Michael KreponDonald Trump’s Challenge to Nuclear Orthodoxy

The belief system in nuclear deterrence makes the most sense at the most elementary level: nuclear weapons deter use by an adversary for fear of retaliation in kind. Deterrence therefore requires survivable nuclear capabilities. It doesn’t take much by way of force structure to convey the threat that leaves something to chance, which is the essence of nuclear deterrence. From this simple construct, a superstructure of belief has been erected, driving requirements and stockpiles to great heights.

The intricacies of this belief system are too hard and off-putting to explain. Discussions of nuclear targeting strategies and plans must be avoided because they sound so Strangelovian. Public opposition would be fanned if they were revealed. True believers accept these tenets without question and ask others to take them on faith.

We also have faith in national leaders to recognize that it would be crazy to fight a nuclear war whose consequences are all too well understood. Consequently, most of us live with the belief system of nuclear deterrence without complaint – or at least try to avoid thinking about these tenets. It takes the prospect of a person with deeply suspect personality traits becoming President for us to focus on how crazy the extrapolations and expansive requirements of nuclear deterrence are. For this, let us give thanks to the candidacy of Donald Trump.

There are two primary arguments against serious debate over this belief system. One is that huge stockpiles aren’t for actual use; they are for deterrence. The second is that nuclear deterrence works, as was evident by their non-use during the Cold War. The rejoinder to both assertions is the same: What if deterrence breaks down and use is required? How does the belief system in nuclear deterrence and its expansive requirements help us then?

A third argument for holding on to a belief system in expansive requirements for nuclear deterrence is that Russia’s Vladimir Putin sincerely and deeply believes in all this, and if U.S. leaders don’t, deterrence will be weakened. If the United States gets off this treadmill and Russia stays on, more bad things will happen around Russia’s periphery. Or Beijing, which has not gotten on this treadmill, might decide to hop aboard, and bad things will then happen along its periphery, too. (A fourth argument, that it is unwise to wean U.S. allies from their addiction to extended U.S. nuclear deterrence, including first use, has been dealt with earlier.)

There has never seemed to be a good time for serious debate on the assumptions that have generated extreme redundancy in nuclear capabilities. The Cold War nuclear competition was too intense to debate fundamentals. After the Cold War ended, there seemed no reason for debate because deployed force levels and stockpiles were greatly reduced. And it’s not a good time now, according to the defenders of nuclear orthodoxy, with Vladimir Putin disregarding the sovereignty of neighboring states and China’s power on the rise. Donald Trump’s candidacy for President argues that, to the contrary, this is a perfectly good time to reconsider the ornately encrusted requirements for nuclear deterrence.

One reason is that breakdowns in deterrence are common, as Keith Payne (The Great American Gamble: Deterrence Theory and Practice from the Cold War to the Twenty-First Century, 2008) often reminds us. They even occur between nuclear-armed states that have fought two limited border wars – so far. Serious accidents have happened with nuclear weapons. There have been false alarms and close calls, sometimes during severe crises. Mushroom clouds can appear on battlefields due to the aggressive use of airpower against dual-capable weapon systems. Command-and-control arrangements can break down under severe duress.

There are other ways in which nuclear exchanges could occur. If they do, for whatever reason, the other fundamental assumption behind nuclear deterrence – that huge stockpiles aren’t for actual use, but for deterrence – goes by the boards. Once the nuclear threshold is crossed, everything depends on escalation control, not “credible” deterrence and war-fighting options. These upper stories of the edifice of nuclear deterrence make sense only in abstract reasoning, not in the crucible of actual warfare.

Nuclear deterrence theory works until it fails, and then the foremost danger is that it will fail catastrophically. The only circumstance under which nuclear use does not fail disastrously rests on extremely limited use, after which national leaders might somehow be able to intervene to control escalation – even though they have failed to prevent a war. What Bridge Colby calls the “pacifying logic” of deterrence theory implodes with first use. Once the nuclear threshold is crossed, the belief system surrounding nuclear orthodoxy falls like a house of cards.

Orthodoxy-busting leaders like Ronald Reagan and Mikhail Gorbachev saw through all this and were fine with junking the whole belief system. They didn’t succeed, but they did tear down many stories from the edifice of nuclear deterrence. The path of least resistance (although it still prompts great resistance) for their successors is to take down one story at a time of what’s left. This process is slow going and prone to disruption. Moving faster requires a hard look at the belief system that has produced this monument to wishful thinking – and having a partner to do so.

Nothing in Putin’s behavior of late suggests that he is this partner – although he, like the next U.S. President, will be hard-pressed to fund the full panoply of strategic modernization program. If Putin opts to jointly tear down another story from this edifice for budgetary reasons, its defenders in the United States will be agog, pointing to Putin’s misdeeds as reasons to hold on to force structure and warheads the United States can’t use.

Going it alone – if Putin is unwilling to join in another round of strategic arms reduction – flies in the face of nuclear orthodoxy. President Obama has rejected this option, while keeping in place a wide array of strategic modernization programs. None of this has assuaged critics on Capitol Hill who continue to chip away at remaining treaty constraints, expand the envelop for national missile defenses directed at Russia and China, while opposing implementation of the nuclear non-proliferation deal with Iran.

The next President has the opportunity to revisit these terms of domestic engagement, which have not served the Obama Administration well. A new administration can alter this equation: However much strategic force structure the next President decides to retain, the significant costs of replacement warrant, in return, the Senate’s consent to ratify the Comprehensive Test Ban Treaty.

Note to readers: For more along these lines, see “Nuclear Orthodoxy after Trump: The Real Requirements of Deterrence,” FOREIGN AFFAIRS, August 15, 2016.


  1. ybutt (History)

    What’s more the ‘ “pacifying logic” of deterrence theory ‘ is just a theory, grounded only in ~70 years of no unintentional/mistaken/accidental use — though there were several close calls.

    In fact, the ~70 years of no unintentional/mistaken/accidental use only places rather weak statistical limits on the theory that nuclear deterrence has kept the peace. Similarly the space shuttle was safe — until it wasn’t. And deep water drilling was safe — until it wasn’t.

    Prof. Martin Hellman at Stanford has suggested in the Bulletin that the Academies look at the issue of the alleged safety of nuclear deterrence, rigorously using proper statistics & quantitative risk analysis:


    “a child born today with a 78-year life expectancy would have almost a 10 percent
    chance of not living out his or her natural life due to our reliance on nuclear weapons —
    a highly unacceptable level of risk”

    footnote 2:

    ” If an event has 1 chance in 1,000 of occurring each year, then the probability of it not happening
    in 78 years is 0.999^78, i.e. 92.5 percent, and the probability of it occurring within that time frame
    is 7.5 percent. In order not to imply unwarranted accuracy, this is rounded to 10 percent”


    One can quibble with some of the assumed numbers but the basic point is that no nuclear use in 70 years should not be terribly comforting to anyone. Bit more here: