a regulation. In today’s world, with the UN, the International Criminal Court, the possibility of arranging international sanctions, and so forth, it seems wise to regulate warfare and to have explicit rules about how a war, just or unjust, can be fought in a just manner. There may come a time when we do not need those rules any more. I alluded to this in the opening section of this chapter. It is of note that not only Immanuel Kant, but also Jeremy Bentham thought that it would be an easy task to do away with wars. According to Bentham, it was just a matter of establishing a free press, abolishing the secret diplomacy (in the way Wikileaks has managed today, I suppose) and of having the colonial powers giving up on their colonial ambitions.170 We know today that this was a bit naive. Actually, Bentham was even more naive than Kant. However, at least some utilitarians keep struggling to find a way of obviating, rather than regulating, war. My personal opinion is that this requires a world government and global democracy. If global democracy can indeed do the trick, it has a utilitarian rationale.
5.1 INTERNATIONAL LAW
If the utilitarian solution to the problem here discussed is so simple, if utilitarianism can provide a rationale behind existing international legislation, why has utilitarianism come into disrepute in the present context? Partly this has to do with a lack of understanding of utilitarianism, on the part of its critics. But partly it has to do with how some utilitarians have themselves argued. This is true in particular of R.B. Brandt. It is a sad fact that this, otherwise brilliant philosopher, has brought the idea of a utilitarian defence of existing international rules of just war into disrepute. He has advocated the following rule, to be seen as an alternative to rules contained in existing international laws:
... a military action ... is permissible only if the utility (broadly conceived, so that the maintenance of treaty obligations of international law could count as a utility) of victory to all concerned, multiplied by the increase in its probability if the action is executed, on the evidence (when the evidence is reasonably solid, considering the stakes), is greater than the possible disutility of the action to both sides multiplied by its probability.171
This is ridiculous. Who can believe that anything of this sort would work? Each party in a conflict would take advantage of the kind of license it gives them. As a matter of fact, this looks like a caricature of utilitarian thinking! Jonathan Glover is right when, politely, he observes that the ‘ ... reason why Brandt’s rule is likely to be inadequate as an instruction to appear in army manuals is that it leaves too much to the judgement of the soldier or strategist’.172
In comparison to Brandt’s suggested rule existing rules gain the upper hand. Perhaps they are not optimal. Perhaps there is some room for some improvement. But Brandt’s suggestion would not be an example of this.
When thinking of possible improvement, however, why not convict the combatants on the wrong side? Their actions are, by and large, wrong, it is safe to assume. The fact that they often mistakenly believe they fight on the right side does not render their actions morally permissible. Two wrongs (one epistemic and one moral) do not make a right. In this moral verdict utilitarianism is no different from deontology or the moral rights theory. Furthermore, it is often true that the combatants on the wrong side could have formed more informed opinions about the war in which they were fighting, and it is then true that they should have done so. If they have to fear post bellum punishment this should make them think more clearly before they enter the war.
And yet, there might exist very strong utilitarian reasons not to punish them afterwards. The reasons are fairly obvious.
First of all, it would be impossible to punish them all. They are too many. It would be extremely costly. And to punish only some of them would invite the objection that it was unfair. It would be difficult to garner general acceptance of such a practice.
Secondly, it might be a good idea not to put this kind of pressure on ordinary combatants, in international law, since this would mean that the governments, who fight unjust wars, would go to even greater length to cover up their bad intentions, to plant evidence in defence of their moves, and so forth, in order to convince the combatants on the wrong side that they are on the right side. Think of the US war against IRAQ. There were no weapons of mass destruction hidden by Saddam, and both George W. Bush and Tony Blair eventually admitted that. This rendered easier a critical stance towards the war, once the facts were on the table.173
Finally, as has been stressed by David Estlund,174 at least in democratic countries it may be a good idea for ordinary citizens to act on the assumption that the elected leaders make better political judgements than they are themselves capable of making. One could here also gesture at Condorcet’s jury theorem to the effect that the majority (if each member of it is right more often than not) is right more often than any individual is, making it up.175 From a utilitarian point of view, this doesn’t make it morally permissible to take part in an unjust war, proclaimed by the democratically elected leaders. In this utilitarianism concurs in the verdict given by the Sanctity-of-Life Doctrine and the moral rights theory. It is still wrong — even if Estlund himself thinks otherwise. However, since the decision to joint the wrong side was made with the aid of a method of decision-making, that a rational person would follow, it seems inappropriate to punish the person who used it. His wrongdoing was blameless (as even Walzer have claimed, we remember). The notion of blameless wrongdoing (and of blameful rightdoing) have a safe place in utilitarian thinking. A problem with both deontology and the moral rights theory is that they can’t find any place for them.
5.2 LEGAL REVISION?
Or, is it after all a mistake to believe that it is a good thing to have such rules in place? When you start reading Jeff McMahan’s important book you may think that this is how he thinks, at any rate:
... the idea that no one does wrong, or acts impermissibly, merely by fighting in a war that turns out to be unjust ... lies at the core of the reigning theory of the just war and also informs the international law of war. Although the presence of this idea in the law is intended to have a restraining effect the conduct of war, the widespread acceptance of this idea also makes it easier, even for independently minded people such as Wittgenstein, to fight in war without qualms about whether the war might be unjust. (Killing in War, p. 3)
When you read this you may get the impression that McMahan wants us to give up the idea that unjust warriors should go unpunished, but that is not his intention (you realise this when you read his entire book). He only wants us to give up the idea of a moral symmetry between combatants on both sides. He wants us to stick to the existing rules, at least for the time being. He shares my intuition that, roughly, the existing rules are as they should be. And he wants to stick to them in spite of all the arguments he levels from a common sense view, a moral rights theory, and deontology (in a perfect blend) to the effect that international law is morally mistaken.
Why should we then keep to existing rules? For purely utilitarian reasons, it seems! Then it strikes a utilitarian reader, like the present one, that this concession should have given McMahan pause. Does not this imply that there is something wrong with the moral theories he has applied to the problem? If utilitarianism gives us the answer to the question what international laws we should adopt, should it not also give us the answer to the question which actions we should take when conducting a war (irrespective of whether we are politicians deciding about it, generals planning it in detail, or combatants fighting it on the battlefield)?
Even if existing legislation is by and large plausible, as assessed from a utilitarian perspective, there may exist ways of improving on it, without radically changing it. David Rodin has proposed that even if the existing rules of jus in bello shall apply to just combatants, stricter rules of jus in bello could be in place for unjust combatants.
... soldiers who fight in an unjust war have no right to use force against just combatants and should be held responsible for unjust killing post bellum, but just combatants do not possess additional in bello privileges.176 Now, if this would work out for the best, a utilitarian rationale exists for the proposal. I doubt it, however. And the reason is that violations of the rules of war are not only punished post bellum but also during combat. If just combatants were to ‘punish’ unjust combatants for doing what they allow themselves to do, there is a clear risk of escalation.177
5.3 A RIGHT TO RESIST UNJUST WARS?
There may be other and less extreme ways of revising the rules, however. McMahan considers the idea that there should be a right to conscientious objection, even among active-duty soldiers, with reference to their moral views. Such a rule could of course be added both to international law and to national laws. He defends it against the objection that such a rule would be disastrous to military efficiency and goes on to compare such a rule with the rule that soldiers should not be complicit in actions that violate the rules of jus in bello:
The suggestion that soldiers could be granted the right to disobey orders from a superior officer if they believed those orders to be unlawful would have been dismissed as utterly incompatible with the maintenance of discipline and thus of organized and coordinated action on the battlefield. Yet today a soldier is legally required to disobey a manifestly unlawful order — provided, that is, that the law the order would violate is an in bello rather than an ad bellum law — and military organizations function no less efficiently as a result. (p. Killing in War, p. 98)
This is a bad idea, however. It is a bad idea, not because it would risk military efficiency, but because it would lessen the resistance to unjust wars by allowing an easy way out for those who recognised that something wrong was going on. By the same token, a conscripted army is preferable to a professional one. And the draft should be quite general, and include both sexes. But it might be a good idea to have an internationally recognised right for combatants not to fight wars that, objectively speaking, are unjust. This right could come as a mere addition to existing international legislation. And it could also be implemented in national legislation. This is how I myself thought (see the preface) when, in the late 60s, I refused to do my military service. I was invited to prove to the military authorities that it was at variance with my conscience to kill other human beings. I refused to refuse on those ground. My reason for my refusal was that, objectively speaking, it would be wrong of me to take part in the military defence.
Is that suggestion absurd? It may seem so at first thought. And, certainly, my wish to refuse on those grounds in the 60’s, were not met with approval from the military authorities. However, at least in democracies with an independent legal system it would be feasible to have, say, a court of law, trying the objection from a combatant that the war her government had initiated was unjust. How interesting, if the court finds the war unjust! In other political systems, where the courts of law follow the political leaders, either because of lack of democracy or because of legal and political traditions, this would be absurd. It would be tantamount to the government itself judging that the war it had initiated was unjust. It is safe to assume that this will never happen.
This idea is interesting, but regardless of whether such a clause should be added to existing international law, I can’t help feeling that, for the time being, existing legal framework about jus ad bellum and jus in bello are, roughly speaking, appropriate. They are close to the optimal ones. If this is so, they do get support from utilitarianism, and from utilitarianism only.
5.4 WALZER ON THE UTILITARIAN RATIONALE
Not only McMahan, but also Walzer recognizes the possibility that existing rules granting innocent bystanders protection in war may have a utilitarian rationale. But while McMahan in the final analysis defends them on this count, Walzer objects to such a defence of them:
But no limit is accepted simply because it is thought that it will be useful. The war convention must first be morally plausible to large numbers of men and women; it must correspond to our sense of what is right. Only then will we recognize it as a serious obstacle to this or that military decision ... (Just and Unjust wars, p. 133)
But this is a utilitarian argument, not against having the laws we have, but against incorporating in the conventions a utilitarian rationale for them. This does not mean that utilitarianism cannot best explain why we should keep to existing rules. As I have argued and as McMahan has admitted, it can. But I think Walzer’s observation is correct. It is correct on more general (utilitarian) grounds. It is never a good idea to incorporate into legislation any explicit or implicit reference to any particular moral rationale behind it. If we did, the support for them would wither.
This has to do with two facts. On the one hand, at least in democracies, legal regulations are in some cases the result of an overlapping consensus. People who hold very different basic moral outlooks find it possible to agree about the law as it appears. On the other hand, in many other cases, an existing law is the result of a compromise. Then it is highly likely that no rationale behind it exists at all. In either case, if any reference to a moral rationale were included, the majority support of the law would be lost.
So Walzer is right when he insists that the rules of war should not refer, explicitly and even implicitly, to utilitarianism. But this is true of legislation in general. And yet, for all that, it seems as if utilitarianism can indeed provide a rationale behind them. As I have indicated, I count this in favour of the plausibility of utilitarianism.
Note furthermore that, in his comments on Nozick, even Walzer resorts to a distinction between wrongdoing and blameworthiness. A way of understanding him is as follows. Those on the wrong side, who kill their enemies, are doing the wrong thing, but they are not blameworthy. We should not blame them. This move is open to the utilitarian, but not to a strict moral rights theorist (it is resisted by Nozick).
5.5 SOME UTILITARIAN IMPLICATIONS
If it counts in favour of utilitarianism that it can provide a rationale behind existing laws on just warfare, does it not count against it, that it has implications in individual cases at variance with these laws? Two examples come to mind.
First of all, international laws condemn terrorist acts, performed both by states and individuals. This is as it should be, if I have understood the implications of utilitarianism correctly. However, there may well exist cases where individual acts of terror are morally obligatory; they may for example put a quick end to a terrible war. Does this count against utilitarianism?
Secondly, even the participation in an unjust war may have a utilitarian rationale. If I join the bad guys this may mean a quick end to the war and a better peace than if I fight them. Then, according to utilitarianism, I ought to take part in the unjust war. Does this count against utilitarianism?
We know this kind of double utilitarian standards from many other contexts. As I have insisted, repeatedly, I see them as a sign of subtlety and strength in the utilitarian creed, not as a weakness. My strong intuition is that we ought to resort to terrorism, killing some innocent people, if there are no bad side-effects, and if by so doing we can put an end to a terrible war and hence save more lives than the ones we sacrifice. And my strong intuition is that it is right to take part in an unjust war, if this means a quick end to the war and a better peace than would otherwise be possible. Other people may have intuitions here, which differ from mine.
The problem with both the Sanctity-of-Life Doctrine (deontology) and the (libertarian) moral rights theory is their idea about the role between the moral law and legal rules. It is not that they claim that all sorts of moral wrongdoing must be made illegal, but they are adamant in the insistence that serious moral wrongdoing should not be left as a legally permitted option. In order to defend existing legislation these theories need the assumption that just and unjust combatants are on equal moral terms. But this cannot be shown, on these theoriess. Those who fight in support of an unjust cause, and kill those who defend a just cause, are murderers (according to the Sanctity-of-Life Doctrine) or aggressors (according to the moral rights theory). This means that, on these theories, there cannot be any way of defending the idea that unjust combatants should have a legal right to kill just combatants. But this idea is crucial to the entire system of international law about just and unjust wars.
Utilitarianism seems to get the upper hand in this contest. It doesn’t need the assumption of moral equality. It can allow that legally we condone moral wrongdoing, if the consequences of such legal rules are optimal. So long as there is war, utilitarianism can provide us with a moral rationale for roughly the kind of system of international law that is presently in place. The point in having this system is that it is difficult to ascertain whether a war is just or unjust, and even more difficult to obtain consensus about this. Then it is a good thing if, at least, those who have violated the rules of jus in bello, can (sometimes) be persecuted and punished. Moreover, utilitarianism can allow for the kind of progress we all hope for, of course, with a world where there is no more war, where no rules about just and unjust wars, let alone rules about just and unjust ways of fighting wars, are needed.
But could not deontologists, who defend the Sanctity-of-Life Doctrine, and libertarian moral rights theorists, give up their insistence that murder should (always) be illegal? Could they not — as Jeff McMahan seems to be prepared to do — rely on utilitarianism when deciding upon legislative measures, and rely on their own favoured theory only in private affairs?
This move is not open to a deontologist who accepts Kant’s retributivism, of course. But those who rely merely on the moral prohibition against killing, tempered by the principle of double effect, as well as moral rights theorists, could claim that, while their favoured moral theory instructs us about how to live our private lives, utilitarianism guides choices of legislation.
But this move invites the criticism I have levelled against McMahan: if utilitarianism is fine when it come to legislative matters, why not apply it to problems of private morality as well?