Herman Kahn (1922-1983) was, and remains, one of the most misunderstood figures of the atomic age. But he was also one of the most influential, and any serious student of nuclear war needs to grapple with Kahn and his writings.
I'll admit it: I'm a big Kahn fan. Now, this isn't to say that I think Kahn was infallible. In fact, I think that his entire paradigm for thinking about nuclear war had severe limitations. But compared to those who came before him and most of those who came afterwards, his work was far more complete and consistent. At the same time, Kahn was an abysmal writer, which is a major reason why he is so little understood today. I think that On Thermonuclear War is much like Marx's Das Kapital- almost unreadable, often simply wrong, yet ingenious and vitally important both intellectually and historically.
In order to understand why On Thermonuclear War is such a landmark, one has to read the works that it supplanted- for instance, Bernard Brodie's The Absolute Weapon and Henry Kissinger's Thermonuclear Weapons and Foreign Policy. The Absolute Weapon, from 1946, was the original book on nuclear strategy. In it Brodie argued that the atomic bomb was the "absolute weapon" because once a nation possessed enough to reliably destroy the enemy's population centers there was no reason to acquire any more. Brodie failed to take into account the possibility of counterforce targeting, which was the centerpiece of Kahn's strategic thinking. Kissinger's book is a lot more interesting, and disturbing. Kissinger argued that the superpowers would soon begin utilizing tactical nuclear weapons in limited wars- essentially that they would agree not to respond to a few nuclear strikes in a place like Indochina with an all-out nuclear war. Both these works were later repudiated by their authors.
On Thermonuclear War is, first and foremost, an attack on the nuclear policies of the Eisenhower Administration. He charged that the then-current policy of "Massive Retaliation" was crude, destabilizing, unnecessarily brutal, and that if things continued on their then-current course the result would be "a catastrophe that would be much worse than it need be." Personally, I am convinced that, whatever the merits of Kahn's strategic thinking overall, he was right about this critical point. The misguided policies of the Eisenhower years appear to have led directly to the nuclear crises of 1961-2, which we now know came frighteningly close to devolving into the nightmare Kahn was trying to prevent.
Most readers of On Thermonuclear War are distracted by Kahn's lengthy arguments that even a large thermonuclear exchange is survivable, given a great enough investment in civil defense. This is what first led me to Kahn- he was probably the single most important civil defense advocate in American history (although JFK has a serious claim to this title too.) It should be kept in mind that Kahn made this argument, in large part, because he was the world's greatest optimist. He genuinely believed that mankind could overcome even the greatest tribulations. Some of the details of Kahn's arguments about nuclear survival have aged well- for instance, the section on hereditary mutations looks pretty good in light of studies of atomic bomb survivors- but other parts of it have not. However, this does not undermine Kahn's basic strategic logic. It merely changes the shape of the concrete war plans that follow from them.
Kahn essentially argues that there are two kinds of nuclear wars: survivable and unsurvivable. He uses idealized thought experiments to argue that trying to deter enemies using preparations for the latter kind of war is a VERY BAD IDEA. The best explanation of this is in Kahn's discussion of doomsday machines, which he uses to illustrate this point. Although a doomsday machine offers a seemingly ideal "Type I" deterrence, the danger it poses is so great that it is ultimately far more trouble than it's worth.
According to Kahn, the ideal deterrent has six qualities:
5. Nonaccident prone
Kahn concedes that the doomsday machine is a clear winner on the first five of these points. Indeed, he says that "it maximizes the probability that deterrence will work." But ultimately this is a fool's bargain, because "one must still examine the consequences of a failure. In this case, the failure kills too many people and kills them automatically."
Kahn transferred this logic from doomsday machines to more conventional strategies that resemble doomsday machines--in particular, to what would later become known as Mutually Assured Destruction. Kahn did NOT claim that this offered inferior deterrence to his proposed counterforce strategies. But he thought that it didn't offer ironclad deterrence--and given the myriad real-world reasons deterrence could fail, the likely cost was far too high to merit the risk.
Therefore, deterrence needed to be based on the principle that nuclear wars should be fought to be won. The fact that some of the wars Kahn thought were winnable probably aren't does not actually invalidate this argument. It just changes the nature of the deterrent force. Arsenals of thousands of multimegaton hydrogen bombs are right out; instead, a more limited arsenal targeted primarily at enemy strategic forces makes the most sense. Indeed, something vaguely approximating this was adopted by the two superpowers, and the world has been considerably safer ever since. Because of this, I am of the opinion that Kahn's influence on human history has been overwhelmingly positive.
It's also important to remember that Kahn's 1960s strategic prescriptions were meant to be temporary. He thought that the bipolar arms race would be long over by the year 2000, but he feared that this would be the result of the development of a multipolar world with dozens of nuclear powers. He thought that preventing the use of nuclear weapons would be impossible under these circumstances. In any case, he did not intend for the deterrence strategies he championed to be made into a permanent fixture of international relations- they were intended to protect US interests for the next decade or two. The bizarre scenario that actually occurred- where the USSR collapsed, the United States achieved global near-hegemony, and yet the arms race and deterrence somehow failed to disappear- did not occur to him fifty years ago.
So what's wrong with Kahn? Well, he bought too much into contemporary pessimism about the possibility of negotiating meaningfully with the Soviets. (To his great credit, though, he emphasized that the USSR was a status-quo power. Back in the 1950s, that put him well ahead of much of the defense establishment.) He was also too dismissive of the positive roles culture plays in the function of nuclear deterrence. In Kahn's writing, culture is just a mental block against thinking seriously about nuclear warfare; but in the real world, it can keep decision-makers from escalating a conflict even when they logically should. Current scholarship on nuclear strategy is moving away from the realist explanations of behavior and motivation championed by Kahn to more nuanced explanations, and scholars like Graham Allison have made very convincing arguments that rational actor theory does not account for the behavior of the United States and Soviet Union in actual nuclear crises. This effectively upends Kahn's entire paradigm for understanding nuclear strategy, but it also arguably undermines deterrence theory in general.
If one wants to read Kahn, I recommend his 1962 book Thinking About the Unthinkable rather than On Thermonuclear War. Intended to be more readable, Kahn is also less extreme in this volume. I particularly like its chapter on alternative strategies for the Cold War, which has fourteen options ranging from surrender to preemption. For those who only know Kahn from the crude caricatures drawn by his critics, the conclusions of Thinking About the Unthinkable will probably come as something of a shock- he endorses world government as the ultimate solution to (or perhaps outcome of) the geopolitical problems posed by nuclear weapons. His writing as a futurist is also worth looking into. A full-blown techno-optimist, Kahn retired from nuclear strategy in the 1970s to debunk the Club of Rome, Paul Ehrlich, and the other then-trendy eco-doomsters. Before he died in 1983 he lamented the crippling of nuclear power in the energy market. Indeed, at the time of his death he was better known as a futurist than as a strategist. But when either contemplating the possibility of nuclear annihilation or the potential of civilian nuclear energy, Kahn was forever the ebullient optimist.