Michael KreponThe Psychology of Deterrence

Henry Kissinger wrote two early, influential books on the Bomb: Nuclear Weapons and Foreign Policy (1957) and The Necessity for Choice (1960). I bought these used books at bargain-basement prices at the Princeton University bookstore. The name on the inside cover suggests that I was the beneficiary of Klaus Knorr’s decision to thin out his library. I was at Princeton on a Council on Foreign Relations fellowship to write Strategic Stalemate: Nuclear Weapons and Arms Control in American Politics (1984).

The Necessity for Choice appeared at a time of great apprehension about Soviet strategic advances and the Eisenhower administration’s seemingly sluggish response. Kissinger, like many others, was under the misapprehension that the missile gap was real and that the Soviet Union was running laps around the United States in the strategic competition. The subsequent discovery that the Missile Gap was imaginary did not calm strategic anxieties because nuclear technologies and delivery vehicles were advancing so rapidly that no one could confidently predict a safe passage in this competition. Kissinger’s concerns resonated greatly at the time. Some have continued relevance. Here are a few passages:

A nation unsure about the circumstances that impair its safety can hardly be expected to address itself with confidence to its positive tasks. It will be torn between complacency and premonitions of catastrophe, between obsession with military security and dismissal of it. When it thinks itself in jeopardy, it will act as if military security were its only problem. When the danger does not materialize immediately it will lapse into euphoria. Its measures are likely to be fitful. Its national mood will alternate between hysteria and smugness…

A gesture intended as a bluff but taken seriously is more useful as a deterrent than a bona fide threat taken as a bluff…

In any given situation a country may be inferior militarily but superior psychologically. It may be able to deter not because it is militarily stronger but because it values an objective highly enough – or can make its opponent believe this – so that it can make plausible a threat to exact a price its opponent is unprepared to pay…

A gap inevitably opens up between deterrence and the strategy we are prepared to implement should deterrence fail…

Since deterrence depends not only on the magnitude but also the credibility of the threat, the side which has a greater reputation for ruthlessness or for a greater willingness to run risks gains a diplomatic advantage…


  1. Jonah Speaks (History)

    “A gesture intended as a bluff but taken seriously is more useful as a deterrent than a bone fide threat taken as a bluff” Here is deterrence and deterrence failure in a nutshell.

    One problem with issuing too many nuclear threats is that the enemy may be tempted to call your bluff. If the enemy calls your nuclear bluff and it was only a bluff, deterrence fails “safe,” albeit with some loss of credibility to one’s other threats. If the enemy calls your bona fide threat, deterrence fails “deadly.” If this is nuclear deterrence we are talking about, the results will be catastrophic.

    “the side which has a greater reputation for ruthlessness or for a greater willingness to run risks gains a diplomatic advantage” Here is a positive motivation for an enemy to test (even though it is risky) one’s threats. Given that the enemy will always be uncertain about whether a given threat is real or fake, there will always be room for a mistaken “test” of one’s genuine threats. Eventually, nuclear deterrence will fail deadly, because a testing mistake against a bona fide threat is inevitable.

  2. P (History)

    > Since deterrence depends not only on the magnitude but also the credibility of the threat, the side which has a greater reputation for ruthlessness or for a greater willingness to run risks gains a diplomatic advantage…

    There’s something here that I’ve never understood… Let’s suppose we have a crisis that doesn’t threaten the survival of a nuclear state/alliance, and a leader irrational enough to make use of nuclear weapons if its own demands are not met. What proof do we have that this leader is nonetheless still rational enough not to use nuclear weapons even if its demands are met?

    My point is that, to make use of nuclear weapons in such a situation, one has to be very irrational. So much that one is at least very close to shooting them anyways and no matter what. So, statistically, the probability that such a leader is still rational enough so that the war can be avoided is small enough to be negligible. (And the more so since people’s rationality evolves over time, and a victory would only embolden him).

    Thus, it seems to me that the only rational reactions to any nuclear threat are to either ignore them, if they’re not deemed to be sufficiently credible, or react to them by shooting first, if they are. Anything in-between, like trying to accomodate the other side by giving in to its demands, looks like an extremely risky gamble – far more risky than a first strike.

    But perhaps I’m the one that’s missing something. Are there any texts about convincing the other side that one’s leader is irrational enough to shoot if its demands are not met, but still rational enough not to shoot if they are met?

    • krepon (History)


      Not sure I’m following your logic train. Suggest we consider a case that’s often considered — a crisis involving North Korea.

      North Korea makes nuclear threats, manufactures crises, and acts provocatively, sometimes by shooting first. This behavior has still not prompted nuclear threats of a first strike by the United States.

      Going back to Kissinger’s quote, it’s hard to see how North Korea’s “ruthlessness” has led to diplomatic gain. It’s led to diplomatic isolation.


    • P (History)

      In my above comment, when I wrote about “shooting”, I meant “shooting with nuclear weapons”. Sorry for being unclear. Of course, North Korea may shoot with non-nuclear weapons, but I see this as an entirely separate issue.

      My point is that, in addition to not leading to any diplomatic gain, all North Koreans *nuclear* threats can be safely ignored by the US. That’s because, if it’s a bluff, there’s obviously no point giving to it, and if it’s not, then they’re so irrational that they’re likely going to launch a nuclear attack even if their demands are honored, so there’s no reason to give to it either; the only difference would be that, if the US was convinced it was in the latter case, a nuclear first strike would make sense.

      Then, following this logic, I’m questionning the wisdom of making any nuclear threat. If the only two rational reactions, for a nuclear-armed country that receives a nuclear threat, are to either ignore it or react to it by performing a nuclear first strike (actually performing, not threatening to perfom), then are we sure making any nuclear threat actually makes sense?

      (Going back to your suggested example about the US not having made any nuclear first strike threats to North Korea, well, the above logic cuts both ways. While it could make sense for the US to *perform* a nuclear first strike, if it were absolutely convinced that the nuclear war is definitely unavoidable, it doesn’t makes any sense to *threaten* to perform one, as it could only tempt the other side to preempt it.)

    • krepon (History)


      I understand better now.

      A couple of reactions: Even tho the United States has not foreclosed the first use of nuclear weapons, this is most unlikely. There was a fair amount of discussion about preemptive N strikes to stop the Soviet Union from gaining the Bomb (and some when China was on the N threshold), but Presidents couldn’t pull this trigger. In the Soviet case, doing so would have been likened to Pearl Harbor, which was a show stopper. Plus, we didn’t know what to aim at, and we probably wouldn’t have been successful.

      Now, in the case of North Korea, we probably have a good idea what to aim at, but after almost 70 years on non-use, the first (or preemptive) use of nuclear weapons seems
      far fetched.

      North Korea’s leaders seem to see value in making nuclear threats. It’s part of their MO, whether to mobilize public support, or perhaps this is a true reflection of their paranoid view, in which case, their N threats “work” — the United States has been “deterred” from attacking since they acquired N weapons.


    • Jonah Speaks (History)

      P’s question raises two issues: deterrence vs. compellence and rationality vs. irrationality. Deterrence can be described as “drawing a line in the sand” and acting only if the adversary crosses it; in contrast, compellence “requires that the punishment be administered until the other acts rather than if he acts” as in deterrence. See, e.g., http://en.wikipedia.org/wiki/Coercive_diplomacy

      A domestic example of compellence is a robber who threatens to kill in order to compel his victim to give him money. Rarely do we suppose that the robber (whether rational or irrational) is likely to kill no matter what, even if he gets the money. A domestic example of deterrence is someone who carries a gun to deter robbers. Deterrence is generally safer than compellence; armed robbery is a poor career choice.

      Krepon’s North Korea example is one of deterrence; North Korea wants to deter invasion or attack. If North Korea tried to use nuclear threats to compel South Korea, China, or Russia to give up territory, that would be an example of a highly risky (and likely unsuccessful) compellent use of nuclear weapons.

    • krepon (History)

      Deterrence sometimes succeeds and sometimes fails.
      Compellence rarely succeeds.

    • Jonah Speaks (History)

      “Compellence rarely succeeds.” This sentiment seems intuitively plausible, but I am not sure it is correct.

      Here are two examples of nuclear compellence that apparently succeeded: The U.S. dropped two bombs on Japan; Japan surrendered. Kennedy insisted that Khrushchev remove nuclear missiles from Cuba, and he did. In both cases, the U.S. had nuclear dominance over its opponent. (I can’t think of any additional attempts at nuclear compellence–if no more exist, 2 successes out of 2 attempts.)

      The more interesting case is whether, starting from nuclear parity or nuclear inferiority, it would ever make sense to use nuclear threats or nuclear explosions to compel an opponent to do something? I don’t know of any historical examples.

      Here are some contemporary situations: China wants Taiwan, but cannot (currently) attain it through conventional war alone. Pakistan wants all of Kashmir, but cannot attain it through conventional war and/or terrorism.

      Both countries have nuclear weapons and could pursue the following limited nuclear war strategy: Drop bombs one at a time until the opponent accedes to its demands. A country adopting this strategy would have to be prepared for similar nuclear treatment by its opponent, until one of them gave in or called a ceasefire. Neither country has attempted, or even threatened, such a limited nuclear compellence strategy. So far as I am aware, neither country has debated even the possibility of adopting such a strategy. Why not?

      One possible reason is the fear that a limited nuclear war may escalate into a full-scale nuclear war. Another possible reason is that even a limited nuclear war would entail a substantial cost, perhaps much greater than the prize being fought over. Hence, for an attempt at nuclear compellence to make sense, the stakes must be really high — much higher than what Taiwan means for China or Kashmir means for Pakistan.

      Possibly, the stakes would be this high for the U.S. during the Cold War, if the Soviet Union had invaded Western Europe and the U.S. wanted to compel a withdrawal. Even so, many nuclear theorists would have doubted the credibility of such a nuclear threat, after the invasion were consolidated, but seemed willing to support nuclear use while an invasion was in progress and nuclear threats beforehand as an “extended deterrent” to invasion.

  3. Bradley Laing (History)

    ISLAMABAD — Media reports here have outlined that Pakistan is set to increase funding for the armed forces and the national nuclear body, the Pakistan Atomic Energy Commission (PAEC), under the forthcoming 2014-FY2015 budget.

    The budget would be just over US $81 million for the PAEC, up from nearly $63 million the previous year (which was later increased to $66 million).

    Mansoor Ahmed, from Quaid-e-Azam University’s Department of Defence and Strategic Studies, who specializes in Pakistan’s national deterrent and delivery programs, said although the figures earmarked for the national nuclear body are mainly for a civilian power generation project, there are national security implications.

    “This sum is primarily geared toward the construction of the two 1,000-megawatt generation-III safeguarded Chinese nuclear power reactors to be established at Karachi, K-1 and K-2, that were recently initiated by the prime minister,” he said.

    However, he added, “Additional financial allocations are most likely earmarked for the unsafeguarded Khushab Nuclear Complex where the fourth plutonium production heavy water reactor is reportedly nearing completion.”


  4. Bradley Laing (History)

    COLORADO SPRINGS: Australia, Britain, Canada and United States have signed a symbolically important Memorandum of Understanding committing them to “a partnership on combined space operations.”

    …Brian Weeden, a space expert with the Secure World Foundation who alerted reporters to the UK announcement, once served at U.S. Strategic Command’s Joint Space Operations Center (JSpOC). He noted that the statement refers to “Combined Space Operations” which “doesn’t necessarily involve everyone being physically co-located in the same place. It is more likely that each partner will have their own national space ops center and some level of coordination/communication between them.

    Weeden thinks that this agreement may have arisen from one of the Schriever war-games, this one held at Nellis Air Force Base in 2010.