The Paradox of Deterrence
Consider the following questions:
- Is it morally OK to kill someone in order to save someone else?
- Is it morally OK to kill one person in order to save 10 people?
- Is it morally OK to kill one person in order to save 100 people?
- Is it morally OK for you to kill 10 million innocent people in another country in order to save 10 million innocent people in your own country?
I won't try to answer these questions as they are part of the introduction to the main topic which I'll outline shortly. Consider now the following situation. You are the President of a nuclear power which is threatened by an enemy country. Your enemy is about to mount a nuclear attack which, we suppose, will kill 10 million of your citizens. Your only way of preventing this attack is for you to attack first, killing 10 million citizens of your enemy country. So consider the following question:
Again, I won't try to answer this question but it is relevant to what is called the “paradox of deterrence”. This paradox arises in situations like the one above. Your country is facing an aggressor A. A might be about to invade, attack you with nuclear weapons, subject you to nuclear blackmail or carry out some other form of aggression. Suppose, now, that the only reliable way you have of deterring A is by threatening nuclear retaliation. And suppose further that bluffing won't work – if your country is a democracy then it would be very difficult to successfully bluff as your actions would be under scrutiny and it would be hard to maintain the secrecy required for bluffing. Thus, the only way of successfully deterring A would be by threatening and fully intending to retaliate if attacked. The problem is that retaliation is morally unjustifiable as it is obviously morally wrong to kill 10 million innocent people.
Furthermore, suppose that your threat is unsuccessful and A attacks killing 10 million people. Then, firstly, there is no point in your retaliating because it won't save your own people, so retaliation is pointless and secondly, it is immoral because it is wrong to kill 10 million innocent people. Retaliation is therefore both pointless and immoral. Therefore, as a moral agent you can neither retaliate nor even intend to retaliate. The reason that you can't even intend to retaliate is that you can't intend to do something that you will not do. This would be similar to believing something that you know is false.
In short the paradox is this. If a threat of retaliation is all it takes to prevent a catastrophic attack, then the threat would seem to be morally justifiable as the making of a threat is clearly preferable to the huge loss of innocent human life which would result from the attack. However, only a serious threat of retaliation can be successful and a moral agent cannot seriously threaten to retaliate. What follows is that, in the sort of situation described above, a rational and moral agent cannot adopt a policy of deterrence.
Note that the morality of retaliation is not like question 4 above. Question 4 is a more difficult question (compare question 3). The morality of retaliation in the paradox situation is much more clear-cut. If 10 million innocent people have already been killed then the killing of a further 10 million people achieves nothing. The only point would be revenge and this has no moral basis – “Two wrongs don't make a right”.
Some Questions- The paradox arises only for a moral agent. Might it be better in this situation to abandon morality and adopt a policy of deterrence on pragmatic grounds? i.e. threatening nuclear retaliation may not be a moral policy but if it works in avoiding aggression, then why not adopt it? Can morality be abandoned so easily? Should it be? If it is abandoned at this point can it also be abandoned elsewhere when convenient?
- If A knows that I am a rational agent then A knows that I will not retaliate. And if so, A has a free hand to act at will.