1. Conway is also focused on the deterrence account of c.p.  Unlike in Aquinas, the concern in the Conway piece is with protecting innocent lives, not producing a virtuous community.  Thus, the Conway piece only deals with the case of murder.  Conway is responding to van den Haag’s arguments.

• The central question for Conway is whether deterrence works.  Conway argues that we have no evidence that it works.
• The preference argument: Criminals prefer life imprisonment (l.i.) to death.  Hence, death deters more.
• But, maybe, many murderers are not sufficiently rational to care, and those who are sufficiently rational would be deterred by l.i.
• Conway: Moreover, there may be a threshold beyond which there is no further deterrence.  Suppose Country A decrees that for murder you get 23 years in jail and Country B that you get 24 years in jail.  Any criminal will prefer 23 years to 24.  But would it make any difference with respect to deterrence?  One doubts it.
• A further argument not in Conway’s piece.  Many murders are committed by young people who do not think of their own death—who act as if they were immortal.  Such a person might be more deterred by l.i. than by c.p.
• Or consider the fact that if a doctor told us that our arms are gangrenous and the only way to save our lives was by having them both amputated.  Most of us would go for it, I think.  But it does not follow that amputation of both arms would be less deterrent.  In fact, one imagines that the horror of it might deter many people much more than death, because it is a much more down-to-earth horror, something more imaginable than death.  Likewise, l.i. might be more imaginable than death, and hence more horrible.

2. One of the major moral views nowadays is virtue ethics.  This is inspired by Aristotle’s insistence—we can see a similar thing in Socrates—that human flourishing is the constant exercise of virtue.  The purpose of life according to virtue ethics is to become virtuous human beings and to live virtuously.  Aquinas’ apparent focus on the human community as a community of virtuous human beings—as something endangered by vice and crime—marks him as at least sympathetic to this.

• But now note that the above considerations have all neglected one class of people.  We looked at criminals and their potential victims.  But there is a third relevant class of people: executioners (both the people who actually pull triggers, flip switches, etc., and the judges who pass death sentences).  These people have killing fellow human beings as their job.
• Moreover, it is not a matter of killing human beings who are actively threatening innocent people.  Rather, the executioners kill helpless human beings. 
• The concern now is two-fold.  (a) Could a virtuous person do this?  (b) Would doing this help tend to make and keep one a virtuous person?
• If c.p. is in itself morally acceptable for the reasons that Aquinas or van den Haag give, one could probably imagine a virtuous person performing it, a person whose will is focused on the good of the community, etc.  So the answer to (a) might be positive if either of these thinkers is right, and if so, then virtue ethics could allow c.p.
• But there is a further question: Would such action tend to help to promote virtue in the executioner?  If on the contrary it would cause harm, then a focus on the promotion of human virtue might make c.p. too risky because of the moral harm to executioners.
• What form might this harm take?
• To make one’s job easier, in killing someone, one is likely not to think of him as a human being.  The kind of brutal treatment that the murderer sentenced to death gets in Decalogue 5 illustrates this.  But a habit of not thinking of human beings as human beings is a very unhealthy one.
• Such killing of innocent people, in cold blood, without any urgency, might help deaden a feeling that we are all fellow humans, that we are all brothers and sisters.  Such fellow-feeling is very important: Empirical data suggests that it is precisely this kind of feeling that led to people rescuing and protecting Jews from the Holocaust.
• Furthermore, if the justification is deterrence, then the killing is done in order to prevent other people from committing murder, and so the criminal is used as an object, is used as an example (pour l’encourager les autres).
• This doesn’t mean that c.p. is always wrong.  In a situation where it is clear that people’s lives are being protected—e.g., when executing a homicidal escape artist who has escaped from jail dozens of times in a country where the lock technology is very poor—one can more easily focus on the lives of the people one is protecting, and in doing so, maintain a healthy motivation, one that recognizes that the criminal is indeed a fellow human being, but one who endangers others.

3. But there another kind of justification of c.p., in terms of what people deserve.

• If I have done you something good, you owe me, at least, gratitude.  The amount of gratitude you owe me depends on what I have done for you.  People who have done good things deserve rewards, whether praise, or medals, or something else.  If when I was in need you helped me, then there would be something wrong with me if I did not help you in the future when I was no longer in need but you were in need.
• Moreover, my rewarding your good deeds shouldn’t merely be there in order to encourage you to do similar things later.  I should reward your good deeds even if you are dying and my rewarding you will not lead you to any more of them.
• Nor should this reward be there merely to encourage others to imitate your example.  If that is why it is there, then somehow you fall off my radar screen: it is not that my gratitude to you moves me to reward you, but it is considerations having nothing to do with you and with what you deserve.  You become just a means, a tool of propaganda.  In fact, if the point of the reward is to encourage others, then it follows, absurdly, that I should reward you if people think you did something virtuous even if I happen to know you didn’t.  (E.g., people think you saved a drowning girl, but I know that I did it, but I don’t want to brag about it and to call you a liar.)
• But there is a flip side of what we deserve.  It seems quite natural to suppose that if your good deeds deserve gratitude and reward, then your bad deeds deserve a negative attitude on the part of others, and even punishment.  Moreover, just as reward is to be apportioned to virtue, so too punishment is to be apportioned to vice.  The worse your deeds, the worse your punishment should be.
• Just as I shouldn’t reward you simply to encourage you and others to act similarly, neither should I punish you simply to discourage you and others from acting similarly.  If that is why I punish you, I am using you as a tool of propaganda—to yourself or to others.
• Rather, it seems, my reason for punishing you should be that you deserve the punishment.  Just as I owe rewards to you for your good deeds, out of a recognition that you did them as a responsible person, so too I owe punishments to you for your ill deeds. 
• Intuitively, if you risk your life to save mine, I owe it to you to risk mine to save yours, or, if I do not actually owe it, I have very good reason to risk mine to save yours—everyone would say that the virtuous person would do that.  Well, it seems that by the same token, if you have taken the life of a member of society, society owes you death.  That is what you deserve.
• This is a retributive justification of c.p.: the criminal deserves punishment, and that is the primary reason why she is being punished.  The magnitude of it is set via the lex talionis, the “law of retribution”: an eye for an eye, a tooth for a tooth, a life for a life (Bible and other ancient near east texts).
• If the punishment is seen as something owed to the criminal, as something done to recognize the criminal as a person who has made choices that deserve consequences, then it seems the punishment could in theory be inflicted in a way that respects the criminal’s personal dignity. 
• The quick and private c.p. done in the U.S., without an audience, without much ceremony does not, however, recognize the criminal’s personal dignity as much.  A quick lethal injection suggests more “getting rid of the criminal” than affirming the just deserts of her actions.
• A pure deterrence view seems to say that we should punish people when society would benefit from it.  But suppose that it would benefit society a lot to get rid of pickpockets, that it would benefit society enough to make it worth killing pickpockets.  Still, this surely does not make killing pickpockets right.  The retributive view sets a limit on punishment: no more than an eye for an eye, a tooth for a tooth, a life for a life.
• A consequence of this account of punishment is that it suggests that it can be acceptable to punish criminals who have changed.  For it is acceptable to reward someone for doing X even if she later repudiated this action (e.g., she saved a policeman’s life, but later joined a radical political group which says that policemen are pigs whose lives should not be killed;  yet, we can reward her for having saved that life).

4. Objections

• Does society have the right to impose such penalties given that society itself is guilty of many things, such as neglect of the poor?  (Cf. Jesus saying that he without sin should take up the first stone.)  This also suggests another danger to the moral life of the executioners: it might make them self-righteous.
• Does this mean we should rape rapists and torture torturers?
• Response: The punishment is supposed to reflect the dignity of the criminal, as someone whose choices matter.  Raping people dehumanizes them, and hence does not reflect their dignity.  Perhaps torture does the same thing.  So rape and torture are not owed to the criminal, for we owe to no one to dehumanize them.  It is only a dignified killing that we can owe.  Moreover, as a matter of fact, rape and torture damage the character of the people who do it, and hence do damage to the human dignity of the executioner.
• Surely we can show mercy.  But according to the retributive view we owe punishment to the criminals.
• Maybe we do not strictly owe it.  After all, if you save your life at a risk of yours, maybe it is not immoral for me to not save yours at the risk of mine.  After all, it isn’t as if we entered into a mutual life-saving contract by your saving my life.  Your saving my life was a gift to me.  A real gift does not absolutely require a return.  But it does give a good reason for a return, even if this reason is not conclusive.  Thus, by analogy, maybe we have good retributive reason to punish criminals, but still there may be reasons against it.  Clemency could, especially if the criminal has changed her life, give one such a reason.
• But this last consideration suggests that perhaps even though retribution gives a strong reason to kill murderers, other reasons might override that.  For instance, the value of the murderer’s life might override this.  This kind of a thing might lead to a hybrid account on which both retribution and deterrence would be needed to justify c.p.  Here’s one such account.  Retribution gives us a good reason to kill criminals.  But the value of the criminal’s life and the dangers to the character of the executioner give good reason not to kill criminals, indeed reasons that if they were all there is would override the retributive reason.  But the value of deterrence then overrides these last considerations.
• That is, if the deterrence account works, if no lesser penalty could do the job.  But I think a lesser penalty could do the job in our society.  (Maybe not in a poorer society which cannot afford to keep people in jail.)  Hence, on my view, the hybrid account would not permit c.p. in our society: retribution would be served by c.p., but the disvalue of death would give one reason not to engage in retribution.