Jeffrey LewisPlague and Cholera

Julian Borger has a post up on the Nuclear Posture Review, in which I liken choosing among the two options on declaratory policy to the choice between plague and cholera.

Chris Jones over at PONI literally doesn’t understand the argument, which leads me to think that if a smart guy like Chris can be so confused, then I should explain more.

My preferred option for declaratory policy is to state the purpose for possessing nuclear weapons, rather than speculating on when a future President might use them:

The United States maintains nuclear weapons to deter and, if necessary, respond to nuclear attacks against ourselves, our forces, or our friends and allies.

My reason for phrasing it this way, as Josh Pollack captured, is to avoid the “what-if questions meant to exhume sinister contradictions” in our nuclear policies. That’s much better than I said it.

The authors of the NPR appear to have internalized this message — to talk about the purpose, rather than speculate on use — but the two options for the final document, as described by Borger and Josh Rogin, are very unsatisfying to me.

The first choice is to state that the the primary purpose is deterrence. This, in my opinion, merely draws attention to any secondary purposes, which is precisely the conversation to be avoided. It’s better not to have a declaratory policy that raises an obvious question unless you know the answer to that question in advance. And it seems to me that no good can come of answering what “secondary” purposes might exist. The fact is that decisions about the size, composition and posture of our nuclear forces are all made in the service of deterrence. Anything else is a lesser included case that is not a fit subject for discussion in polite company.

The second choice is to say that our goal is for some future President to someday be able state that the “sole purpose” is deterrence. (For the record, I am not wedded to the adjective “sole.”) To articulate the preferred outcome as a goal, rather than a description of current policy, is is almost, though not quite, as bad.

This is a simple question: Why do we have these awful things? Setting sole purpose as a “goal” leaves this question unanswered. We know what the purpose is not (solely deterrence), but not what is. We are left to guess at which purposes might prevent the Obama Administration from answering this simple question forthrightly. Rather than lamely admitting that the posture (and the posture review) is in some sense a disappointment, one might as well defend the role of extending nuclear deterrence to conventional attacks against allies.

A second downside of admitting that the reality of a nuclear policy falls short of our ideal is the degree to which it implicitly undermines the goal set in Prague. The President committed the United States to seek the “peace and security of a world without nuclear weapons.” The logical corollary of that statement, is that all other threats — including those to our allies — could be met with what Bundy, Crowe and Drell called “prudent conventional readiness.” The argument, especially in Paris, has been that it is easy for the United States to seek a world without nuclear weapons given our vast conventional military power. If the United States today doesn’t have enough conventional capabilities to relegate nuclear weapons to the task of deterring nuclear attacks, no country ever will.

Still, it must be admitted that “sole purpose” is an admirable goal for US nuclear weapons policy, even if admitting that the reality of US nuclear policy falls short of it makes the Prague Speech look a little naive.

Yet, I do not understand why we can’t simply state that the purpose of the weapons narrowly, while declining to speculate on their use. After all, as a practical matter, the United States maintains nuclear weapons today for purpose of deterring, and if necessary responding to, nuclear attacks against the United States and our allies. Any other scenario is, at best, a lesser-included case.

The President should just say so.

All this is terribly disappointing, but fortunately it is not the sort of disappointment that can’t be overcome with a stiff Manhattan at the University Club with an old friend. I guess in that way it really isn’t like plague or cholera. Until tomorrow …


  1. anon (History)

    Your formula works if you are trying to use the declaratory policy as a guide post in sizing and structuring the nuclear force. If you are sized and structured to deter, and, if necessary, respond to nuclear attacks, then you have enough size and the right structure to do anything else you might want to do (that’s the meaning of a lesser included case, you don’t rule it in or out, but you can do it if and when you decide you need to do it.)

    But your formula does not work if you are trying to demonstrate that you have reduced the role of nuclear weapons, or, in other words, reduced the number of circumstances when you might consider using nuclear weapons. If you don’t rule them out, you allow for these circumstances as “lesser included cases.”

    Personally, I’m fine with this formula. I don’t think we need to rule in or rule out other cases, I like the ambiguity for deterrence. But your formula does not get to the fundamental question that many people want an answer to: Would we use nuclear weapons to deter or respond to a CW or BW attack? Or a conventional attack.

    Or have I misunderstood you, too?

  2. Brent Logan (History)

    I’m certainly no expert in arms control, but how does the mere existence of nuclear arms provide any deterrent effect without those being deterred believing they might be used? Asked another way, wouldn’t their deterrent effect be lost if the president were to say that the nuclear arms would never be used under any circumstances?

  3. kme

    Brent Logan: Even under your last hypothetical, the deterrent value would not be completely lost. The potential aggressor must factor in to his calculations the probability of the President changing his mind on the matter when actually pressed by an attack.

    This obvious fact (that words are not deeds) is why we have mutual verification in arms control treaties, after all.

  4. Doug (History)

    how about a nuclear posture mulligan? as I recall you observing, the prospective enthusiasm about the potential outcome of the NPR process consistently overlooks the NPR’s structural constraints. Could the NPR begin a wider discussion?

  5. Jeffrey Lewis (History)


    I think we agree.

    I believe it is a mistake to answer the question of whether we would (or would not) use nuclear weapons to respond to a conventional, CW or BW attack. I want to leave existential deterrence in place. (I happen to think it is a function of the existence of the weapons, but that is another conversation.)

    Still, “reserving the right” is conflated with “intends to, maybe tomorrow” — an equally if not more problematic formulation. It does violence to conventional deterrence, creates perverse incentives and is, I suspect, a bluff. (I object to nuclear bluffing.) And — for me this is the key — routine U.S. nuclear weapons decisions are interpreted in the US and elsewhere in light of these peculiar lesser included cases, undermining the role that the weapons do play.

    The solution, it seems to me, is to observe that there are no guarantees in life, but that lesser-included (or I would say incidental) cases should not be confused with the purpose of the forces.

    Does that demonstrate a reduced role for nuclear weapons? No, but I don’t care for that formulation — Arnie Kanter convinced me that it is logically backward.

    The role for nuclear weapons is declining for external reasons (no ideological rivalry, better conventional forces, etc.). Our declaratory policy recognizes the new environment, it doesn’t create it.

    On the other hand, this policy does a good bit to correct the record. Domestic and foreign perceptions of US nuclear weapons policy are, in my opinion, somewhat distorted.

    Stating the purpose of the forces doesn’t, strictly speaking, reduce their role — but the audience comes away understanding that the role is less than they thought.

    Were I advising the President, which of course I am not, I would have encouraged him to simply assert that these weapons play a much smaller role than ever before and that our task is to make sure our forces, policies and posture reflect that reduced role.

    And that I file this one under “policies.”

  6. anon (History)

    I agree with you. I, personally believe that, with declaratory policy, the less said the better. But the pressure is on for the Admin to be definite about whether it intends to use nukes to deter CW and BW. Not that they’d listen to me, but I think that is a mistake. Silence, on that front, is golden. If they say nothing, they will be pushing back from the Bush policy that explicitly threatened nukes for almost everything. They should have many other changes to point to that demonstrate the reduced role for nuclear weapons.

  7. George William Herbert (History)

    So – question.

    Do you feel it is more likely that some country would within the forseeable policy future consider, threaten or launch a nuclear attack on civilians, or a CBW attack on civilians?

    Controlling nuclear weapons and their use is important, but CW and BW are considered weapons of mass destruction for good reasons as well.