Science1 publisher2 min readPublished
Brookings finds nuclear crises clustered where one side held a big warhead advantage
Melanie Sisson's Brookings report counts historical episodes of nuclear threat because the two-peer deterrence claim cannot be tested directly, and it finds those episodes more common in periods of lopsided arsenals than of rough parity.
The Scientist · Science desk
What happened
- A Brookings Institution report by senior fellow Melanie W. Sisson presents evidence that increasing the size of the U.S. nuclear arsenal could make nuclear crises more likely.
- Because the claim cannot be tested directly, the report draws on historical instances in which the United States made or received nuclear threats and compares the balance of forces at the time.
- Episodes such as the Cuban Missile Crisis, confrontations with North Korea and Russia's threats over Ukraine were more common when one side held a substantial advantage, usually the United States.
- Where and how many warheads were deployed, and the doctrines governing U.S. nuclear policy, showed no consistent association with how often crises occurred.
- The Bulletin of the Atomic Scientists puts the U.S. at around 5,000 warheads, Russia at a little less than 5,500 and China at around 600.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- decision Anyone weighing extra warheads against submarine and command-system work now has an empirical argument on one side of that spending choice, and its strength depends on episode counts that so far only Brookings has run.
- contradiction Two camps read one arsenal in opposite directions: the two-peer argument holds that it cannot credibly deter China and Russia together, while the report holds that it could still retaliate against a combined attack.
- constraint This experiment is impossible to run, so both the case for expansion and the case against it rest on how a limited set of past episodes is defined and coded.
- exposure If AI-enabled attacks on command and control are the live vulnerability, the exposure sits in systems that extra warheads leave untouched.
An episode enters Sisson's sample when the United States made or received a nuclear threat [6]. If a case qualifies because a threat was made, then the concentration of episodes in asymmetric periods partly measures who was willing to make threats, and the report puts the United States on the advantaged side in most of them [8].
These cases leave the direction of causation open. A state far behind in warheads is both a plausible target for coercion and a state with reason to build, so the gap and the crisis can share a cause without either one producing the other.
Deployment locations, deployed numbers and declared doctrine showed no consistent association with crisis frequency [10]. Those nulls carry less weight than the positive finding sitting beside them: a small set of historical episodes will often return that answer even where a real association exists. The phys.org account of the report does not say how many episodes were counted.
The current gap is roughly eight American warheads for every Chinese one [1]. Russia and China together hold just under 6,100 [2], and Sisson's position is that the existing U.S. arsenal would still be able to retaliate if the two combined against it [12].
Each episode named in the account is a confrontation between the United States and a single adversary [3]. The frequency comparison is therefore measured on one-on-one crises. The claim it answers is newer: the two-peer argument holds that a stockpile long treated as adequate for deterrence [4] cannot credibly cover China and Russia at the same time [5].
Her alternative goes to the survivability half of deterrence, the condition in which the ability to strike first does not eliminate the other side's ability to strike back [15]. Keep strengthening the nuclear submarine fleet, she argues, and defend nuclear command and control against AI-enabled cyberattacks that could degrade or disable the capability to respond [13][14].
What to watch
- An independent recount of the crisis cases under a different definition of a nuclear crisis would test whether the parity result survives recoding.
- The Bulletin of the Atomic Scientists' next estimate for China: a sharply higher number moves the present balance toward the near-parity region of Sisson's own data.
- Whether U.S. budget requests fund submarine survivability and command-and-control defense or additional warheads.