Another Wednesday, another Ward Wilson entry.  This one tackles two of my favorite cases: the  Yom Kippur and Falklands Islands  Wars.

In case you are interesting, Parts 1 and 2 are also available.

Doubts about Nuclear Deterrence, Part III: Yom Kippur and Falkland Islands

by Ward Wilson

In 1945 there was nothing that nuclear weapons couldn’t do. Secretary of State James Byrnes told a friend that nuclear weapons would assure success in negotiations. Others said that a nuclear arsenal would make a country entirely safe. Who would launch an attack (conventional or nuclear) against a country armed with these fearsome weapons? And it didn’t take long before the United States was trying to think of ways to use these multifaceted weapons to protect its friends and allies. Eventually they developed “extended deterrence”–protecting your friends with a nuclear umbrella. Four remarkable capabilities: provide diplomatic power, prevent conventional attacks, prevent nuclear attacks, and protect our friends.

But these early, sky-high expectations have not been fulfilled. The history of nuclear deterrence since 1945 has been a steady, step-by-step retreat. The circle of nuclear deterrence capabilities keeps getting smaller. It’s discouraging.

First came the post-World War II negotiations over the shape of Europe. Byrnes left for Europe confident. He came back chastened. The Russians, he reported, were “stubborn, obstinate, and they don’t scare.” Subsequent negotiations (like those between the United States and North Vietnam) ratified Byrnes’s judgment: you can’t always rely on nuclear weapons in diplomatic negotiations. So the first of the four capabilities of nuclear weapons was disproved.

But nuclear proponents were not unduly discouraged. They said, “You can’t use a weapon as horrible as nuclear weapons as a threat. It’s a weapon of last resort. You can’t threaten to blow some country off the face of the map if they don’t give you mining concessions or whatever. It’s just not credible.” And they had something of a point. It is hard to threaten something so out of proportion and be believed.

The notion that nuclear weapons prevented countries from launching wars against nuclear-armed countries, though, remained intact. The reason is that war is a fundamentally incalculable event. You don’t need Clausewitz to tell you that war is unpredictable, it’s end is enveloped in impenetrable fog, and that of all human events, war is the one most likely to swirl out of control. Whenever you engage in war with a nuclear-armed country, you risk a spiral ending in nuclear attack. Wars just get out of control that way. People get angry, they feel a desire for revenge, they get swept up in blood lust or some idiot doesn’t get the word. Going to war with a nation that has nuclear weapons is extraordinarily risky. Even if the object of the war is limited and you imagine that the war will be short. (People always imagine that wars will be short, by the way. Geoffrey Blainey argues that the main cause of wars is too much optimism.) Anyone with the slightest knowledge of history would know that engaging in a war with an adversary armed with nuclear weapons is no joke. So nuclear deterrence should prevent conventional war with ease.

But it didn’t. Twice. And these two events stand as a stark challenge to the assertion that nuclear deterrence is a powerful and reliable capability when it comes to conventional war.

Middle East War 1973

When nuclear proponents talk about the 1973 Middle East War they mostly talk about the deterrence “success.” In the last days of the war, Henry Kissinger ordered U.S. nuclear forces on alert worldwide. The move was intended to send a signal to the Soviets not to send paratroopers to reinforce Egypt (which they were planning to do.) And, according to nuclear deterrence proponents, it worked. The Soviets did not, in fact, air lift paratroopers to Egypt and some people claim it was the nuclear threat that stopped them.

But that doesn’t face up to the real problem. The real problem is what were Sadat and Assad thinking? The leaders of Egypt and Syria must have known that Israel had nuclear weapons. It had been reported in the New York Times. They must have known that Israel is a pretty small place and any attack that breaks through has the potential to be in Tel Aviv the next day. We’re not talking about invading Russia here. In Russia you can break through and drive for weeks, as the Germans found out, and still not even be close to Moscow. Any attack on Israel can quickly because an existential attack. How is it that Sadat and Assad were not deterred? How could they launch a shooting war, in which people would bleed and die, and not fear that it would spiral out of control?

Proponents say that Egypt and Syria knew that the war would be limited. They were, after all, only attacking the occupied territories. They weren’t attacking Israel proper. And such a strictly limited war would not cause Israel to reach for the nuclear option. They could reasonably assume, proponents claim, that Israel would not use nuclear weapons in a limited war.

I have real doubts about the notion that it’s rational to assume that a war can be kept limited. That’s not the way I read the history of war. That’s not why Clausewitz asserts that war tends to violent extremes. But let’s leave the proponents with the last word for the moment.

Falkland Islands War

The second fly in the ointment is the Falkland Islands War. Again, it’s curious that most of the discussion you read about this war has to do with Margaret Thatcher telling the French that if they didn’t give up the codes to the Exocet missiles that Argentina was using to sink British ships, she’d have to use nuclear weapons against Argentina. Proponents smile and shake their heads (as if to say, “That Maggie. She’s a tough one.”) They imply that the danger of nuclear war was crucial to convincing the French into surrendering the codes. And it is true that they gave up the codes.

What they don’t talk about is what the leaders of the Argentine junta were thinking. The British had nuclear weapons. Argentina did not. How is it possible that you could get into a shooting war with a nuclear-armed nation and not worry about nuclear attack? Even if you weren’t worried that the Brits would attack you out of hand, you ought to have at least been concerned that the war might take some unexpected turn and escalate to nuclear conflict. For example, you sink a bunch of British ships, the Brits get furious, they launch a bombing raid against one of your cities, you sneak some ships off the British coast and shell Plymouth, and then things get out of hand. If the fear of nuclear war is so powerful, why weren’t the Argentines deterred?

Some proponents tut-tut and say (condescendingly, usually) that this is all completely understandable. Of course nuclear deterrence didn’t prevent the Falkland Wars or the Middle East War. Only a naive person would expect it to. (In fact, a fellow made this very point in a recent off-the-record briefing I was giving. Although more politely.) Proponents of nuclear weapons say, “Nuclear weapons are weapons of last resort, and as such they only deter attacks on vital interests. Attacks on far flung islands or occupied territory simply don’t count. If the Argentines had attacked London, now that would have been a different matter.” They argue what I would call the “vital interests” exception to nuclear deterrence.

There is a certain logical consistency to this. It could make sense that because nuclear weapons are so horrible, they can only be used in really dire circumstances. But if that’s so, if you want to argue that nuclear weapons can only deter attacks against vital interests, then you have to throw over NATO and much of current U.S. foreign policy. Because much of that policy is intimately linked to extended deterrence–the idea that you can extend nuclear protection over distant friends and allies. The United States uses nuclear guarantees to defend Europe, Japan, South Korea and other important places. Unfortunately, it isn’t possible to make a case for extended deterrence if you argue the vital interests argument. If you can’t extend deterrence over far flung islands that are part of your own country, how are you going to extend it over, say, Japan? If you can’t extend deterrence over territories that your troops are occupying and your civil authorities are administering, how are you going to make people believe that you can extend it over Germany?

By arguing the vital interests exception for nuclear deterrence, you can save deterrence from the contradicting cases of the 1973 Middle East War and the Falkland Islands War. But you do have to surrender 60 years of U.S. foreign policy. And you should probably spend some time trying to figure out what other magic glue you are going to use to bind its allies to the United States. So your choice: explain away the cases that contradict reliable deterrence of conventional war and give up extended deterrence, or keep extended deterrence but have to live with two irreconcilable contradictions in the heart of the “deters conventional war” evidence.


What do these potential failures of nuclear deterrence prove? What is there significance? It seems to me that the real problem here is that no one is debating these questions. Check the literature on the Middle East war and you’ll find that even critics of nuclear weapons spend most of their time talking about the Kissinger threat and almost none on what Sadat and Assad could possibly have been thinking. It’s not entirely ignored. James Acton found a book about this by Jewish scholars (in Hebrew). But it’s not a big topic in the debate. Given the dangers involved with nuclear deterrence (in case you’ve forgotten “the dangers involved with nuclear deterrence” include catastrophic nuclear war), wouldn’t you expect that these sorts of things were being rigorously gone over and double checked? Should all possible failures of nuclear deterrence be completely understood and explained? Blithely relying on nuclear deterrence without digging into and examining the doubtful cases doesn’t seem exactly prudent to me.

Of course, there are those who argue that although the three other kinds of deterrence have proved doubtful (diplomatic power, preventing conventional war, extended deterrence) that the most important kind of deterrence–preventing nuclear attacks–is still reliable. There has, after all, never been a nuclear war. And maybe they’re right. But the circle has been contracting for sixty years. Proponents claim that although they were wrong in the past, although the area of nuclear deterrence’s influence has continually shrunk, that now the shrinking has finally stopped. Even though those other uses for nuclear deterrence turned out not to be so effective, this last use of nuclear deterrence will work for sure. They say, in effect, “This time for sure.” Somehow I’m not reassured.