Whither Nuclear Deterrence?

By Gregory F. Treverton

NOTE: The views expressed here are those of the author and do not necessarily represent or reflect the views of SMA, Inc.

Looking back from today on the nuclear stand-off during the Cold War, it seems in retrospect dangerous but simple. In game theoretical terms, the United States and the Soviet Union were engaged in a two-party game. They two had the opportunity to communicate, and in the later years of the Cold War arms control agreements provided more visibility into the actions of the other side. Those arrangements, plus cool heads, prevented the nuclear scare of 1983—the most dangerous nuclear crisis of the Cold War after the 1962 Cuban missile crisis—from unleashing nuclear war. By the end of the Cold War, in the words of my teacher, friend and mentor, Thomas Schelling, perhaps Moscow and Washington were akin to people about to cross a busy street: they weren’t deterred by the traffic from stepping into the street, they just knew better.

The rub, then, was that the United States and NATO sought more from nuclear weapons than to deter a nuclear attack. America sought to extend deterrence to Europe by threatening a nuclear response to a Soviet military sweep into Western Europe with conventional military force. Seeking to make that threat credible led to the tortured logic and language of “extended deterrence.” In that sense, the deterrence challenge before us is easier, for we no longer worry about a massive Russian attack on Western Europe. And even a nuclear North Korea should be easily deterred: it may believe that nuclear weapons protect it from “regime change,” but it knows that any use of nuclear weapons would be the end for it.

What Has Changed?

Even if the nuclear future is less dangerous than the high Cold War, deterrence is more complicated, in at least three ways. First, there are now more than two players in the game. In theoretical terms again, that simple increase in numbers increases uncertainty and decreases the ease of communication. Some of the players, like India, Pakistan, and Israel, are not of much concern to the United States, yet just the increase in numbers increases the chances of miscalculation leading to a use of nuclear weapons in anger, bringing implications for the international politics that are hard to fathom. One scenario in the 2017 Global Trends, the National Intelligence Council publication that I had the honor to oversee, included the use of a nuclear weapon, which in that scenario frightened leaders into pulling back and start talking, rather than signaling that all nuclear inhibitions were off. But in fact, we cannot know. And we should hope we don’t find out.

Yet unlike the Cold War, China can no longer be excluded from consideration. Its arsenal, while modernizing, remains small, it retains its doctrine of not using nuclear weapons first, and it has insisted it is too small to be part of any nuclear negotiations. Yet its nuclear weapons, in the context of its broader military modernization, raise awkward questions for deterrence: should, for instance, the United States seek to extend deterrence to Taiwan now, in a way it did for Western Europe in the Cold War? And one of the objections to the late, lamented Intermediate Nuclear Forces (INF) treaty was that it did not include China.

Second, advancing technology complicates warning, and shortens the time for decision. The increasing accuracy of missiles enables them to strike targets that formerly would have required a nuclear weapon to destroy, and the Chinese way of war, in particular, contemplates using conventional missiles early in any conflict with the United States, to destroy bases and staging locations. However, the proliferation of missiles, and their possible use, raises hard issues for warning: is the incoming ballistic missile nuclear-tipped? That was a critical issue during the Cold War as well, but it was simpler to deal with in a two-party game.

Third, doctrinal changes, especially by Russia, seek to make the use of small-yield nuclear weapons more thinkable. Russia now emulates Cold War NATO in threatening a first use of nuclear weapons. Its “escalate to de-escalate” doctrine really means “escalate to win,” opening the possibility of a “fait accompli” land grab, say of a strip of Lithuania between Russia and Kaliningrad. While Russia would surely lose a major conventional war with NATO, it does have superiority in areas close to home. It might grab land, then threaten or use nuclear weapons when NATO counterattacked, thus putting the onus on NATO to escalate still further. China might do something similar in invading Taiwan, thus forcing the United States to face the decision of whether to go nuclear in response.

Cross-Domain Deterrence

The future of deterrence is further complicated by what is called, slightly confusedly, “cross-domain” deterrence. NATO’s first-use threat during the Cold War might be thought of as an example: NATO sought to use the nuclear domain to deter conventional aggression. Now, though, the term has been broadened to various domains: can economic threats deter threatening actions, like Iran’s use of proxies in the Middle East? More pointedly, should nuclear threats not be ruled out in the effort to deter major cyber attacks?

Certainly, the prospect of relief from sanctions was a major reason Iran negotiated over its nuclear program in 2015, but that was a response to actual sanctions, not threatened ones. And the reimposed U.S. sanctions have not visibly deterred Iran’s actions in the region.

Cyber presents a more complicated case. Perhaps the wisest course would be some ambiguity, not threatening but not ruling out a nuclear response to a very damaging cyber attack. There would then be grave questions about what constituted “very damaging.” And, like Cold War nuclear threats, the implied threat might be rational to make but would be insane to carry out.

Indeed, I thought we had left behind the agonies of early Cold War nuclear strategy, which contemplated automatic responses or “leaving something to chance” in an effort to make the incredible credible. The period from the end of the Soviet Union until now looks, in retrospect, simple: the sole purpose of nuclear weapons was to deter nuclear attacks. Alas, that period now seems ending.

Gregory F. Treverton stepped down as Chair of the U.S. National Intelligence Council in 2017. He is now Professor of the Practice at Dornsife College, University of Southern California, Chair of the Global TechnoPolitics Forum, and an SMA Executive Advisor.