Does nuclear deterrence work?

How long can we rely on an irrational threat?

Welcome back to our Substack Seminar on just-war principles!

If you are just joining us, I’d encourage you to look back on previous episodes, to gain a better sense of what we’re about and to learn the “ground rules” for this discussion. For the full experience, read up from the bottom of the segments listed here. But you don’t need to know everything, and you can always catch up later; we’re happy to have you join us right now!

Once again let me encourage readers (and listeners) to join in our discussion, asking your questions or making your arguments in the Comments section. I know from experience that this topic provokes lively and enlightening debates, and I’m waiting for our little Substack Seminar to reach what you might call a “critical mass.” As I explained a few months back, I’ve set aside the outline that I’d presented for our discussion, in hope of jump-starting those debates.

“War tends toward the utmost exertion of forces,” wrote the great Clausewitz. If you are losing the skirmish, you call in the artillery. If you are losing the battle, you call up the reserves. When you need them, and if you have them, you will always use more powerful weapons.

And yet, in the several decades since the world’s greatest military powers have developed nuclear weapons, they have not used them. (It is significant, perhaps, that the Bomb was used when only one nation had that capacity— before the age of nuclear deterrence.) The world’s two greatest military powers, the US and the Soviet Union, each was accepted a defeat at war— in Vietnam and Afghanistan respectively— without escalating to ensure victory. Have nuclear weapons made that observation by Clausewitz obsolete? Does deterrence work?

In my last two essays in this series I have argued first that no nation can have a rational incentive to use nuclear weapons, and then that no nation can have a moral justification for using them. (In both cases I am referring to the use of nuclear weapons in an all-out war; the limited use of tactical nuclear weapons, in special circumstances, calls for a different sort of analysis.) Even those who disagree with me on either or both of these points will usually agree that the world would be a happier place if the Bomb did not exist. But it does exist, and even if we don’t want to use them— don’t even want to have them— we are left with the hard practical problem of what to do with the nuclear weapons that now exist.

If peaceful and righteous nations disarm, the world will be left vulnerable to nuclear blackmail by belligerent and unjust nuclear powers. Unless we can somehow achieve a universal, complete, verifiable, and enforceable disarmament— a goal that seems helplessly beyond our reach, given both the tensions that prevail in the world and the spread of nuclear technology— how can we live with the weapons we have?

The theory that nuclear weapons could serve purely as a deterrent, without ever being used, is at best an uneasy compromise. To date deterrence has worked— at least insofar as no nuclear exchanges have taken place— but a single failure would be catastrophic, and no one can guarantee against such a failure.

At the height of the Cold War, strategists touted the doctrine of mutually assured destruction, aptly known as MAD. The reasoning behind the doctrine was that no nuclear power would use its weapons in a first-strike attack, because its adversary would answer with its own massive response. Neither side could win, although both sides (and innocent bystanders) could certainly lose.

However MAD was fundamentally unstable; it relied on the assumption that a superpower under attack would launch its own all-out strike, even though it had no rational incentive to do so. Moreover the logic behind the MAD doctrine— the idea that even a great superpower would inevitably be destroyed by its enemy’s counter-strike— was itself unstable. If one power could gain enough first-strike capacity to paralyze the other’s forces, wiping out his capacity to retaliate, then the deterrent would fail. Thus each superpower had an incentive to add constantly to its nuclear arsenal, making it less likely that a pre-emptive strike would be successful. And yet even that sort of response was destabilizing. If country A built hundreds of new missiles, in order to ensure that country B could not wipe out its arsenal with a first strike, then country B would notice that A might be acquiring its own first-strike capacity. So the arms race escalated; in a perverse version of the Clausewitz doctrine, the superpowers used “the utmost exertion of forces” in building up a military capacity they had no wish to use.

MAD was built on the implicit promise by the two superpowers that if they were struck, they would strike back. Now consider: what sort of retaliatory strike would be effective?

The main component of the nuclear deterrents used by the US and USSR during the Cold War were intercontinental missiles, which could be launched from hardened silos. These missiles could be targeted either to destroy the adversary’s missiles, in what would be known as a counter-force strike; or to destroy the enemy’s cities, in a counter-value strike. A counter-value strategy, contemplating the deliberate of killing millions of civilians, would be clearly immoral. But a counter-force strategy, which might theoretically be justified in moral terms, would give the adversary an incentive to strike first, before his missiles could be eliminated. Yet again MAD proved unstable.

One response to this balance of terror that seemed both sensible and morally desirable was the development of anti-missile defense systems. A system designed to defend— to save lives rather than destroy them— should always be preferable, shouldn’t it? But proponents of MAD, especially in the old Soviet regime, argued that anti-missile systems— ABMs— were themselves immoral because once again they were destabilizing, threatening to give one side the ability to withstand an all-out attack. Thus a system that was inherently defensive and morally justifiable was opposed on the grounds that it would interfere with MAD: a doctrine that was unstable and morally bankrupt!

So what is the solution? Early on in the Cold War, the US pledged never to launch a nuclear first strike. (That pledge, naturally, did not satisfy those who doubted the word of US leaders, who claimed that the US was the aggressor in any conflict— and who could point out that the US was the only nation that ever did use the Bomb.) Still, assuming that the pledge was made in good faith, what purpose did the US nuclear arsenal serve? If we would not launch a first strike, the missiles could only be used in retaliation— in other words, when it was too late, when they had failed in their deterrent capacity.

And at that point, what could the US achieve by a nuclear strike? A counter-force attack would serve no purpose, if the Soviet missiles had already been launched. A counter-value attack could kill millions, but would serve no military purpose.

In 1983, when the Catholic bishops of the US issued their letter on nuclear weapons, The Challenge of Peace, they joined the ranks of ethicists and military strategists who had rejected the MAD doctrine as immoral. But their pastoral letter was ambiguous on possession of nuclear weapons as a deterrent. Writing in Commentary in June 1983, Albert Wohlstetter remarked:

Having observed long ago that not even Genghis Khan avoided combatants in order to focus solely on destroying noncombatants, I was grateful, on a first look at this issue in the evolving pastoral letter, to find the bishops on the side of the angels. Unfortunately, a closer reading suggested that they were also on the other side. For, while they sometimes say that we should not threaten to destroy civilians, they say too that we may continue to maintain nuclear weapons—and so implicitly threaten their use as a deterrent— while moving toward permanent verifiable nuclear and general disarmament; yet we may not meanwhile plan to be able to fight a nuclear war even in response to a nuclear attack.

The MAD strategy has now fallen into disrepute, and rightly so. But it has not really been replaced by a workable theory of how we might maintain a nuclear deterrent that provides greater stability, and avoids giving nuclear powers an incentive for immoral actions.

And although I realize that I promised last week that I would wrap up my analysis of nuclear warfare in this essay, I realize now that I will need more time and space. So next week I’ll try to address the most important remaining questions, which include:

• Is it immoral to threaten to do something (such as a counter-value strike) that it would be immoral to do? Is the threat itself immoral, even if it is only implicit?

• What are the real risks of nuclear war, and how can they be eased?

• Do some categories of nuclear weapons have potential uses that could be both militarily advantageous and morally justifiable?

• Nuclear deterrence has “worked,” insofar as it has prevented an all-out world war. But has it given rise to local wars, proxy wars, and other lesser conflicts, in which superpowers encourage wars between less powerful nations?

• A related question: Is a nuclear capacity any help at all in a conflict with a non-nuclear power? Does possession of the Bomb inhibit the nuclear powers themselves?

• Is a “limited” nuclear war possible? If not, can it be made possible?

• And if nuclear war is never the right option, can a nation develop a war-fighting capacity sufficient to deter potential attackers without relying on a nuclear deterrent?