Illustration by Thomas Gaulkin
Illustration by Thomas Gaulkin

Can minimum deterrence save nuclear arms control?

By Stephen J. Cimbala, Lawrence J. Korb | June 10, 2024

Prospects for Russian–American strategic nuclear arms control seemed unfavorable in the aftermath of Russia’s invasion of Ukraine in February 2022 and Russia’s later suspension of formal participation in discussions about New START and possible post-New START options.[1] If official conversations about strategic nuclear arms limitations between the United States and Russia (and possibly China) do occur, however, a different conceptual framework for negotiating parties might prove fruitful.[2]  One such framework focuses on the concept of minimum deterrence, which would call for nuclear arsenals that are smaller than the two major nuclear superpowers now possess, but large enough to allow sufficient retaliatory strike capability as to retain strategic stability. In examining some of the broader politico-military and arms control contexts for the viability of any minimum nuclear deterrence regime, we consider “how much is enough” with respect to US and Russian strategic nuclear forces. We acknowledge that the rise of China as a strategic nuclear superpower presents challenges for nuclear-strategic stability and a minimum deterrence regime, but we believe that these challenges are not insurmountable.

Defining minimum deterrence. The meaning of “minimum” deterrence is not necessarily obvious without having addressed the question “compared to what?” Nuclear strategists would probably agree that minimum deterrence lies somewhere between assured destruction, as emphasized during Cold War discussions about nuclear strategy, and nuclear abolition.[3] Exactly where “minimum” lies on this continuum is more debatable.[4] At least four kinds of variables are in play in classifying nuclear strategies:

  • The political and military objectives for which forces are tasked.
  • The specifics of nuclear targeting plans, related to retaliatory objectives but not necessarily reflecting the actual intent of policy makers.
  • The numbers of weapons and launchers deployed and their assumed rates of survivability against first or later strikes, and
  • The command-control systems and operational protocols of the states’ nuclear forces, including their dependency on high states of alert or prompt launch for survivability.

During the height of the Cold War, assessment of these variables might have led to a spectrum of possible nuclear deterrent strategies as summarized in the following table.

Attributes of Cold War Nuclear Deterrence Strategies

Counterforce-warfighting Assured destruction Minimum deterrence

and targeting

Victory denial or “prevailing” in a protracted conflict by imposing escalation dominance on the opponent at any phase of conflict— ambitious targeting of enemy nuclear forces and C3

Inflicting retaliatory strikes sufficient to impose “unacceptable” damage on any attacker, including its remaining forces, C3, industry, and population

Imposing unacceptable damage to the attacker’s society and civilian population and-or national infrastructure, although with forces less than those required for assured destruction

Numbers of weapons-launchers required

Numbers of survivable weapons capable of attacking or holding at risk military, C3, industry, and population targets, if necessary through phases of a protracted war; may also require antimissile defenses for protecting population and/or forces; requires numbers of deployed warheads in the thousands, well above the threshold for assured destruction

Numbers of survivable weapons capable of attacking military, C3, industry, and population targets and inflicting “unacceptable” damage; allows for flexible targeting but does not envision fighting a protracted nuclear war to a successful conclusion; requires numbers of deployed warheads in the thousands, fewer than required for counterforce-warfighting strategies

Numbers of survivable weapons sufficient to destroy major infrastructure and the sinews a modern national economy, while not necessarily emphasizing the destruction of urban-industrial areas, but also not necessarily guaranteeing “city avoidance”; usually requires numbers of deployed warheads in the hundreds

Command-control and alert-launch protocols

Political and military C3 must be not only survivable against initial attacks but enduring through various phases of a protracted conflict; some proportion of the force will be on hair trigger alert even in peacetime

Political and military C3 must be survivable for second strike retaliation and for post-attack negotiation for war termination—no forces on high alert required in peacetime but not precluded either

Political and military C3 must be survivable for second strike retaliation; no forces on high alert in peacetime

Sources: Keith B. Payne and David J. Trachtenberg, Deterrence in the Emerging Threat Environment: What is Different and Why It Matters (Fairfax, Va.: National Institute for Public Policy, National Institute Press, 2022); David A. Cooper, Arms Control for the Third Nuclear Age: Between Disarmament and Armageddon (Washington, D.C.: Georgetown University Press, 2021); Fred Kaplan, The Bomb: Presidents, Generals, and the Secret History of Nuclear War (New York: Simon and Schuster, 2020);  Stephen J. Cimbala, The United States, Russia and Nuclear Peace (New York: Palgrave Macmillan, 2020); Paul K. Davis, “What Do We Want from the Nuclear Command and Control System,” draft paper for presentation at NC3 and Global Security Workshop, Stanford University, January 2019; Andrew Futter, The Politics of Nuclear Weapons (London: Sage Publications, 2015); Robert Jervis, The Meaning of the Nuclear Revolution: Statecraft and the Prospect of Armageddon (Ithaca, N.Y.: Cornell University Press, 1989), pp. 74-106; Charles Glaser, “Why Do Strategists Disagree about the Requirements of Strategic Nuclear Deterrence,” Ch. 2 in Lynn Eden and Steven E. Miller, eds., Nuclear Arguments: Understanding the Strategic Nuclear Arms and Arms Control Debates (Ithaca, N.Y.: Cornell University Press, 1989), pp. 109-171; Scott D. Sagan, Moving Targets: Nuclear Strategy and National Security (Princeton, N.J.: Princeton University Press, 1989, esp. pp. 58-97; Desmond Ball, “The Development of the SIOP, 1960-1983,” Ch. 3 in Desmond Ball and Jeffrey Richelson, eds., Strategic Nuclear Targeting (Ithaca, N.Y.: Cornell University Press, 1986), pp. 57-83; Robert Jervis, The Illogic of American Nuclear Strategy (Ithaca, N.Y.: Cornell University Press, 1984), esp. Ch. 3-4; Ball, “U.S. Strategic Forces: How Would They Be Used?,” in Steven E. Miller, ed., Strategy and Nuclear Deterrence (Princeton, N.J.: Princeton University Press, 1984), pp. 215-244.  

The preceding table cannot capture all the nuances or possible variations within, and among, these three kinds of strategies because states’ declaratory strategies are not always consistent with their operational policies.[5]  It does illustrate some of the qualitative and quantitative points of similarity and difference among these kinds of generic nuclear strategies.

In today’s world, minimum deterrence implies that US, Russian, and Chinese arsenals would each be limited to a maximum number of 1,000 operationally deployed strategic nuclear weapons, or fewer if possible. “Fewer if possible” means that for Washington and Moscow to go below 1,000 deployed weapons on transoceanic or intercontinental launchers, other acknowledged nuclear weapons states would have to commit themselves to proportional reductions or limitations. A global minimum deterrence system might allocate a maximum of 500 operationally deployed weapons to states other than the Big Three. Sub-strategic nuclear weapons, including tactical or operational weapons that are deployed on land, at sea, or in the air, have both political and military-operational contexts requiring separate discussion. There is certainly the possibility that, in any multilateral, constrained nuclear proliferation regime, some weapons of medium or intermediate range might have to be included as “strategic,” based on their potential effects against likely regional adversaries.

Measuring minimum deterrence. The charts that follow illustrate the deterrence stability of two minimum deterrence regimes.[6] In the first case, US and Russian strategic nuclear forces are limited to a maximum of 1,000 operationally deployed weapons for each state. In the second case, a lower limit of 500 deployed weapons is imposed.[7] For these larger and smaller forces, we have calculated the expected numbers of second-strike surviving and retaliating warheads under four operational conditions of alertness and launch protocols:

  • Generated alert, and launch on warning.
  • Generated alert, riding out the attack and retaliating.
  • Day-to-day alert, and launch on warning, and
  • Day-to-day alert, and riding out the attack.
Formal risk assessments and nuclear arms control: exploring the value of modern methodologies

Day to day alert refers to the status of nuclear forces during normal peacetime conditions.  Bomber and non-strategic nuclear forces are not loaded with nuclear weapons day-to-day.  A portion of the ballistic submarine force is maintained at sea day to day to ensure survivability.  The bulk of the ICBM force is on alert day to day.  During a crisis, alert levels would be raised (generated) for some or all of the forces based on decisions by the National Command Authority.

One might anticipate that, in general, the numbers of surviving and retaliating warheads would diminish as we proceed from the first option above through the last. But that progression is not necessarily automatic, depending on the specific circumstances of attack and response.

The results of this analysis appear in Chart One and Chart Two.  Chart One summarizes the numbers of second strike-surviving and retaliating warheads for the United States and for Russia under each of the operational conditions listed above, in the case of 1,000 maximum deployed weapons for each country. Chart Two provides information equivalent to that summarized in Chart One, but for the more restrictive case in which maximum deployments are capped at 500 weapons for each.

Surviving and retaliating warheads

Estimates have been made of the numbers of weapons available for retaliatory attack under four possible force structures for each state.  For the United States, the four structures include: a balanced triad of ICBMs, SLBMs and bomber delivered weapons; a dyad without ICBMs; a dyad without bombers; and a force based entirely on SLBMs.  For Russia, the four force structures include: a balanced triad; a dyad of ICBMs and  bombers; a dyad of ICBMs and SLBMs; and a force made up entirely of ICBMs.  The alternatives to triads for each country reflect differences in their force modernization priorities based on past experience: ICBMs for Russia, and SLBMs and bombers for the United States.  Current U.S. and Russian modernization plans call for continued modernization of triads; hypothetical alternatives are included for comparative purposes.

In the canonical case often used for analysis (but not necessarily reflecting the likelihood of actual operations), either an American or Russian strategic nuclear triad, under conditions of “generated alert, riding out the attack,” could provide for more than 500 retaliating weapons under a deployment limit of 1,000 warheads.  For a deployment limit of 500 warheads, a U.S. retaliatory triad on generated alert and riding out the attack could provide more than 300 surviving and retaliating weapons; under the same conditions, a Russian triad would yield more than 200 retaliating warheads.

Objections to minimum deterrence. Objections to the preceding analysis and argument can be anticipated along the following lines:

First, China might not play. Unless and until China feels it has an essential nuclear-strategic parity with the United States and Russia, Beijing might be disinterested in nuclear arms control. On the other hand, this objection might apply only or mainly to the bean-counting aspect of nuclear force reductions and to required inspections.

The United States and Russia would benefit from including China in multilateral discussions about the nature of strategy and strategic thinking, about threat perceptions and the international environment, and about the need for international collaboration on issues including nonproliferation and nuclear risk reduction. Eventually China might find a comfort zone as one of the Big Three nuclear weapons states, appreciate the symbolic and substantive benefits of sharing arms control preeminence in international forums, and conclude that it would be smarter to be included in the arms control tent than standing apart from it.

China can be expected to ease into nuclear arms control gradually – unaccustomed as China is to the transparency required for arms limitation and reduction talks. Nevertheless, China could eventually decide that the alternative to participation would be an apparent obliviousness on the part of its leadership to the possibility of an uncontrolled nuclear arms race.

Part of the concern about a rising nuclear China is the fear that China and Russia might combine their strategic nuclear forces and pose an insuperable threat to current plans for US nuclear modernization. In a worst case of deterrence failure, the United States could find itself attacked simultaneously in Europe and in Asia by large scale conventional forces supported by the implicit or explicit threat of a joint Sino-Russian resort to tactical or strategic nuclear weapons. On the other hand, it cannot be assumed that an outbreak of wider war in Europe would necessarily cause China to combine with Russia under all circumstances. Nor would Russia necessarily come to China’s aid in case of a crisis in Asia. Much would be circumstantial with respect to each side’s political motivations.

A second objection to minimum-deterrence proposals might come from military planners and those who make decisions about nuclear employment policy. Would a minimum-deterrence regime permit the United States to field a force that provided for more flexible options than a one-variety “wargasm” that limited policy makers’ range of choices under exigent conditions of deterrence failure?

The answer to this concern is that, so long as the US arsenal is capped at the same number as those of potential adversaries, no single power should be disadvantaged relative to the others with respect to its capabilities for flexible targeting and escalation control. In fact, compared to the strategic nuclear forces of Russia and China, the United States’ would provide more flexibility than those of Russia or China because more of the US deterrent resides on ballistic missile submarines, the most survivable leg of any strategic nuclear triad. In theory, Russia and China might gang up on the United States and grossly outnumber the Americans in operationally deployed weapons. In practice, however, it is difficult to imagine either Russia or China riding the coattails of the other party into a holocaust—even under the best of conditions for them.

A third objection to a minimum-deterrence regime for strategic nuclear forces is that it does not deal with the problem of nonstrategic nuclear weapons. The concern is well placed, but nonstrategic nuclear weapons come in so many varieties that separate negotiations will be needed to agree a formula among the United States, Russia and China, as well as among other nuclear weapons states or aspiring NWS. But the United States and Russia should prioritize bilateral talks about what to do with nonstrategic nuclear weapons now deployed in Europe.

A fourth concern about the preceding analysis is that it does not provide any prescription for preventing the rise of new nuclear weapons states, especially those with revisionist aims relative to the existing international order.[8] Guilty as charged. A nuclear Iran is the most immediate worry here, but also, additional misbehavior by North Korea that encourages South Korea, Japan, or other American allies in Asia to become nuclear weapons states.

Preserving the nuclear test ban after Russia revoked its CTBT ratification

No magic bullet is offered here to guarantee against an expanded club of nuclear weapons states. But a combination of good example set under a minimum-deterrence regime—with successful deterrence or other persuasion against additional nuclear weaponization—is one possible pathway to reviving an otherwise endangered nonproliferation regime.[9]

One option for avoiding a nuclear arms race. Minimum deterrence is a concept, and the need for adaptive concepts is urgent because the international system has changed significantly since the end of the Cold War. Minimum deterrence may be a practical consolation prize for advocates of nuclear abolition or, at the same time, a symbol of US nuclear stasis or decline for those favoring more ambitious strategies.[10] At the very least, minimum deterrence should contribute to serious discussions about limitations in strategic nuclear force sizes among the United States, Russia, and China. An unlimited arms race only invites instability.

In addition to the challenge of “how much is enough” with respect to nuclear weapons and launchers, minimum deterrence will also have to adjust targeting plans to fit the new technological environment of the 21st century. Weapons for destruction of “war supporting industry” or national economies will be less relevant than the ability to disable or destroy the sinews of state political control and the brain and central nervous system of armed forces. Therefore, the earliest strikes will likely be made against cyber systems and space-based assets for reconnaissance and warning, navigation, C3 (command, control, and communications), and other vital military functions.

Finally, experts disagree about whether nuclear superiority correlates with favorable outcomes in coercive bargaining between rivals. For example, University of Virginia expert Todd S. Sechser and Teas A&M political scientist Matthew Fuhrmann argue that “states rarely behave as if nuclear superiority provides them with advantages in coercive bargaining.”[11] On the other hand, Georgetown University government professor Matthew Kroenig elaborates a “superiority-brinkmanship synthesis” theory that “military nuclear advantages increase a state’s willingness to run risks in international conflicts.”[12] Both arguments are supported with considerable statistical analysis and case studies.

It would seem that, the weaker the connection between nuclear superiority and favorable outcomes in bargaining, the stronger the case for minimum force sizes and capabilities. But in deterrence and in war, the enemy always gets a vote. Therefore any US strategic nuclear posture and supporting arms control regime should continue to include a “hedging” capability for unforeseen challenges and nuclear “black swans.”

Stephen J. Cimbala

Stephen J. Cimbala is Distinguished Professor of Political Science at Penn State University, Brandywine.... Read More

Lawrence J. Korb

Lawrence J. Korb is a senior fellow at the Center for American Progress. He is also an adjunct professor of security studies at Georgetown... Read More


  • [1] For the New START treaty, see: Treaty between the United States of America and the Russian Federation on Measures for the Further Reduction and Limitation of Strategic Offensive Arms (Washington, D.C.: U.S. Department of State, April 8, 2010) [LINK].  For assessments, see:  Steven Pifer, “The US and Russia must re-assess their strategic relations in a world without New START,” Bulletin of the Atomic Scientists, June 13, 2023 [LINK] and Stephen J. Cimbala and Lawrence J. Korb, “Reviving Arms Control, post-Ukraine: Why New START Still Matters,” Bulletin of the Atomic Scientists, July 13, 2022 [LINK].
  • [2] On the significance of China’s emergence as a nuclear peer competitor with the U.S. and Russia, see: Brad Roberts, et. al., China’s Emergence as a Second Nuclear Peer: Implications for U.S. Nuclear Deterrence Strategy, Report of a Study Group Convened by the Center for Global Security Research (Livermore, Calif.: U.S. Department of Energy, Lawrence Livermore National Laboratory, Spring 2023) [LINK].
  • [3] Minimum deterrence has, in fact, a considerable pedigree, dating back to some of the earliest U.S. debates on nuclear strategy and deterrence.  “Minimum deterrent” strategies have variations and are sometimes referred to as “deterrence only” strategies.  See Herman Kahn, On Thermonuclear War, Second Edition (New York: The Free Press, 1969), pp. 7-13; and Kahn, On Escalation: Metaphors and Scenarios (New York: Frederick A. Praeger, 1965), pp. 281-284.  See also: John Baylis, “Nuclear Weapons, Prudence, and Morality: The Search for a ‘Third Way’,” Ch. 5 in Baylis and O’Neill, eds., Alternative Nuclear Futures, pp. 70-86 inclusive, esp. pp. 78-81.
  • [4] On minimum deterrence strategies, see: James Wood Forsyth Jr., B. Chance Saltzman and Gary Schaub Jr., “Minimum Deterrence and Its Critics,” Strategic Studies Quarterly, No. 4 (Winter, 2010), pp. 3-12; Forsyth, Saltzman and Schaub, “Remembrance of Things Past: The Enduring Value of Nuclear Weapons,” Strategic Studies Quarterly, No. 1 (Spring, 2010), pp. 74-89; and Stephen M. Walt, “All the nukes you can use,” Foreign Policy, May 24, 2010 [LINK].
  • [5] On this point, see especially Desmond Ball, “The Development of the SIOP, 1960-1983,” Ch. 3 in Desmond Ball and Jeffrey Richelson, eds., Strategic Nuclear Targeting (Ithaca, N.Y.: Cornell University Press, 1986), pp. 57-83; and Ball, “U.S. Strategic Forces: How Would They Be Used?,” in Steven E. Miller, ed., Strategy and Nuclear Deterrence(Princeton, N.J.: Princeton University Press, 1984), pp. 215-244.
  • [6] Calculations in this study were made using the Arriving Weapons Sensitivity Model (ASWM@) developed by Dr. James Scouras. The model calculates the outcomes of nuclear force exchanges and provides graphic and tabular information. For additional information, see: Stephen J. Cimbala and James Scouras, A New Nuclear Century (Westport, Ct.: Praeger Publishers, 2002), and Cimbala, War Games: U.S.-Russian War Games and Nuclear Arms Control (Boulder, Colo.: Lynne Rienner Publishers, 2017), pp. 241-243. Dr. Scouras has no responsibility for analysis or arguments in this study.
  • [7]  Forces are hypothetical structures, although not unrealistic ones, used for analytical purposes, not predictions of actual deployment decisions.  For estimates of Russian and U.S. present and future deployments, see:  Hans M. Kristensen, Matt Korda, Eliana Johns, and Mackenzie Knight, “Nuclear Notebook: Russian nuclear weapons, 2024,”Bulletin of the Atomic Scientists, March 7, 2024 [LINK]; Hans M. Kristensen & Matt Korda (2023), “Nuclear Notebook: United States nuclear weapons,”2023, Bulletin of the Atomic Scientists, January 16, 2023 [LINK]; and U.S. Congressional Budget Office, Approaches for Managing the Costs of U.S. Nuclear Forces, 2017 to 2046 (Washington, D.C.: CBO, October 2017), pp. 33 and 44 [LINK].
  • [8]On the current status of nuclear nonproliferation, see:  Rebecca Davis Gibbons, The Hegemon’s Tool Kit: US Leadership and the Politics of the Nuclear Nonproliferation Regime (Ithaca, N.Y.:  Cornell University Press, 2022); and Campbell Craig, S.M. Amadae, Nicholas L. Miller, Michal Onderco, Tom Sauer, and Rebecca Davis Gibbons, Roundtable Review 15-35, March 22, 2024, Robert Jervis International Security Studies Forum, < 15-35.
  • [9]For original thinking about future options for arms control and nonproliferation, see: Henry D. Sokolski, “Nuclear Non-Proliferation…If You Can Keep It,” January 6, 2023 [LINK]; and Amy J. Nelson and Michael O’Hanlon, “All START: a proposal for moving beyond US-Russia arms control,” Bulletin of the Atomic Scientists, March 16, 2023 [LINK].
  • [10] See, for example: Madelyn Creedon, Chair, and Jon Kyl, Vice Chair, Congressional Commission on the Strategic Posture of the United States, America’s Strategic Posture: Final Report (Washington, D.C.: October, 2023).
  • [11] Todd S. Sechser and Matthew Fuhrmann, Nuclear Weapons and Coercive Diplomacy (Cambridge: Cambridge University Press, 2017), p. 254 and passim.
  • [12] Matthew Kroenig, The Logic of American Nuclear Strategy: Why Strategic Superiority Matters (Oxford: Oxford University Press, 2018), p. 3 and passim.
Get alerts about this thread
Notify of
1 Comment
Newest Most Voted
Inline Feedbacks
View all comments
Wesley Parish
Wesley Parish
9 days ago

While reading this article I kept thinking to Daniel Ellsberg’s recorded statement “A Common Insanity”, and wondering how much extinction is not enough? Perhaps that should be included in the diplomatic discussions?