Table of Contents

The best nuclear force would be one that is:

  • credible enough to deter adversaries and reassure allies and partners;
  • least likely to provoke escalation if deterrence fails but could survive adversary escalation if it occurred; and
  • would not cause more destruction than necessary to meet wartime objectives.

Such a force is hard to design in theory and harder to maintain in practice. Nuclear weapons systems take years to fund, design, develop, produce, and deploy. Once they are deployed, they are expected to remain in the arsenal for decades unless they are eliminated through arms reductions or obsolescence. The longevity of the arsenal may impede its adaptability to changing global security dynamics.

And so, the triad. For five decades, the United States has had a triad of nuclear weapon delivery systems, based on sea, air, and land. This arrangement developed through competition between the Air Force and the Navy for a share of the nuclear mission, and the evolution of available delivery technologies thereafter.1 Only after the triad was in place did strategists and officials enshrine its virtues.2 It gradually became sacrosanct and vital to the Air Force, the Navy, the eleven states that host triad delivery systems, and the many other states and enterprises that produce these systems. Alongside it, an NC3 system was designed to survive the stresses of nuclear conflict.

In 2021, the executive and legislative branches will continue to debate whether and how to pay for, produce, and deploy delivery platforms and warheads. The current plan started with a commitment from the Obama administration in 2009 to “modernize or replace” the strategic triad and its constituent delivery platforms and warheads nearing the end of their operational lives. Since then, the services have mostly proposed plans to replace old systems with new ones, rather than modernize extant systems. These plans include the development of new nuclear warheads to be deployed on new delivery platforms in the next two decades.3

In order to evaluate these plans, this chapter proceeds through each component (sea, air, and land) of the triad, analyzing arguments for and against changes that have been proposed within each category. It then offers recommendations and closes by discussing NC3 and offering principles to guide its modernization effort.

Summary of main force structure recommendations:

  • Continue the B-21 Raider, LRSO (Long-Range Standoff weapon), B61-12, and Columbia SSBN programs, and maintain the current B61 posture in Europe.
  • Extend the life of the Minuteman III ICBM and seek bilateral reductions with Russia in ICBM forces before further development of a replacement ICBM.
  • Cancel the nuclear-armed sea-launched cruise missile (SLCM-N).
  • Maintain the low-yield version of the Trident SLBM pending answers to questions detailed below.



Strategic bombers are well suited for the types of nuclear crises the United States is likely to face—emanating from major conventional war in Europe or Northeast Asia. These bombers provide visibility, flexible attack trajectories, and multiple long-range conventional and nuclear employment capabilities suited to modern deterrence requirements. (Bombers are the only triad leg to provide substantial conventional capabilities.) The new B-21 Raider will provide a stealthy, modernized replacement for conventional B-1B bombers and a small, expensive fleet of stealth B-2s built in the 1990s, complementing the aging fleet of B-52s that dates from the 1960s. (The B-52s will undergo a modernization in the near future to extend their life until 2050.4)

The ability to flexibly deploy bombers is important for allied assurance.5 Aircraft are the politically visible representation of the U.S. nuclear arsenal (even when stealthy in flight). Indeed, the United States has extensively used nuclear-capable bombers in peacetime extended deterrence signaling through bomber assurance and deterrence (BAAD) missions flying over or traveling near allied territory.

Strategic bombers are well suited for the types of nuclear crises the United States is likely to face—emanating from major conventional war in Europe or Northeast Asia.

Bombers have some downsides. First, their effectiveness may be degraded by adversaries’ improving air defenses. The United States has long recognized that the B-52H will have to operate as a standoff weapon and employ long-distance cruise missiles. However, the B-2s retain substantial capabilities to reach targets deep in adversary territory. A key unanswered question is how U.S. nuclear, conventional, cyber, or other nonkinetic operations could be used to reduce the potential of an adversary’s air defenses such that U.S. bombers would be effective. Attacks to suppress enemy air defenses (SEAD operations) may themselves be escalatory if their physical destructiveness were extensive.

Second, bombers are not on day-to-day alert with weapons deployed. Thus, they are theoretically vulnerable to surprise attack before being prepared for nuclear operations. Some experts argue therefore that bombers should not count as part of the deployed strategic deterrent. However, it is practically inconceivable that the U.S. homeland would face a strategic attack from Russia, China, or North Korea in peacetime with such short warning that the bomber force could not be made ready for quick mobilization and dispersal. Instead, the potential need to conduct nuclear strikes would arise from crises and, more likely, major armed conflict in which the bomber force would already be at least in heightened readiness. Bombers, when alert and conducting operations, are less susceptible to surprise attack.

The B-21 Raider should be fully funded for the currently requested fleet of 100 aircraft.

Given the B-21’s conventional capabilities, it is possible the Air Force will increase the number of B-21 bombers it requests. The commander of the Air Force Global Strike Command, which manages the nuclear and conventional long-range strike missions for U.S. bombers, has stated his interest in growing the bomber force for conventional missions.6 To satisfy the nuclear mission, there is no need to increase the current request for 100 aircraft—34 more than the current number of nuclear-capable bombers in the U.S. Air Force—which will be certified for nuclear operations within two years of deployment.7

One concern of some analysts is that adversaries would be unable to discern whether B-21 bombers threatening them were carrying conventional or nuclear weapons. This concern can be addressed by basing nuclear bombers separate from conventional ones and utilizing well-known arms control, transparency, and confidence-building procedures to confirm that conventional bombers are indeed being sequestered from nuclear operations. As the United States continues to pursue dual-capable bombers, it is in its interests to be unambiguous to Russia, China, and North Korea regarding bomber operations in peacetime and especially during crises.

If concerns grow that B-21s could be preemptively destroyed by nuclear or conventional attack against air bases, creative basing arrangements should be the priority. At one time, the Strategic Air Command operated nuclear-capable bombers from dozens of bases in the continental United States and abroad.8 Although this number shrank at the end of the Cold War, the U.S. Air Force could explore the feasibility of operating B-21s in rare contingencies from more bases, with a capability to deploy to them rapidly if needed. This could be a way to increase the current target set of three strategic bomber air bases to complicate an adversary’s first strike planning.9


The United States has maintained an air-deliverable nuclear-armed cruise missile for decades. The planned LRSO will replace the current AGM-86B air-launched cruise missile (ALCM), which dates from the 1980s. The LRSO is designed to preserve the standoff capabilities of the entire nuclear bomber force, including legacy B-52H bombers. It will provide a stealthy means to penetrate air defenses. When bombers are generated for nuclear operations, the share of survivable weapons increases.

Some analysts have expressed concern that if a conventional variant of LRSO were fielded simultaneously with a nuclear version, its use would be destabilizing.10 An adversary—Russia or China—detecting a bomber would not be able to discern whether it is carrying nuclear-armed or conventional cruise missiles. Russian or Chinese officials could then be tempted to assume the former and be more likely to unleash their own nuclear weapons. For now, official U.S. government sources plan for this weapon to be exclusively nuclear. The fiscal year 2020 National Defense Authorization Act repealed a prior congressional requirement for the Air Force to develop a conventional variant.11

As with the current nuclear air-launched cruise missiles, it is possible that Russia or China could mistake the radar signature of the future conventionally armed long-range air-launched JASSM-XR (Joint Air-to-Surface Standoff Missile–Extreme Range) cruise missile for that of a nuclear-armed LRSO. The United States has used such conventional cruise missiles numerous times in other conflicts in recent decades without triggering alarm in Russia or China. Of course, in a direct conflict with Russia or China, their militaries could mistake an incoming U.S. conventional long-range cruise missile for a nuclear-armed ALCM. However, given the JASSM and LRSO’s stealth capabilities, any ambiguity likely would derive from the use of bombers (particularly B-52s) that can and perhaps would carry nuclear weapons as well as conventional ones. Beyond avoiding war in the first place, this problem needs to be managed by direct military-to-military communication, arms control transparency measures to distinguish nuclear-capable from non-nuclear bombers, and bomber preparation (“generation”) as means to signal U.S. intentions. Chapter 6 covers these issues in greater detail.

The LRSO should be fully funded as an exclusively nuclear-armed weapon (as is the ALCM it would replace).

B61 Bomb

The B61 has two roles—a gravity bomb delivered by B-52H and B-2A bombers based in the United States, and a forward-based weapon for NATO dual-capable aircraft (DCA). A program is underway to improve the bomb’s capabilities, including accuracy and earth penetration, and consolidate most of its various forms into one type, the Mod 12. (One variant, the B61 Mod 11, a specialized earth-penetrating gravity bomb, will remain in service.) The B61 has a variable yield, including low-yield options.12

Within NATO, the main argument for keeping B61s is that they embody the alliance’s commitment to maintain nuclear deterrence. Indeed, the B61 is the only nuclear weapon under NATO’s nuclear-sharing arrangement. Any move to remove B61s could disrupt NATO unity and hearten Moscow. B61s are most controversial in Germany, where a domestic political debate has erupted over the continued presence of U.S. nuclear weapons in the country, triggered by Germany’s April 2020 decision to replace its aging DCA.13 NATO Secretary General Jens Stoltenberg has argued that the B61s should be maintained in the current basing locations and has sought to keep the NATO consensus on nuclear sharing.14

Other experts question the military utility of the B61. The forward-based weapons and DCA are based in five well-known locations, demated from nonalert aircraft. This arrangement leaves them vulnerable to attack—especially by short- or intermediate-range missiles from Kaliningrad or near the Eastern European border—if they have not been alerted and readied for use. It also may be difficult to deliver weapons to target against modern Russian air defenses. The NATO DCA mission is currently fulfilled by F-16s and PA-200s (Panavia Tornados), which were built in the 1970s and 1980s and do not have stealth capabilities. The addition of the F-35 as a NATO DCA will somewhat improve its military utility against modern air defenses but may not credibly guarantee NATO’s ability to rely on the B61 in a conflict. However, all DCA options utilize short-range fighters, raising questions regarding what targets they may be able to strike and adding the complication of aerial refueling to the DCA mission. This increases doubts about the B61’s utility and informs advocacy of the LRSO and/or the low-yield warhead for the submarine-based Trident D5 missile.

From a military perspective, this weapon is not optimal. It is a legacy of Cold War–era NATO deterrence planning. But the political and deterrence consequences of withdrawing the B61s to the continental United States without NATO’s concurrence would be more costly than the disarmament gain unless Russia significantly reduced its regional nuclear arsenal.

We recommend keeping the B61s forward-deployed under current nuclear sharing arrangements with European allies, until NATO requests its removal with reciprocal arms control or confidence-building steps by Russia.



Ballistic missile submarines will remain the mainstay of the U.S. nuclear deterrent for decades to come. The current Ohio-class submarines with Trident D5 SLBMs provide a reliable and survivable second-strike capability, with at least eight submarines at sea at all times.15 The first of the current operational SSBNs was deployed in 1984 and is planned to reach the end of its service life in 2027. The last Ohio is scheduled to retire in 2040. Both the Obama and Trump administrations decided to proceed with a replacement, the Columbia-class.16 The Columbia is intended to ensure the sea-based second-strike force for the foreseeable future.

SSBNs are vital because they are the most survivable basing mode for the strategic deterrent force. Ever since submarines began carrying ballistic missiles in the 1960s, they have become quieter and evolved techniques for avoiding detection.17 Moreover, with several SSBNs at sea on alert at all times, U.S. nuclear forces remain ready for any potential contingency.

The United States should take all steps necessary to ensure that the Columbia-class is introduced at the pace required to maintain the at-sea deterrent capability while the Ohio-class submarines are decommissioned.18 To hedge against future antisubmarine warfare breakthroughs, the Navy and relevant research organizations should prepare alternative operational practices.

Trident SLBMs

The Trident D5 SLBM is highly accurate—capable of holding at risk an adversary’s nuclear forces—and can penetrate missile defenses. These characteristics enable it to perform nuclear counterforce attacks, which are central to traditional U.S. damage-limitation plans. In the past decade, enhancements have bolstered the D5 W76 warheads’ capacity to kill hard targets.19

Tridents could substitute for vulnerable ICBMs as effective counterforce weapons. Depending on the launch location, they may have shorter flight times and, unlike ICBMs, varied trajectories. 20 Maintaining close proximity to an adversary’s territory, launching from unseen locations, and utilizing depressed flight trajectories would make D5s more capable of destroying targets such as silo-based missiles, missile launchers, and bombers before an adversary employs them in counterattacks.21 Because these submarine-launched weapons are invulnerable, unlike with silo ICBMs there would be no “use-them-or-lose-them” pressure to launch. The missiles offer yield flexibility: 90 kt (W76) or 475 kt (W88) warheads can be deployed on them. (As discussed further below, the much lower-yield W76-2 is also available for D5 now.)22 The Trump administration also announced it will pursue the W93 warhead, a “higher-yield” warhead that will help reduce the Navy’s “excessive reliance on less destructive W76s.”23 In contrast, the ICBM only deploys warheads with greater than 300-kt yields.24

If new strategic threats to the United States or its allies require additional deployed nuclear weapons to deter or defeat, the Trident D5 provides upload potential—that is, additional warheads can be added to each deployed missile.25 This could be done, for example, if New START (Strategic Arms Reduction Treaty) were to expire and Russia responds by deploying additional warheads, or if China decides to radically increase its number of nuclear warheads.

Finally, the D5’s long range—almost that of Minuteman ICBMs—enables it to be launched from locations far away from Russia’s and China’s improving coastal defenses and concentrated antisubmarine warfare capabilities. This range helps mitigate potential advances in adversary antisubmarine warfare and maintain survivability for the SSBN fleet.

A life extension program is underway to modernize the D5’s components and existing warheads.26 These activities should continue, and the Navy should continue the second life extension program—D5LEP2—to prolong the missiles’ service life past the 2040s.27

The arguments for the new W93 SLBM warhead are unclear, given the ongoing modernization of the W88 warhead. Further study of the rationale for the W93 is warranted before making a final decision on whether to add a third SLBM warhead to the U.S. stockpile.

Low-Yield Trident D5 (LYD5)

The 2018 NPR described a need to “supplement” the triad in order to respond to a potential limited nuclear attack by Russia. It then said that the United States would modify a “small number” of the W76 warheads designed for SLBMs to have a lower yield of approximately 5–7 kilotons.28 The weapons were deployed in late 2019 and are designated “W76-2.”29

The Low-Yield Trident D5 is intended to fill a perceived gap in U.S. nonstrategic nuclear weapons (NSNW), sometimes referred to as low-yield nuclear weapons, against Russia. Its advocates highlight two potential indicators of this gap. First, Russia has (and is modernizing) a large stockpile of NSNW. The exact number is not publicly known, but the 2018 NPR said it could be “up to 2,000.”30 Second, Russian military writings over the past two decades have suggested that attacking NATO targets with a small number of low-yield nuclear weapons could motivate NATO to stop an ongoing aggression into Russian territory.31

To counter these Russian battlefield weapons, and thereby deter or defeat Russian aggression, the United States has long retained limited nuclear options intended for theater use, including in Europe. These weapons usually had lower destructive yields than the weapons based on ICBMs and SSBNs. If Russia initiated use of nuclear weapons against NATO territory, NATO could respond in kind. The demonstrated resolve and capability to respond to a Russian limited nuclear attack, the theory posits, would deter Russia from escalating. If the scenario were different, and NATO were losing a conventional conflict with Russia and NATO leaders decided that only a nuclear strike would cause Russia to stop, using low-yield weapons would be the least escalatory way for NATO to cross this threshold. In both scenarios, U.S. and NATO leaders would seek to use types of nuclear weapons that they think would be least likely to cause Russian leaders to escalate a conflict to all-out nuclear war.

The United States possesses air-based NSNW suitable for limited nuclear war or regional deterrence. These include ALCMs from strategic bombers based in the continental United States and the B61 Mod-12 gravity bomb carried by NATO DCA.32 Proponents offer several arguments why the LYD5 is better than these alternatives.

First, unlike NATO-based aircraft, the D5 missile can deliver its warhead to target without a risk of interception by advanced Russian air defenses. The F-35, which will be certified in 2024 as a replacement DCA to carry the B61 bomb, will help redress (though not eliminate) NATO concerns about Russian air defenses.33 In the meantime, however, the B61 will be without a modern, forward-deployed, effective delivery system for several years.

Second, even before NATO DCAs could be launched, their bases could be attacked by Russian conventional or nuclear weapons. Indeed, these air bases and aircraft would be prime targets in an escalating conventional war, and their locations are publicly known.34 If Russia wanted to signal that NATO military actions had crossed its nuclear threshold, it would contemplate attacking such bases with low-yield nuclear weapons, as in its Cold War–era plans.35 Russia may legitimately view attacks on DCA bases in the early stages of a conventional war as justified, as the aircraft may be involved in conventional operations below the nuclear threshold.

Third, compared to an ALCM that would have to come from bomber bases in the continental United States, an SLBM provides a prompter option. Launch orders could be given, and targets could be hit within an hour by at-sea SSBNs.

Fourth, compared with air-delivered weapons, the use of D5 missiles would not require destructive attacks to “pave the way” through air defense systems if electronic countermeasures or cyber attacks could not disable or circumvent these systems. SEAD operations could require a substantial number of offensive long-range conventional cruise missile strikes on air defenses, radar installations, and command and control. These strikes could lead to a greater loss of life and render the adversary’s national territory more exposed to later attacks. The use of a single or limited number of LYD5s could help limit overall destruction and theoretically could better contain escalation than massed conventional strikes (alone or as precursors to DCA delivery of B61 nuclear bombs).

Lastly, offshore limited nuclear options are politically preferable to additional land-basing in Europe, which would be unpalatable to European allies. With the U.S. withdrawal from the INF Treaty in August 2019, NATO has explicitly stated its intention not to deploy land-based nuclear missiles.36 Submarine-based low-yield weapons provide greater flexibility and survivability, and do not exacerbate domestic political strife in NATO basing countries.

Offshore limited nuclear options are politically preferable to additional land-basing in Europe, which would be unpalatable to European allies.

Notwithstanding these arguments, formidable questions remain about the advisability of the low-yield submarine-launched warhead. The basic argument critics make is that a U.S. president would be more likely to use this weapon than he or she would be to use a much more destructive high-yield one. Critics worry that Russian leaders therefore would become more inclined to initiate nuclear strikes in order to beat U.S. leaders to the punch—perhaps perceiving that such actions would deter the United States from undertaking larger “damage-limiting” attacks on strategic forces to follow. The overall effect would be to “lower the threshold” of nuclear war. Though this argument must be taken seriously, it neglects the fact that NATO and the United States already deploy low-yield weapons for such scenarios. These weapons have not been particularly controversial within NATO or elsewhere. Moreover, deterrence of Russian aggression would be less effective if NATO and the United States had no options other than 100, 300, or 400-plus kt strategic weapons. Furthermore, in terms of international humanitarian law, the low-yield W76-2 should be preferred over the more indiscriminately destructive W76-1.

The LYD5 raises more specific questions that should have been answered before it was deployed and which were not adequately addressed by the Trump administration to Congress and the public. The answers to these questions should determine whether to retain it in the long term.

First, is it necessary for deterrence today or in the future?37 The State Department recently published a paper clarifying the U.S. government’s arguments in favor of the low-yield D5. The paper lists a series of attractive “distinct attributes” of the new weapon—promptness, survivability, and day-to-day presence.38 However, the United States has other means of delivering limited low-yield nuclear strikes against Russian, Chinese, and North Korean targets. The B-2A stealth bomber can penetrate advanced air defenses carrying B61 and B83 gravity bombs which provide multiple yield options. The B-2 also benefits from continual upgrades to ensure the aircraft can better detect and avoid ever-advancing air defenses.39 Further, the B-52H bomber can deliver the nuclear-armed ALCM, which the LRSO is planned to replace. As discussed above, the LRSO, which would be deployable on the B-52H, B-2A, and eventually the B-21, will significantly enhance the air leg’s ability to penetrate ever-advancing air defenses. In a crisis, bombers will likely be sortied and closer to potential targets, cutting down on the amount of time it would take for cruise missiles to reach their targets. However, the LRSO is not planned to be available until approximately 2030, which contributes to the case for developing more immediately available options, such as the LYD5.

Second, assuming that the United States would not plan to detonate 5–7 kt W76-2 warheads on allied territory, what circumstances and targets in Russia would require delivery of one or a few of these weapons at ballistic missile speed?40 Proponents have not, at least publicly, specified categories of such targets or employment plans for the W76-2. The “clarifying” State Department paper of April 2020 pointedly does not describe any possible targets. “Our strategy for deterring limited nuclear war,” it declares, “is not target-based; it is capability based.”41 Yet without a sense of likely targets, it is extremely difficult for officials and analysts in the United States and elsewhere to assess whether these weapons are more likely to stabilize or destabilize crises and escalate or deescalate nuclear exchanges.

Third, what are the probabilities that Russian early warning systems and personnel, and Russian decisionmakers, would accurately discern a detected incoming D5 missile as carrying only one warhead and that the warhead was a 5–7 kiloton W76-2, and not the 90 kiloton W76-1 or the 475 kiloton W88 warhead? The April 2020 State Department paper correctly notes that a “foe’s early warning system . . . would not be able to determine the yield of the weapons on a U.S. bomber or air-launched cruise missile either.”42 While valid, this statement is hardly reassuring given the real risks of inadvertent escalation.

Fourth, are there steps that the United States could take unilaterally, or explore with Russia, in advance or during a conflict, that would significantly mitigate risks of Russian misperception or mistake in assessing whether a LYD5 or high-yield weapon is being used? Here, too, the State Department acknowledges the issue. It has discussed this concern with Russian officials. But in the absence of meaningful strategic dialogue and revived arms control diplomacy, the best the State Department offers is that “there is no strategic rationale for an adversary . . . to launch a massive nuclear attack upon detection of a single SLBM, triggering the unlimited war it is trying to avoid.”43 This is reasonable and an improvement on the 2018 NPR’s case for the weapon, but the challenge remains.

Fifth, is there any basis for concluding that LYD5 use would not be escalatory—specifically, that Russian military and Kremlin leaders would not respond to attacks by LYD5 with further nuclear escalation?44 The point at which the State Department suggests a LYD5 may be useful—during an acute crisis where Russia may be “backed into a corner” and considering nuclear use—may instead trigger a large response by alert Russian nuclear weapons.45 If proponents for LYD5 conclude the risk of further escalation is low, what is the basis for this assertion, especially if the United States refuses to indicate targets for the weapon? If not, then what are the implications? How would U.S. decisionmakers and nuclear operators plan to keep escalation short of all-out nuclear war?46

Sixth, what is the probability that launching one or a few LYD5s would reveal the launching submarine’s location in ways that would enable Russian forces then to successfully attack the submarine and thereby remove up to 10 percent of the U.S. deployed strategic nuclear force?47 What probability of such an outcome would or should make U.S. decisionmakers eschew this option?48

Finally, does the lower yield of the W76-2 make it more likely for U.S. leaders to consider using nuclear weapons first in a conflict, or is it purely a retaliatory weapon given its basing on SSBNs and the U.S. military’s labeling of the weapon as enhancing “deterrence of limited nuclear use?”49 Russia and China may view the LYD5’s combination of lower-yield, promptness, and day-to-day readiness as supporting a first-use role, though this concern should be obviated by the low number of such weapons.

Obviously, answering these questions would involve extremely sensitive information and analysis, much of which understandably would need to remain secret. However, the questions raised here are vital. To date, notwithstanding the April 2020 State Department paper, there is no indication that relevant U.S. political leaders, including in Congress, have been briefed and comprehend these issues. Nevertheless, the W76-2 warhead is now deployed on Trident D5 missiles. This does not remove the need for answers to these questions. Without such analysis and rigorous debate over its conclusions, it is impossible to determine the balance of risks and benefits that this weapon entails, and how to minimize such risks. Thus, in 2021, the Department of Defense, the Department of State, and the intelligence community should conduct such analysis and briefings to better inform future decisionmaking about this weapon and others.

Pending conduct of such analysis and briefings to Congress and (as feasible) to the public, allies, and Russia, we support retaining the LYD5 at least until the nuclear variant of the F-35 is deployed and perhaps until the LRSO is also available. When these new low-yield weapons are deployed, U.S. policymakers should examine the continuing utility of the LYD5. Removing these warheads without clarifying how they are unnecessary, or whether other systems could better serve their purpose, could undermine NATO and Asian allies’ confidence in U.S. leadership and extended deterrence.50 Moreover, we see no gain in removing W76-2 warheads if they would be replaced by their unnecessarily higher-yield predecessors.


The previous U.S. nuclear-armed SLCM—the TLAM-N—was retired in 2010. At the time, the 2010 NPR described it as a redundant capability. Like the SLCM, bombers and nuclear-capable fighter aircraft can be forward deployed in times of crisis. Also, as with the SLCM, ICBMs and SLBMs provide the ability to strike any point on the globe.

The 2018 NPR proposed a new SLCM-N to provide a “non-strategic regional presence, an assured response capability, and an INF-Treaty compliant response” to Russia’s violation. Defense officials believe deploying SLCM-N on elusive attack submarines—which have only carried conventional cruise missiles for nearly three decades—will begin to shift the current U.S.-Russian balance in limited nuclear deterrence options to favor the United States.51 Yet the military role of this system still appears to be redundant. Other forward-deployed and deployable nonstrategic systems also provide a regional presence; these include the B61 bomb and ALCM-armed bombers, as well as the new LYD5. It is difficult to see the unique value of a SLCM-N in a regional crisis or any targets for a SLCM-N that a LYD5, ALCM, or B61 could not address equally well. Certainly, these distinguishing characteristics, if they exist, have not been explained publicly, and a series of questions regarding the intended delivery platform, targets, and capabilities of the missile have yet to be answered.52

Politically, some allies may see value in adding nuclear capability to SLCMs. Others may not. Stationing nuclear weapons on attack submarines or even surface ships could preclude some allies (for example, New Zealand and Japan) from allowing these vessels access to their ports.53 This would reduce the value of nuclear-armed SLCMs for signaling resolve to Russia, China, and North Korea, or for reassuring allies. Moreover, the utility of the SLCM-N as a regional deterrence tool depends on it being aboard forward-deployed naval assets, which would be impossible if the ships are not permitted access to allied ports. It would weaken alliance cohesion if the prospect of deploying such ships stimulated intense political discord in allied countries.

This weapon would have further downsides: deploying nuclear-armed cruise missiles on attack submarines could detract from these ships’ other missions, such as tracking enemy SSBNs and nuclear-armed surface ships; protecting U.S. carrier groups; and conducting conventional attacks on priority land targets such as air defenses, naval bases, and air bases. Policymakers have yet to clarify that adding regional nuclear strike missions would not undermine performance of attack submarines’ priority conventional missions, including antisubmarine warfare.

It has been suggested that the SLCM-N would be valuable as a potential bargaining chip for future Russian NSNW reductions—for instance, to reduce nuclear-armed antiship and land-attack SLCMs deployable on Russian naval vessels.54 However, there is no reason to suspect its introduction will not result in a Russian buildup as opposed to negotiated reduction. We do not advise prolonging the development program for this purpose.

The SLCM-N program is unnecessary and should be canceled.



The ground-based leg of the U.S. triad is composed of 400 ICBMs deployed in Colorado, Montana, Nebraska, North Dakota, and Wyoming.55 The Minuteman III missiles in these silos were deployed in 1970 as the first MIRV-capable ICBM, but today they carry only one warhead each. The service life for the oldest Minuteman ICBMs in the missile force will begin to end in 2029, at which point new Ground-Based Strategic Deterrent (GBSD) missiles are planned to begin replacing them. Procuring the missiles is now estimated to cost between $93 and $111 billion, with a decades-long life-cycle cost of $264 billion.56 These financial estimates are problematic insofar as only one company—Northrop Grumman—bid for the job and faced no competition. Given how costs of nuclear systems historically have exceeded initial estimates, this cost estimate is unrealistically low. Indeed, the costs were recently revised upward from 2017 estimates.57

The highly accurate missiles are on a day-to-day alert and can be launched within minutes of a missile crew receiving an order. Yet because the silos are in fixed, known locations, enemy intercontinental-range nuclear weapons (and potentially long-range precision-strike conventional weapons) can destroy them or the equipment required to operate them. Given a choice, few would construct a force structure with vulnerable silo-based ICBMs, as opposed to mobile ICBMs. Unfortunately, to be survivable, mobile ICBMs require expansive open land areas for the missile launchers to “roam” in order to minimize risks of being destroyed by Russian forces.58 Domestic politics have precluded the United States from deploying mobile ICBMs.

The primary targets for U.S. ICBMs are Russian ICBM bases, both fixed-silo and mobile. Destroying these targets would require a combination of high accuracy and/or yield—a hard target kill capability.59 ICBMs also could target air and naval bases and command and control facilities, but these installations are also susceptible to cruise missiles and SLBMs. The quick-launch capacity of the U.S. missiles and their accuracy make them especially relevant for targeting Russian silo-based ICBMs. The competition with Russia here is circular. Russian ICBMs are primary targets of massive first strikes by U.S. ICBMs. Russia then cites this threat from the U.S. ICBM force as the primary rationale for maintaining its own silo-based first-strike capability with multiple high-yield warheads and a LUA/launch on warning capability. (In the Russian military lexicon, this is known as “reciprocal counterstrike.”) Bureaucratic and military-industrial interests motivate both countries to continue investing heavily in these forces.

The combination of quick-launch, accuracy, and vulnerability suggests that ICBMs could conduct a preemptive (first) strike on opposing nuclear forces. The idea would be to limit the damage that Russia could inflict on the U.S. homeland. The lives that hopefully would be saved (compared to the alternative of not attacking Russia’s nuclear forces) would make such a massive first strike justifiable, affirming the value of ICBMs.60 The second and perhaps more likely scenario for employing ICBMs would be in rapid response upon detection of a Russian nuclear attack on these missiles. In this case, U.S. ICBMs would not target already-launched Russian ICBMs (mobile or silo-based), but rather other assets.

Recognizing the mutual danger and futility of these scenarios, the favored rationale today for silo-based ICBMs is that they deter Russia by requiring it to use so much of its nuclear arsenal to destroy them that Russia would not have enough left to counter U.S. submarine and air-based forces. For whatever number of ICBMs the United States deployed (400 today), Russia would need to expend more of its own warheads to try to destroy them.61 This math, plus the doubt that Russian weapons would in fact destroy U.S. ICBMs before they were launched, provides robust deterrence. However, as noted above, there are significant dangers and liabilities in retaining silo-based ICBMs and planning to launch them in a first strike or before Russia’s presumed incoming nuclear weapons can reach them.

A large-scale U.S. nuclear first strike to disarm Russia would be the riskiest and most cataclysmic attack ever made by any leader in wartime.

A large-scale U.S. nuclear first strike to disarm Russia would be the riskiest and most cataclysmic attack ever made by any leader in wartime. It would have to rely on exquisitely timed attacks on Russia’s early warning capabilities (by cyber, conventional, and/or nuclear means) to ensure that Russia could not launch its own ICBMs on warning. It also would require near-perfect intelligence of the locations of Russia’s mobile missile force. Both assumptions are dubious to say the least.

The second risk, as discussed above, stems from the vulnerability of ICBMs and the consequent plans to retain options to launch under attack (discussed in chapter 3). Technical malfunction or human error in detecting and assessing Russian attacks could lead the United States to launch its ICBMs unnecessarily or in disproportionate numbers, both of which could cause massive escalation. This risk is now compounded by the United States’ reliance on launch-detection and attack-assessment satellites and communications satellites that Russia (and China) could target by cyber or kinetic means in a conventional conflict, even if such attacks were not intending to threaten U.S. nuclear command and control. The United States could misperceive their intent or, in any case, interpret such an attack as a precursor to nuclear strikes, and act accordingly to prepare and/or launch U.S. nuclear forces.62 The ICBM force is more vulnerable to such destabilizing entanglement than other legs of the triad.

ICBMs also seem to be the least compliant strategic delivery system with respect to the law of armed conflict. The warheads currently deployed on U.S. ICBMs, the W78 and W87, both have yields of at least 300 kilotons. If these warheads detonate close to hard-to-kill targets like missile silos and buried command sites, the blasts would loft large amounts of radioactive debris into the atmosphere. Given the location of Russia’s ICBM bases, this is likely to create mass civilian casualties and have extremely pernicious long-term environmental and humanitarian consequences. None of these outcomes seem compatible with the law of armed conflict’s principles of military necessity, distinction, and limiting unnecessary suffering. Further, the high yields and intended targets for ICBMs give them little capability to address the regional deterrence scenarios that are most likely to engage U.S. nuclear forces in the foreseeable future. Use of an ICBM would signal a major escalation. This adds to the central deterrence value of the ICBM but also limits its utility in most likely conflict scenarios.

Possible Alternatives

If retaining silo-based ICBMs entails significant risks, what are the alternatives?

One option would be to unilaterally eliminate all or a significant fraction of the 400 Minuteman missiles and to rely more heavily on the more survivable SSBN and bomber forces. The 2013 review of U.S. nuclear employment strategy concluded that a reduction of deployed strategic warheads to approximately 1,000 to 1,100 would still allow the United States to fulfill current targeting requirements, whether or not Russia reciprocated.63 Though there is little reason to think that Russia would not remain deterred by a U.S. arsenal of 1,000 sea- and air-based weapons, political realities in the United States and the intensified confrontation between the United States, NATO, and Russia make this option infeasible and probably unwise. Even debate over such a unilateral move could jeopardize tenuous bipartisan congressional support for a linked approach to nuclear modernization and arms control negotiations and exacerbate partisan conflict and stalemate in Washington.

Another alternative would be for the United States to eliminate as many ICBMs as could be substituted for by adding warheads to the bomber and submarine forces while keeping under New START’s warhead ceilings (figure 4 illustrates the hypothetical upload capabilities of the air and sea legs as Minuteman missiles are slowly phased out).64 The downside would be that this force structure would rely more heavily on the eight or nine submarines on patrol at any given moment. As Russian and Chinese antisubmarine capabilities improve, the theoretical risk to a much larger portion of the United States’ deployed nuclear deterrent could grow too. These risks could be mitigated as described above, and the deployment of the Columbia SSBN replacement will also improve the survivability and viability of the submarine force far into the future.

Some express concern that the two SSBN bases—in Kings Bay, Georgia, and Bremerton, Washington—could be destroyed by an adversary with only a small number of nuclear or conventional munitions. (These sites are defended by air and missile defenses, though that may be of little comfort to U.S. nuclear strategists.) Thus, if the United States reduced its ICBM force, a Russian disarming first strike could plausibly threaten the U.S. nuclear deterrent.65 However, this scenario is unlikely for two reasons. First, attacking the two submarine bases would still leave the United States with a large portion of the SSBN fleet at sea when the attack is carried out. (Eight submarines at sea could carry approximately 1,000 nuclear warheads, if SLBMs carry the additional warheads suggested here.) Second, in terms of deterrence, Russian experts know that if they attacked a U.S. SSBN base (including with non-nuclear weapons) the United States would take it as an extreme form of escalation and could respond accordingly. As the 2018 NPR states, “Significant non-nuclear strategic attacks include, but are not limited to . . . attacks on U.S. or allied nuclear forces.”66 If the United States reduces its ICBM capacity in favor of SSBNs, the U.S. Navy would be expected to enhance the survivability of the base infrastructure and nondeployed SSBNs at Kings Bay and Kitsap.

The better and more politically feasible option is to pause the GBSD program, and meanwhile retain (and update) the Minuteman ICBM force while seeking negotiation of an agreement with Russia to mutually reduce the total number of warheads and silo-based nuclear missiles and launchers.67

The Minuteman remains effective and launch-ready. If the main value of vulnerable ICBMs is to soak up a large portion of Russian nuclear forces, an expensive new system is not necessary. The U.S. Air Force can maintain a sizable Minuteman arsenal for a longer period by reducing the number of deployed ICBMs. If the deployed ICBM force were 300 or 200 missiles instead of 400, it is difficult to imagine Russian leaders becoming less deterred from attempting a disarming first strike against the United States. Missile bodies and parts from withdrawn missiles can be used to keep the retained ones operational.68 The feasibility of extending the life of Minuteman III is explored in Appendix A.

Reducing the size of the U.S. ICBM force may be necessary regardless of the GBSD deployment. The Air Force plans to deploy the B-21 in greater numbers than the aircraft it is intended to replace, the B-1B and (eventually) the B-2. If a new administration intends to extend New START for five years, as recommended here, B-21 bombers likely will be deployed while the treaty remains in force. (The B-21, along with other major nuclear modernization programs will certainly be relevant to negotiations after New START expires, even if that date is in 2026—see figure 5.) Thus, more “room” will need to be created in the U.S. nuclear force to comply with numerical ceilings.

At a time of growing competition among defense programs, with Republicans and Democrats alike calling for ruthless or “smart” spending prioritization, the U.S. nuclear program must come to grips with the archaic deterrence logic that underpins the land-based leg of the strategic triad, as well as the arms control opportunities it may present. As U.S. nuclear modernization programs enter an expensive, technically difficult stretch, these considerations should not be overlooked.69

Nuclear Command, Control, and Communications (NC3)

Nuclear weapons are only as reliable (physically) as the command, control, and communication systems that inform and implement the decisions to employ or not employ them. These systems (known as NC3) are supposed to:70

  • guarantee effective monitoring and exclusive control at all times over all nuclear forces and strategic operations;
  • support decisionmaking, planning, and operations in all scenarios;
  • provide timely warning of imminent attack;
  • supply situational awareness to the various command levels;
  • assure effective and secure communications to and from national command authority;
  • accommodate and support required maintenance, upgrade, safety and surety operations;
  • withstand efforts to undermine or subvert the reliable transmission of information and guidance between and across command levels; and
  • sustain high standards of safety, security, and secrecy commensurate with the sensitivity of nuclear weapons.

The U.S. NC3 system includes space- and terrestrial-based sensors and communications platforms, as well as the computer architecture and other hardware that make these platforms work. The Pentagon currently plans to modernize the NC3 system wholesale. The Congressional Budget Office estimates that this effort will cost $77 billion from 2019 to 2028.71

NC3 modernization is receiving deserved attention for several reasons. Existing hardware components are numerous and of various historical and technical vintages, which makes it difficult for operators and commanders to understand how the systems work and what their vulnerabilities might be.72 New threats are rapidly emerging, particularly in the form of cyber espionage and potential attack, as well as innovative antisatellite weaponry.73 And new technological capabilities provide opportunities to network systems and integrate military operations across domains of combat from cyber to conventional to nuclear.

The first imperative is to make NC3 secure and resilient against attacks that could destroy, degrade, or corrupt its functionality.74 NC3’s vulnerability to advanced cyber, space, long-range precision-strike weapons, and advanced autonomous weapons systems is an emerging concern. Competition among nuclear-armed states in these areas and the targeting of NC3 heightens risk of nuclear use.75

Other, less obvious risks also must be addressed. Command, control, and communication systems that are shared across warfighting domains—for example, by nuclear and conventional precision-strike forces and commands—could exacerbate risks of inadvertent escalation. Adversary attacks intended as part of conventional war could affect U.S. nuclear forces and NC3 as well.76 U.S. military and political leaders would then face the challenge of interpreting adversary intentions and deciding whether nuclear strikes are imminent or already underway, and how to respond, in the midst of an ongoing conflict. Opportunities for mistakes abound. (Russia and China may face similar risks to the extent that they too have entangled NC3 systems.)77

Improving the functionality, security, and resilience of NC3 and understanding how to address possible modes of failure are noncontroversial priorities. In doing this, policymakers also should be expected to recognize that certain force postures place more stress on different elements of NC3.78 For example, the maintenance of an alert, silo-based ICBM force creates compressed time for presidential decisionmaking. U.S. nuclear forces and operational planning centered on survivable nuclear forces may increase decision time and reduce burdens on the NC3 system.

Deterrence also can help reduce or channel threats to NC3. The 2018 NPR and subsequent remarks by U.S. officials warned that “any harmful interference with or attacks upon such components of our space architecture at any time, even if undertaken only with non-nuclear tools, thus starts to move into ‘significant non-nuclear strategic attack’ territory, and would lead to a significant and potentially drastic escalation of a crisis or conflict.”79 Therefore, an adversary’s attack on space-based command and control hardware—whatever its purpose—may, in the view of the U.S. government, warrant nuclear retaliation.

The Pentagon has recently recognized the urgency of building a resilient and adaptable NC3 system and has put USSTRATCOM in the lead of the NC3 modernization program. USSTRATCOM is well positioned to determine the best course of action for NC3 modernization within the broader context of U.S. nuclear modernization and changes to posture and policy. This report cannot credibly describe how NC3 modernization should proceed. Doing so will require extensive classified review and analysis. However, several basic principles should guide this process:

  • Modernization must carefully navigate the inevitable trade-off between redundancy and cutting-edge capability. In light of the growing threats to NC3, the United States should prioritize building a more resilient if less technologically capable NC3 architecture.
  • Resilience and redundancy will be necessary. For example, communication/early-warning payloads hosted on other constellations of U.S. military satellites could be deployed as alternatives or supplements to a small number of highly capable dedicated satellites. The Pentagon should consider developing survivable emergency backup capabilities such as high altitude, long endurance UAVs (unmanned aerial vehicles).
  • Traditionally, U.S. NC3 assets have been designed to survive for as long as possible in a nuclear war. For example, they have been hardened against electromagnetic pulse attacks, at great financial costs. However, U.S. NC3 may not be sufficiently resilient today to emerging kinds of non-nuclear attacks, such as the jamming or spoofing of communications satellites or “dazzling” through ground-based lasers.80 An adversary could use such techniques at the outset of a conventional conflict. Yet, in attempting to disrupt U.S. conventional military operations, the adversary could disrupt co-located nuclear command and control functions, which could increase the risk of escalation. To better manage such risk, the NC3 systems must be able to survive a broader range of non-nuclear attacks to ensure the viability of U.S. nuclear forces during an ongoing and potentially escalating conflict.


1 These rivalries originated in the late 1940s when the Navy and Air Force competed vigorously for a larger share of the defense budget. See Anand Toprani, “‘Our Efforts Have Degenerated Into a Competition for Dollars’: The ‘Revolt of the Admirals’, NSC-68, and the Political Economy of the Cold War,” Diplomacy and Statecraft 30, no. 4 (2019): 681–706; and Darius Watson, “Rethinking the U.S. Nuclear Triad,” Strategic Studies Quarterly 11, no. 4 (2017): 138–40.

2 In the late 1950s, the Navy then looked to wrest the atomic monopoly away from the Air Force and Strategic Air Command with the Polaris SSBN. In response, the Air Force designed “counterforce”: an attempt to utilize the greater accuracy of bombers compared to submarine-launched missiles (which at the time, were accurate enough to be aimed at cities but not at nuclear forces). The Air Force’s bureaucratic strategy was to suggest a more politically palatable counterforce “limited” nuclear strike that would spare an adversary’s cities; see discussion in Kaplan, The Bomb, 25–36.

3 Patrick Tucker and Marcus Weisgerber, “A New Nuclear Warhead? STRATCOM Chief Can’t Answer Yes or No,” Defense One, February 27, 2020,

4 Valerie Insinna, “US Air Force Nears Battle Over Next B-52 Engine,” Defense News, September 15, 2019,

5 “It should be understood that BAAD missions are simply one part of a multi-faceted assurance effort. By focusing on known BAAD mission timing, this analysis makes every attempt to study the effect of the missions themselves, with as little influence as possible from other extraneous variables.” Todd Robinson and Al Marouni, “Evaluating the Effectiveness of Assurance and Deterrence Missions: AY17 Strategic Deterrence Research Papers,” U.S. Air Force Center for Unconventional Weapons Studies, September 2017,

6 Brian W. Everstine, “USAF’s Goal of 220 Bombers a ‘Living Number,’ Can Evolve as B-21 Comes Online,” Air Force Magazine, September 15, 2020,

7 Roxana Tiron and Tony Capaccio, “Pentagon Seeks $10.3 Billion to Buy the Stealthy B-21 Raider,” Bloomberg Government, February 10, 2020,

8 Consider the Strategic Air Command’s bomber operations during the Cuban Missile Crisis, when deployed B-47 bombers were dispersed to more than thirty airfields to make them less vulnerable to attack, including at civilian airfields. Stephanie Ritter, “SAC During the 13 Days of the Cuban Missile Crisis,” Air Force Global Strike Command Air Forces Strategic-Air, October 19, 2012,

9 For example, see the number of air bases accountable under the START Treaty. “Strategic Arms Reduction Treaty (START) Inspectable Sites in the United States,” Federation of American Scientists, n.d.,

10 Andy Weber and Christine Parthemore, “Cruise Control: The Logical Next Step in Nuclear Arms Control?,” Journal for Peace and Disarmament 2, no. 2 (2019): 453–67, DOI: 10.1080/25751654.2019.1681886.

11 Joseph Trevithick, “Congress Poised to Cancel Non-Nuclear Version if Air Force’s Future Stealth Cruise Missile,” The Drive, December 12, 2019,

12 Hans M. Kristensen, “Video Shows Earth-Penetrating Capability of B61-12 Nuclear Bomb,” Federation of American Scientists, January 14, 2016,

13 Oliver Meier, “German Politicians Renew Nuclear Basing Debate,” Arms Control Association, June 2020,

14 Sebastian Sprenger, “NATO Chief Backs Germany’s Vow to Keep War-Ready US Nukes,” Defense News, May 11, 2020,

15 Hans M. Kristensen, “US SSBN Patrols Steady, but Mysterious Reduction in Pacific in 2017,” Federation of American Scientists, May 24, 2019,

16 “Navy Columbia (SSBN-826) Class Ballistic Missile Submarine Program: Background and Issues for Congress,” R41129 (Washington, DC: Congressional Research Service, October 7, 2020),

17 For example, computer-controlled machinery, unavailable in the 1960s, is now used in the manufacture of submarine propulsion components and can achieve a standard of precision that reduces noise. Kyle Mizokami, “What Makes Submarines So Quiet,” Popular Mechanics, August 15, 2017, The Columbia-class will also benefit from an electric-drive, a first for U.S. SSBNs: “The Columbia class is to be equipped with an electric-drive propulsion train, as opposed to the mechanical-drive propulsion train used on other Navy submarines. The electric-drive system is expected to be quieter (i.e., stealthier) than a mechanical-drive system.” See “Navy Columbia (SSBN-826) Class Ballistic Missile Submarine Program.”

18 The head of U.S. Strategic Command, Admiral Charles A. Richard, warns that “there is no margin to extend the OHIO-class further; therefore, the COLUMBIA-class SSBN must field on time to avoid a capability gap in the triad.” Other senior military officials have similarly stated there is little margin for error in producing the Columbia SSBN on its current schedule: “the bottom line is there’s no additional margin for construction and delivery of Columbia.” See “Statement of Charles A. Richard, Commander United States Strategic Command, Before the Senate Committee on Armed Services 13 February 2020,” Senate Committee on Armed Services, February 13, 2020,; and Sydney J. Freedberg Jr., “Inside America’s Aging Nuclear Missile Submarines,” Breaking Defense, July 16, 2018,

19 See Hans Kristensen, Matthew McKinzie, and Theodore Postol, “How US Nuclear Force Modernization Is Undermining Strategic Stability: The Burst-Height Compensating Super-Fuze,” Bulletin of the Atomic Scientists, March 1, 2017,

20 The Trident II SLBM’s capabilities have been compared favorably to the Peacekeeper MX. See “Triad Hearing Follow-Up,” GAO/PEMD-93-28R, Government Accountability Office, August 6, 1993, p. 6, “We found the D-5 Mark 5 SLBM to be equivalent to the Peacekeeper ICBM on the measure of capability to destroy hard targets. In contrast, MM III is considerably less accurate than either the Peacekeeper or D-S/Mark 5, and hence would be much less lethal against hard targets. On other measures, such as communications speed and system reliability, MM III is essentially the equal of the two other systems, while sharing the same deficiency as Peacekeeper with regard to easy locatability, given its basing mode in fixed silos. In sum, the D-5 with Mark 5 (W88) warheads would be considerably more capable with regard to hard targets than the MM III.”

21 Lisbeth Gronlund and David C. Wright, “Depressed Trajectory SLBMs: A Technical Evaluation and Arms Control Possibilities,” Science and Global Security 3 (1992): 101–59, See also Lieber and Press, “The New Era of Counterforce,” 24n44 (and more broadly Lieber and Press’ discussion of the SLBM’s counterforce capabilities).

22 A life extension for the W76 (now called the W76-1) was recently completed, and the W88 is undergoing a life extension now. The Department of Energy and U.S. Navy recently announced the new W93 warhead for the Trident II, which likely will be deployed by 2040. Aaron Mehta, “Inside America’s Newly Revealed Nuclear Ballistic Missile Warhead of the Future,” Defense News, February 24, 2020,

23 John M. Donnelly, “Trump Team’s Case for New Nuke Cites Risks in Current Arsenal,” Roll Call, July 29, 2020,

24 Robert S. Norris and Hans M. Kristensen, “U.S. Nuclear Warheads, 1945–2009,” Bulletin of the Atomic Scientists 65, no. 4 (2009): 72–81, DOI: 10.2968/065004008.

25 The Trident D5 can carry eight W88 or twelve W88 warheads, but the United States limits the number of warheads on each missile to meet the limits of New START (approximately four warheads per deployed Trident D5). See Congressional Budget Office, “The Potential Costs of Expanding U.S. Strategic Nuclear Forces if the New START Treaty Expires,” August 2020,, 10.

26 Amy F. Woolf, “Defense Primer: Strategic Nuclear Forces,” IF10519, Congressional Research Service (updated January 10, 2020),

27 Statement of Vice Admiral Johnny Wolfe, “FY 2021 Budget Request for Nuclear Forces and Atomic Energy Defense Activities,” March 3, 2020, p. 5,

28 Hans M. Kristensen and Matt Korda, “United States Nuclear Forces, 2019,” Bulletin of the Atomic Scientists 75, no. 3 (2019): 123.

29 John Rood, “Statement on the Fielding of the W76-2 Low-Yield Submarine Launched Ballistic Missile Warhead. Department of Defense,” U.S. Department of Defense, February 4, 2020,

30 2018 NPR, 53. However, in 2011, then principal deputy undersecretary of defense for policy James Miller told the House Subcommittee on Strategic Forces that Russia had 2,000 to 4,000 such weapons. See “The Current Status and Future Direction for U.S. Nuclear Weapons Policy and Posture, Hearing Before the Subcommittee on Strategic Forces of the Committee on Armed Services, House of Representatives,” November 2, 2011, There has been no public U.S. government statement about this discrepancy, which suggests that Russia has actually reduced its nonstrategic nuclear warheads in the past decade.

31 Olya Oliker, “Moscow’s Nuclear Enigma,” Foreign Affairs, November/December 2018,

32 The term “mod” designates a different version of the bomb; here, mod-12 designates the latest B61 following a life extension program.

33 Stephen Carlson, “Lockheed Martin Contracted for F-35 Flight Testing and Nuclear Capability,” Space Daily, November 16, 2018,

34 Kristensen and Korda, “United States Nuclear Forces, 2020,” 124.

35 According to some Soviet war plans, NATO airfields would be targeted with tactical nuclear weapons once a NATO–Warsaw Pact conflict began: “The Warsaw Pact would have used many more smaller ‘tactical’ nukes against NATO command posts, army bases, airfields, equipment depots and missile and communications sites.” See “This Is How the World Could Have Ended,” War Is Boring (Medium), January 14, 2014,

36 See the statement by Jens Stoltenberg at NATO Defence Ministerial: “Defence Ministers confirmed that NATO had no intention to deploy new land-based nuclear missiles in Europe, and did not want a new arms race.” See “NATO and the INF Treaty,” NATO, August 2, 2019,

37 As mentioned earlier, the United States would need to use advanced conventional strike options where possible to better adhere to the law of armed conflict. The LYD5 is likely designed for targets where conventional weapons have little utility; this paper does not contest the assignment of nuclear or conventional weapons to unknown targets.

38 See Bureau of Arms Control, Verification and Compliance, “Strengthening Deterrence and Reducing Nuclear Risks.”

39 Kris Osborn, “Air Force Upgrades B-2 Stealth Bomber as Modern Air Defenses Advance,”, April 24, 2015,

40 For an extensive discussion of AirLand Battle and potential uses of tactical nuclear weapons, see Douglas W. Skinner, “Airland Battle Doctrine,” Center for Naval Analyses, September 1988, Though NATO countries worried about the possibility of nuclear use on friendly territory, the doctrine clarified the issue: “It is important to note, however, that tactical use of nuclear weapons is generally associated with the deep battle far away from friendly territory” (p. 17).

41 Bureau of Arms Control, Verification and Compliance, “Strengthening Deterrence and Reducing Nuclear Risks,” 3.

42 Bureau of Arms Control, Verification and Compliance, “Strengthening Deterrence and Reducing Nuclear Risks,” 5.

43 Bureau of Arms Control, Verification and Compliance, “Strengthening Deterrence and Reducing Nuclear Risks,” 6.

44 For instance, have U.S. policymakers closely examined Russian military doctrine which considers any WMD use on Russian territory as warranting a nuclear response? Dara Massicot, “Anticipating a New Russian Military Doctrine in 2020: What It Might Contain and Why It Matters,” War on the Rocks, September 9, 2019,

45 See Konstantin Bogdanov, “Not-so-Nuclear War,” Russian International Affairs Council, March 10, 2020,

46 Then USSTRATCOM commander General John Hyten stated of the LYD5 in February 2018, “I don’t see it as escalatory at all I see it as that is a logical extension of our capabilities that we should have as warfighters to present to the President of the United States.” General John Hyten, “Nuclear Posture Review Discussion at National Defense University,” USSTRATCOM, February 16, 2018,

47 Austin Long, “Discrimination Details Matter: Clarifying an Argument About Low-Yield Nuclear Warheads,” War on the Rocks, February 16, 2018,

48 As some experts have assessed, the detection of launch, tracking of underwater targets, and utilization of prompt antisubmarine (possibly nuclear-armed) weapons are complex technical problems that must be overcome to destroy an SSBN that has launched an LYD5, and undoubtedly stealthily moved several miles away from the launch point. See Austin Long, “Location, Location, Location: Evaluating Risks to Submarines From Low-Yield Warhead and Submarine Missile Launch Detection,” Lawfare, March 11, 2018,

49 Hyten, “Nuclear Posture Review Discussion at National Defense University.”

50 It is notable that U.S. allies in Europe or Asia did not raise concerns regarding the LYD5 deployment.

51 Aaron Mehta, “The US Navy’s New Nuclear Cruise Missile Starts Getting Real Next Year,” Defense News, February 21, 2020,

52 Mehta, “The U.S. Navy’s New Nuclear Cruise Missile.”

53 See Michael Pugh, “Nuclear Warship Visiting: Storms in Ports,” The World Today 45 no. 10 (1989): 180–83,

54 Aaron Mehta, “Will the US Trade Its New Sub-Launched Cruise Missile for Russian Arms Treaty Compliance?,” Defense News, February 6, 2018,

55 Under the New START force structure, 400 ICBMs is the maximum deployed by the United States. And additional 50 silos are empty but kept in “warm” status, which would allow 450 ICBMs to be deployed if desired. See “Fact Sheet on U.S. Nuclear Force Structure Under the New START Treaty,” n.d.,

56 Anthony Capaccio, “U.S. ICBM to Replace 1970s Minuteman May Cost $111 Billion,” Bloomberg, October 1, 2020,

57 Kingston Reif, “New ICBM Replacement Cost Revealed,” Arms Control Today, March 2017,

58 The history of U.S. administrations’ efforts to win approval for mobile ICBMs should disabuse anyone of the political feasibility of this approach. See John T. Correll, “Peacekeeper by Fits and Starts,” Air Force Magazine, February 26, 2019,

59 Matthew Kroenig, “The Case for the US ICBM Force,” Strategic Studies Quarterly 12, no. 3 (2018), 61–62,

60 Kroenig, “The Case for the US ICBM Force,” 60–61.

61 Of course, the possessors of silo and mobile ICBMs also use hardening, camouflage, dummy targets, and other operational practices to increase the chance that incoming warheads may malfunction or miss the target.

62 Acton, “Escalation Through Entanglement.”

63 “Report on Nuclear Employment Strategy of the United States”; and Hans M. Kristensen, “New Nuclear Weapons Employment Guidance Puts Obama’s Fingerprint on Nuclear Weapons Policy and Strategy,” Federation of American Scientists, June 20, 2013,

64 See Kristensen and Korda, “United States Nuclear Forces, 2020”: “Each Trident SLBM can carry up to eight nuclear warheads, but normally carry an average of four or five warheads, for an average load-out of approximately 90 warheads per submarine.”

65 Dennis Evans and Jonathan Schwalbe, “Intercontinental Ballistic Missiles and Their Role in Future Nuclear Forces,” Johns Hopkins Applied Physics Laboratory, 2017, p. 11,

66 2018 NPR, p. 21.

67 Some experts suggest pursuing a conventional ICBM—something the United States considered in the 2000s as part of the Conventional Prompt Global Strike program—to reduce the destabilizing disarming first-strike dangers posed by vulnerable nuclear U.S. silo ICBMs. There may be some merit to this approach: the United States may benefit from a long-range conventional strike option that does not rely on expensive, potentially vulnerable, and limited-in-number air and sea platforms; a conventionally armed ICBM is also more compliant with the law of armed conflict. However, the efficacy of such systems at attacking an enemy’s nuclear forces, even with conventional warheads, means the change from nuclear to conventional ICBMs is unlikely to reduce first strike stability concerns that the land-based leg of the triad creates in Moscow, Beijing, or Pyongyang. Scholars outside of government have suggested several novel ideas to retain the Minuteman for years beyond the end of its projected lifespan, including adjustments to flight test frequency and maintenance practices. See Steve Fetter and Kingston Reif, “A Cheaper Nuclear Sponge,” War on the Rocks, October 18, 2019,

68 Nick Adde, “Minuteman III Replacement Program Moves Toward Next Phase,” National Defense, November 2, 2018,; and Fetter and Reif, “A Cheaper Nuclear Sponge.”

69 Elbridge Colby, Mackenzie Eaglen, and Roger Zakheim, “How to Trim the Defense Budget Without Harming U.S. Security,” Foreign Policy, September 30, 2020,; and Rachel S. Cohen. “Smith: Coronavirus Should Spur Fiscal Responsibility, Defense Industry Action,” Air Force Magazine, April 7, 2020,

70 John R. Harvey, “U.S. Nuclear Command and Control for the 21st Century,” NAPSNet Special Reports, May 24, 2019,

71 This is a $19 billion increase of the ten-year estimate from 2017. Congressional Budget Office, “Projected Costs of U.S. Nuclear Forces, 2019 to 2028,” January 2019,

72 See, for background, Peter Hayes, “Nuclear Command, Control and Communications (NC3): Is There a Ghost in the Machine?” (PowerPoint presentation, April 9, 2018),

73 See Defense Intelligence Agency, “Challenges to Security in Space,” Defense Intelligence Agency, January 2019,; and Jon Lindsay, “Cyber Operations and Nuclear Weapons,” NAPSNet Special Reports, June 20, 2019,

74 Creating wholly separate conventional and nuclear command, control, and communications hardware is financially and operationally infeasible, and current NC3 modernization plans will continue the practice of integrated NC3 for the foreseeable future. As a result, the United States must look to make the current NC3 system, with modern hardware and updated practices, address growing entanglement risks.

75 James N. Miller Jr. and Richard Fontaine, “A New Era in U.S.-Russian Strategic Stability: How Changing Geopolitics and Emerging Technologies Are Reshaping Pathways to Crisis and Conflict,” Center for a New American Security, 2017, p. 5,

76 Acton, “Escalation Through Entanglement.”

77 See for further discussion, Leonid Ryabikhin, “Russia’s NC3 and Early Warning Systems,” NAPSNet Special Reports, July 11, 2019,

78 For instance, LUA increases requirements for timely, secure communications with the president.

79 Ankit Panda, “Space-Based Nuclear Command and Control and the ‘Non-Nuclear Strategic Attack,’” Diplomat, April 8, 2020,

80 Defense Intelligence Agency, “Challenges to Security in Space”; and