“For half a century, ACA has been providing the world … with advocacy, analysis, and awareness on some of the most critical topics of international peace and security, including on how to achieve our common, shared goal of a world free of nuclear weapons.”

– Izumi Nakamitsu
UN High Representative for Disarmament Affairs
June 2, 2022

Human Augmentation and Nuclear Risk: The Value of a Few Seconds

March 2022
By Marina Favaro and Elke Schwarz

Nearly 30 years ago, the fate of humankind hung in the balance when a Soviet Union early-warning system indicated that a series of U.S. intercontinental ballistic missiles (ICBMs) were headed toward Soviet soil. The alert came with a “high reliability” label. At the height of the Cold War, Soviet doctrine prescribed that a report of incoming U.S. missiles would be met with full nuclear retaliation—there would be no time for double checking, let alone for negotiations with the United States.

U.S. soldier tests out a brain-computer interface device as part of a U.S. program to expand the use of technology to augment the performance of military forces. (Photo by U.S. Army)The officer on duty that day, Soviet Lieutenant Colonel Stanislav Petrov, had a quick decision to make. Reporting the incoming strike flagged by the system would result in a nuclear strike by the Soviet forces; not reporting it could risk making an error that would prove devastating for the Soviet Union. After a moment of consideration, Petrov went with his gut feeling and concluded that the likelihood of a system error was too great to risk a full-scale nuclear war. Indeed, the system had misidentified sunlight reflected from clouds for missiles. The worst had been avoided.

That incident, now a well-known chapter in nuclear weapons history, occurred in 1983, when the pace of weapons technology advancement was comparatively slow, when aspirations to accelerate decision-making through real-time computational technologies were not yet within reach, and when advances in military human enhancement were comparatively limited. The current military-operational context is very different.

Today, renewed great-power competition is being intensified by technological advancements that allow for the real-time relay of information that requires decisions be made within seconds. Human beings are embedded much more intricately into the military technological systems logic, as operator and functional element, in the pursuit of speed and optimization. In order to function within a highly scientific, technologically sophisticated conflict environment, those involved in the action chain, including operators and fighters, are themselves in need of a tune-up.

Indeed, military human enhancement is one of the new frontiers in emerging weapons technology, as advanced militaries across the globe are planning to enhance and augment the capabilities of their war-fighting forces. This development is shifting the parameters of decision-making, including nuclear decision-making. What, for example, might have happened if Petrov had been more intricately woven into the computer system that reported the erroneous satellite signal, perhaps via an implantable neural interface to facilitate speedier human-machine communication for more efficient decision-making? Notwithstanding its operational and strategic importance, human augmentation is not typically discussed in nuclear policy. This oversight needs redress as ministries of defense begin to focus on human enhancement as the “missing part” of the human-machine teaming puzzle.1

Definition and History of Human Augmentation

Human augmentation is a vast field with many linkages to other areas of study. There is no commonly agreed definition, and it is known by many names, which are often used interchangeably. “Augmentation” usually refers to the transformation of capabilities to include a new or additional capability, but “enhancement” refers to the fortification of existing capabilities. Both concepts can be broadly defined as “the application of science and technologies to temporarily or permanently improve human performance.”2 As with all emerging technologies, human enhancement technologies exemplify aspects of continuity and change.

The development of physically and mentally resilient soldiers has a long history, involving techniques and technologies that work toward fortifying the human body and mind with the objective of extending capacities and limiting vulnerabilities in war. Such technological transformations begin with straightforward tools such as a soldier’s armor, the crossbow, the machine gun, the rocket launcher, and a range of natural and synthetic substances with pharmacological effects. Roman and Greek legionnaires strengthened their bodies with leather and bronze and their resolve with wine, beer, rum, and brandy. Opioids and amphetamines have long been used in battle to gain a greater edge in fighting.3 Pain relievers and other synthetic drugs are instrumental in alleviating pain and facilitating healing. Meanwhile, propaganda, systematic training, and enemy dehumanization are used to augment or suppress the war-fighter’s emotions. Human augmentation programs today extend much further into the development of the soldier as fighter, as well as the solider as operator, through various modes of scientific-technological inscriptions, shaping bodies toward greater operational efficiency and effectiveness.

As human and machine become increasingly entwined, the concept of the “super soldier,” where human tissue and technological circuitry fuse for maximum performance, is taking shape. In May 2021, the UK Ministry of Defence and the German Bundeswehr co-published a report that conceptualizes “the person as a platform” and heralds the “coming of the Biotech age.”4 Half a year earlier, China and France published reports indicating their readiness to augment military personnel physically, cognitively, psychologically, and pharmacologically.5,6 In the United States, the Defense Advanced Projects Research Agency has been investing in neurotechnology since the 1970s and today is expanding the frontiers of the field, with a focus on neural interface technology.7

Current trends in military human enhancement focus on external enhancements such as augmented reality, exoskeletons, wearables, and biosensors and internal augmentations through pharmacological supplements. Implantable chips for a medical or curative purpose are already on the horizon.8 Directly enhancing human capabilities, however, is only half of the equation. The other half is that human augmentation will become increasingly relevant to security and defense because it is the binding agent between humans and machines.9 Beyond these external and internal enhancements, there is considerable interest in developing technologies that facilitate smoother and more functional teaming between the human and computational systems through so-called neural interface technologies.

The importance of effective integration of humans and machines is widely acknowledged, but has been primarily viewed from a technocentric perspective. Many of the existing solutions are technology focused, such as “building trust into the system” by making artificial intelligence (AI) more transparent, explainable, and reliable. Although this is necessary for cultivating trust in human-machine teams, it does not account for the human element in the teaming equation. Proponents of human augmentation argue that it is the necessary adjustment for a world in which future wars will be won by those who can most effectively integrate the capabilities of personnel and machines at the appropriate time, place, and location.10

As militaries increasingly incorporate automated and autonomous processes into their operations, brain-interface technologies could serve as a crucial element in future human-machine teaming.11 Brain-interface technologies offer methods and systems for providing a direct communication pathway between an enhanced, or wired, brain and an external device.12 In other words, they enable the transfer of data between the human brain and the digital world via a neural implant. This has implications for all spheres of military operations. What might this mean for the future of nuclear decision-making?

Human Augmentation and Nuclear Stability

The United States is a good case study for understanding how human augmentation might intersect with nuclear decision-making, given that it is more transparent about its nuclear decision-making protocols than other states possessing nuclear weapons.

There is a clear sequence of events involved in the short period from considering a nuclear strike to the decision to launch.13 When the president decides that a launch is an option, they convene a brief conference of high-ranking advisers, including members of the military, such as the officer in charge of the war room. Whatever the president decides, the Pentagon must implement. For a strike decision, the next step is to authenticate the order, then the encoded order goes out via an encrypted message. Once the launch message has been received by the submarine and ICBM crews, the sealed authentication system codes are retrieved and compared with the transmitted codes in a further authentication step before launch. The missile launch then is prepared. If launched from a silo, it takes five ICBM crew members to turn their keys simultaneously for a successful launch. This entire process from decision to ICBM launch can be completed within five minutes, 15 minutes if the launch is executed from a submarine. This is already a quick decision process with very little room for mediation, deliberation, or error. With human augmentation, the timelines would be compressed even further.

What is the effect on nuclear decision-making if brain-computer interface technologies tie humans more closely to computers and speed communication? That is one question that needs greater examination in nuclear policy, authors Marina Favaro and Elke Schwarz write. Here, members of the 576th Flight Test Squadron monitor an operational test launch of an unarmed Minuteman III missile at Vandenberg Air Force Base, Calif. in 2015. (U.S. Air Force photo/Michael Peterson)To understand the relevance of human augmentation to nuclear decision-making, three scenarios, each positioned at different points in the decision-making process, are illustrative and reflect a military context in which nuclear weapons interact with a brain-computer interface. Such an interface could detect when certain areas of the brain are cognitively activated (e.g., by certain thoughts) and transmit this signal, thereby enabling brain-controlled action or communication. The scenarios focus on the incorporation of a brain-computer interface into nuclear decision-making because of their potential to accelerate communications. The ability of an interface to compress temporal timelines enables us to probe the boundaries of the question “What is the value of a few seconds?”

The first scenario involves the advisory and decision chain. Assuming the above decision chain, the president, upon learning of a potential threat and considering the need to give a nuclear launch command, could shorten the deliberation time frame of their human advisers by transferring data directly to military participants via a brain-computer interface. In addition, the senior military commander could be plugged into an AI system that can compute many possible scenarios and give recommendations through the brain-computer interface in real time.

The upshot of human enhancement in this scenario might be that the military adviser has more data available through the interface, but critically, there is no guarantee of the quality or accuracy of this data. Moreover, with civilian and military advisers partaking in the advisory conference, there is a risk of the same algorithmic bias that is evident in human-machine teaming with AI systems. This refers to the tendency of humans to give uncritical priority to decisions that are ascertained with the help of technology, also called automation bias. This may make an advisory team superfluous and weaken the quality of advice in a critical situation. The mandate to act faster based on technologically derived advice could prevail and shorten the deescalation window.

In the second scenario, involving the executive chain, the launch crew is assumed to be partially or wholly networked through brain-computer interfaces. Perhaps the transmission of the launch codes takes place directly through computational networks, shortening the time between receiving the order, authenticating the codes, and executing the order. The submarine and ICBM crews executing the order by coordinated action are networked to facilitate the launch. In addition to the obvious vulnerabilities that any network inevitably produces, such as information and network security compromises, this would have the consequence of accelerating action. Any errors may be overlooked or not acted on in time. Particularly concerning is the possibility that a given action could rest on flawed initial inputs or skewed calculations, which would greatly increase the risk of unwarranted escalation.

Finally, there is the Petrov example, or the predecision phase. As suggested above, the decision chain does not really begin with the president’s decision to launch but with the input that the president receives from those in the military chain of command. If Petrov, the Russian duty officer whose job was to register apparent enemy missile launches, had been operating with a brain-computer interface in place and received the information transmitted directly to his brain and the brains of other military personnel, would there have been the impetus or indeed the opportunity to question the information from the system? Would he have had enough time to understand the context, draw on his experience, and make a considered judgment; or might he have felt compelled to uncritically execute the recommendations made by the system, which may well be indistinguishable in his mind to his own judgments?

In all three scenarios, the human is less able to exercise important human judgment at critical nuclear flashpoints. Algorithmic bias, increased system fog, lack of overall situational awareness, cognitive overload, and an accelerated action chain are consequences of intricate human-machine teaming through interfaces. This blurring of boundaries could obscure where machine input starts and human judgment ends. It could reduce the scope for cognitive input from the human and increase the extent to which algorithmic decisions prevail without serious oversight. Human experience and foresight based on noncomputational parameters are likely to be bracketed considerably. Is that wise in a nuclear context? As military operations prioritize speed and networked connectivity, slotting the human into a computer interface in the nuclear context may significantly exacerbate nuclear instability.14

Other Major Concerns

Regardless of its impact on nuclear stability, military human enhancement raises a myriad of ethical, legal, political, and other concerns that need to be explored. Among the questions that arise in this context, one involves consent, namely, can soldiers give free and informed consent to these enhancements, especially those that require a surgical procedure or invasive treatment, without pressure from their employer or peers? Will soldiers who consent to these interventions become part of an elite class of super soldiers, and what would be the impact of two classes of soldiers, enhanced and unenhanced, on morale? For how long would soldiers consent to these enhancements? Can an enhanced soldier ever go off duty or retire from service? In other words, when soldiers leave service, are they able to reverse these interventions? If not, what kinds of additional issues could this create for those who leave the service, who already encounter difficulties adjusting to a civilian environment?15 These consent issues are magnified in the many countries that maintain conscription.

In the area of brain-interface technologies, some American leaders worry about falling behind China, which experts say has a more coherent plan for using the technology, including to enhance military forces. (Photo by Andrea Verdelli/Getty Images)Second, has the practice of military human enhancement already created an arms race, with states scrambling to out-enhance each other?16 Already locked in an ostensible arms race for dominance in military AI,17 the United States, China, and Russia, among other states with advanced militaries, are all keeping a close eye on who is enhancing their fighting forces and how.18 If human enhancement is the missing link in perfecting human-machine teaming, a race in this arena is perhaps inevitable. Oversights and flaws associated with arms racing then become a cause for concern, especially given that human integrity is on the line.

Third, what kinds of information security concerns does human enhancement create? If service personnel use dual-use computer technologies in the same way as civilians use them, is it possible that this technology could create vulnerabilities and reveal sensitive data?19 Everything that is digitally connected can, in principle, be hacked. What kinds of vulnerabilities will be created by connection of the brain to a computer or by an increased number of networked devices, such as biosensors? How does the mere potential of cyberattacks on wearable or transdermal devices erode trust in the system? How can mass personal data collection and use be done without infringing on privacy?

Finally, who has this technology, and how do they use it? What kinds of interoperability concerns might military human enhancement create between allies? How might an asymmetry between “red” and “blue” forces using this technology impact nuclear stability? On a more philosophical note, if there is an accelerated action chain in a nuclear conflict, can such wars be won?

Averting Nuclear Risk

Notwithstanding the risks that embedding a brain-computer interface into nuclear decision-making could create, some types of enhancements could significantly benefit nuclear stability. It would be alarmist to focus exclusively on the risks created by human augmentation without also highlighting the potential opportunities. Indeed, brain-computer interfaces can provide new ways of accessing vast amounts of information and new ways of communication if human judgment is not sidelined in the process.

One significant opportunity created by military human enhancement relates to minimizing the risk of accidents. Nuclear weapons duty is known to be conducive to serious behavioral problems due to isolation, monotony, and confinement. During emergencies, sleep deprivation and heavy responsibilities may cause inaccuracy in judgment, hostility, or paranoia.20 In a prolonged nuclear alert, missile crews have reported visual hallucinations, balance disturbances, slowed movements, and lack of vigilance. The advantages of biosensors and bioinformatics to identify, predict, and treat such symptoms or at least give operators a break when needed could minimize the likelihood of nuclear war as a result of miscalculation, misunderstanding, or misperception. Indeed, evidence suggests that the world has been lucky, given the number of instances in which nuclear weapons could have been used inadvertently as a result of miscalculation or error.21 Future research should examine historical cases of nuclear near use and determine whether human augmentation could have minimized the likelihood of these disturbingly close calls. Meanwhile, bioinformatics could play a key role in identifying commanders and staff with the right cognitive and adaptive potential for command and control roles.22

There are also indications that biosensors could assist in the detection of chemical, biological, radiological, and nuclear agents embedded in smart clothing;23 better identify signs of nuclear activities,24 including weapons development; monitor and respond to radiation, including proposing treatment options in response to nuclear fallout;25 and prevent the illegal transportation and transfer of nuclear materials.26 This is not an exhaustive list; there are certainly other unforeseeable ways in which human augmentation will be relevant to the nuclear order, nuclear disarmament, and nuclear policy.

As is the case with all emerging technologies, there are risks and opportunities related to human augmentation and nuclear decision-making. On the one hand, human augmentation could minimize the risk of accidents and enable better human-machine teaming. On the other hand, there are profound legal, ethical, information security, and personnel-related questions that are overdue for rigorous examination. More research programs should consider how to mitigate the risks associated with human enhancement technologies, while remaining cognizant of their potential benefits. Ultimately, the question is, What is the value of a few additional seconds or minutes in the nuclear decision-making process, and are the trade-offs worthwhile?

Fortunately, the world is still some distance from a future of ubiquitous human augmentation, and the hype that suggests otherwise must be met with skepticism. Even so, human augmentation and nuclear decision-making have long been bedfellows; and the changing nature of war in the 21st century demands that all citizens, not just political and military leaders and technology experts, think deeply about what new risks and opportunities this intersection could unleash.



1. UK Ministry of Defence, “Human Augmentation—The Dawn of a New Paradigm,” May 2021, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/986301/Human_Augmentation_SIP_access2.pdf.

2. Ibid.

3. Norman Ohler, Blitzed (New York: Penguin, 2017).

4. UK Ministry of Defence, “Human Augmentation.”

5. Elsa B. Kania and Wilson VornDick, “China’s Military Biotech Frontier: CRISPR, Military-Civil Fusion, and the New Revolution in Military Affairs,” China Brief, Vol. 19, No. 18 (October 8, 2019), https://jamestown.org/wp-content/uploads/2019/10/Read-the-10-08-2019-CB-Issue-in-PDF2.pdf.

6. Pierre Bourgois, “‘Yes to Iron Man, No to Spiderman!’ A New Framework for the Enhanced Soldier Brought by the Report From the Defense Ethics Committee in France,” IRSEM Strategic Brief, No. 18 (February 24, 2021), https://www.irsem.fr/media/5-publications/breves-strategiques-strategic-briefs/sb-18-bourgois.pdf.

7. U.S. Defense Advanced Research Projects Agency (DARPA), “DARPA and the Brain Initiative,” n.d., https://www.darpa.mil/program/our-research/darpa-and-the-brain-initiative (accessed February 13, 2022).

8. Emily Waltz, “How Do Neural Implants Work?” IEEE Spectrum, January 20, 2020, https://spectrum.ieee.org/what-is-neural-implant-neuromodulation-brain-implants-electroceuticals-neuralink-definition-examples.

9. NATO Science & Technology Organization, “Science & Technology Trends 2020–2040: Exploring the S&T Edge,” March 2020, https://www.nato.int/nato_static_fl2014/assets/pdf/2020/4/pdf/190422-ST_Tech_Trends_Report_2020-2040.pdf.

10. UK Ministry of Defence, “Human Augmentation.”

11. Anika Binnendijk, Timothy Marler, and Elizabeth M. Bartels, “Brain-Computer Interfaces: U.S. Military Applications and Implications,” RAND Corp., RR-2996-RC, 2020, https://www.rand.org/content/dam/rand/pubs/research_reports/RR2900/RR2996/RAND_RR2996.pdf.

12. Ibid.

13. Dave Merrill, Nafeesa Syeed, and Brittany Harris, “To Launch a Nuclear Strike, President Trump Would Take These Steps,” Bloomberg, January 20, 2017, https://www.bloomberg.com/politics/graphics/2016-nuclear-weapon-launch/.

14. Olivier Schmitt, “Wartime Paradigms and the Future of Western Military Power,” International Affairs, Vol. 96, No. 2 (March 2020): 401–418, https://doi.org/10.1093/ia/iiaa005.

15. Sarah Grand-Clement et al., “Evaluation of the Ex-Service Personnel in the Criminal Justice System Programme,” RAND Corp., RR-A624-1, 2020, https://www.rand.org/content/dam/rand/pubs/research_reports/RRA600/RRA624-1/RAND_RRA624-1.pdf.

16. Yusef Paolo Rabiah, “From Bioweapons to Super Soldiers: How the UK Is Joining the Genomic Technology Arms Race,” The Conversation, April 29, 2021, https://theconversation.com/from-bioweapons-to-super-soldiers-how-the-uk-is-joining-the-genomic-technology-arms-race-159889.

17. Richard Walker, “Germany Warns: AI Arms Race Already Underway,” Deutsche Welle, June 7, 2021, https://www.dw.com/en/artificial-intelligence-cyber-warfare-drones-future/a-57769444.

18. Thom Poole, “The Myth and Reality of the Super Soldier,” BBC News, February 8, 2021, https://www.bbc.co.uk/news/world-55905354.

19. Alex Hern, “Fitness Tracking App Strava Gives Away Location of Secret US Army Bases,” The Guardian, January 28, 2018, https://www.theguardian.com/world/2018/jan/28/fitness-tracking-app-gives-away-location-of-secret-us-army-bases.

20. A.W. Black, “Psychiatric Illness in Military Aircrew,” Aviation, Space, and Environmental Medicine, Vol. 54, No. 7 (July 1983): 595-598.

21. Patricia Lewis, Benoît Pelopidas, and Heather Williams, “Too Close for Comfort: Cases of Near Nuclear Use and Options for Policy,” Chatham House, April 28, 2014, https://www.chathamhouse.org/2014/04/too-close-comfort-cases-near-nuclear-use-and-options-policy.

22. UK Ministry of Defence, “Human Augmentation.”

23. Richard Ozanich, “Chem/Bio Wearable Sensors: Current and Future Direction,” Pure and Applied Chemistry, Vol. 90, No. 10 (June 12, 2018), https://doi.org/10.1515/pac-2018-0105.

24. “New Biosensor Could Help Search for Nuclear Activity,” CBRNE Central, February 22, 2017, https://cbrnecentral.com/new-biosensor-could-help-search-for-nuclear-activity/10601/.

25. M. Gray et al., “Implantable Biosensors and Their Contribution to the Future of Precision Medicine,” The Veterinary Journal, Vol. 239 (September 2018), pp. 21–29, https://doi.org/10.1016/j.tvjl.2018.07.011.

26. Thamir A.A. Hassan, “Development of Nanosensors in Nuclear Technology,” AIP Conference Proceedings, Vol. 1799, No. 1 (January 6, 2017), https://doi.org/10.1063/1.4972925.

Marina Favaro is a research fellow at the Institute for Peace Research and Security Policy at the University of Hamburg, focusing on the impact of emerging technologies on international stability and human security. Elke Schwarz is a senior lecturer in political theory at Queen Mary University of London, focusing on the political and ethical implications of new technologies.

Brain-interface technologies could greatly augment human capabilities and even create “super soldiers.” If they shorten decision-making time, what is the impact on nuclear stability?

The Biden Nuclear Posture Review: Obstacles to Reducing Reliance on Nuclear Weapons

January/February 2022
By Adam Mount

President Joe Biden entered office with two objectives for nuclear weapons policy: declaring that “the sole purpose of our nuclear arsenal should be to deter—and, if necessary, retaliate against—a nuclear attack” and implementing the sole purpose policy as part of a broader effort to reduce the role of nuclear weapons. Although Biden has not clearly defined either goal, his support for both has signaled an intention to produce a significant shift in U.S. nuclear weapons policy. Public hints about the structure and the content of his administration’s Nuclear Posture Review (NPR) suggest, however, that Biden will not achieve his goal of reducing reliance on nuclear weapons in U.S. plans and posture.

During a January 11, 2017 speech to the Carnegie Endowment for International Peace, Vice President Joe Biden said that he and President Barack Obama “strongly believe we have made enough progress that deterring—and if necessary, retaliating against—a nuclear attack should be the sole purpose of the U.S. nuclear arsenal.” (Photo by Chris Kleponis/AFP via Getty Images)If Biden’s NPR walks back these commitments, it would be only the latest example of a president trying and failing to reduce the nation’s reliance on nuclear weapons. This pattern is the result of concerted opposition from partisan opponents and Pentagon officials, structural impediments to the president’s ability to shift policy, and the failure of political appointees to learn the lessons of past attempts. Even more than in previous nuclear policy reviews, these trends have been publicly visible throughout the 2022 NPR process and represent a cautionary tale for future administrations.

The Past as Prologue

For Biden, the 2022 NPR process is a familiar story. With Biden at his side, President Barack Obama entered office in 2009 also determined to reduce the nation’s reliance on nuclear weapons. Although this was the subject of his first major international address, Obama did not come equipped with a firm plan for how to do it. He left the issue up to the NPR process. For example, the Obama review explicitly did not adopt the concept of sole purpose as the role of nuclear weapons although it promised to “work to establish conditions under which such a policy could be safely adopted.”1 Instead, the review stated that the United States “would only consider the use of nuclear weapons in extreme circumstances to defend the vital interests of the United States or its allies and partners.”2 The administration reopened the question of the sole purpose concept in its last year in office, but Obama’s secretaries of defense, state, and energy all argued against it.3

In January 2017, nine days before he would leave office, Vice President Biden delivered a wistful speech to a Washington audience, reporting that “the President and I strongly believe we have made enough progress” toward creating the conditions for the sole purpose doctrine.4 Despite their conviction, they had not made the change. After eight years, the U.S. nuclear arsenal was still configured to deter the same threats for the same reasons.

The 2018 NPR reversed course and instead took steps to increase the nation’s reliance on nuclear weapons. Worried that adversaries could conduct a limited nuclear strike for coercive purposes, the Trump administration argued that new nuclear weapons were the answer. The 2018 review proposed new low-yield warheads, a new sea-launched cruise missile, and a delay in retiring the last megaton-class gravity bomb. That review also produced new language warning that the United States would consider employing a nuclear weapon in response to a “non-nuclear strategic attack,” a vague phrase the administration never defined. After critics warned that the new policy could permit a nuclear response to a cyberattack, officials hastened to dispute the claim, but never really clarified it.5 To this day, it is not clear who wrote the document. In short, it was not the kind of process that the Biden team should want to emulate.

In the past year, former President Donald Trump’s allies and advisers have worked hard to prevent the Biden administration from revisiting these decisions. Instead, they apparently hoped the new team could be coerced or cajoled into abandoning Biden’s stated goals and reaffirming the Trump policy.6 In a series of hyperbolic articles, they have argued that any change to U.S. policy would alarm allies and embolden adversaries, despite the fact that the Biden administration has not fully articulated its policy. According to this view, raising the bar for a U.S. nuclear response would give a green light to attacks that fall below that bar. Rather than engaging with any specific formulation of the sole purpose doctrine, these arguments tend to conflate the policy with a no-first-use strategy and object generally to any related change in existing policy.7

Admiral Charles Richard,head of the U.S. Strategic Command, has said that the purpose of Biden's NPR should be “validation, that we like the strategy we have.” (Photo by Department of Defense)In October, former Trump administration officials released a declassified document that was required to report on presidential guidance for nuclear employment plans that had been issued in April 2019.8 Although legally required to inform Congress of the change before it occurred, the outgoing team only sent the document to Capitol Hill in December 2020, after Trump lost the election. It is a strikingly partisan document that explicitly refutes policies to which Biden had committed on the campaign trail by arguing that “the United States sees no benefit and significant risk in adopting a ‘sole purpose’ policy” and claiming that doing so “would dispirit allies and partners.”9 The document does not provide Congress with information about Trump’s employment guidance, but rather serves as a handbook for civil servants, military officials, and sympathetic officials in allied countries who intended to resist Biden administration policies. It is more a partisan strategy than a nuclear strategy.

Consistent with that document, Admiral Charles Richard, who oversees the nation’s nuclear forces, said that the purpose of Biden’s NPR should be “validation, that we like the strategy we have.”10 With that perspective, Richard went before Congress in April to argue against options that the Biden administration was then considering, including a sole purpose policy and any changes to existing plans for acquiring new weapons.11 Further, an unnamed Pentagon official stated that it was “not likely” that sole purpose or no-first-use policies will be presented as options.12

Closing Off Options

This campaign effectively is an attempt to deprive Biden of the ability to set his own nuclear weapons policy. In this context, it would require a concerted effort to advance the president’s objectives. In practice, the Biden administration has taken steps in the structure and staffing of the review that further constrain its ability to pursue the president’s goals.

Biden’s first budget request, submitted in April, was an early opportunity to build leverage and set the tone of the review. Rather than pause or cancel questionable programs to preserve decision space, the request fully supported Trump’s accelerated schedule to procure new air-launched cruise missiles and continue developing a new sea-launched cruise missile (SLCM) and the low-yield submarine-launched warhead that Biden had called “a bad idea” during the campaign. The budget request is 28 percent higher than projected two years ago.13 These decisions guaranteed that the default position in the NPR would be to retain the existing policy, ensuring that debate would center around low-hanging fruit such as the SLCM that had been carefully positioned by the previous administration to divert attention from other policies.

The crucial moment for an administration seeking to shift nuclear weapons policy comes when the National Security Council (NSC) issues presidential guidance to initiate, indicate the president’s expectations for, and structure the NPR. For the Biden administration, this took the form of a public interim national security guidance document and a classified presidential study directive. Neither document referred to a sole purpose policy directly. Rather than explicitly direct that the Pentagon develop the president’s preferred options, the guidance was negotiated among a range of offices across the government, including officials from the Joint Chiefs of Staff. Meanwhile, there is no political appointee on the NSC staff empowered to represent and interpret the president’s guidance in the NPR process. Instead, the responsibility is divided between the offices of the NSC senior directors for defense and nuclear issues. The director for strategic capabilities in the defense office is customarily a uniformed general officer and so will tend to be more comfortable implementing settled policy than defining a shift in policy such as a sole purpose policy.

A visitor to the Peace Memorial Museum in Hiroshima walks by images of the mushroom cloud that erupted when the atomic bomb was dropped in 1945. An estimated 70,000 people were killed instantly and afterward, many thousands more died from radiation. Nearly four decades later, nuclear weapons remain a serious threat. (Photo by Junko Kimura/Getty Images)In the months leading up to the NPR, Leonor Tomero, the deputy assistant secretary of defense in charge of managing the review, came under fire from Senate Republicans and civil servants who worried that her views were too progressive.14 In particular, she was accused of favoring the sole purpose declaration that Biden supported and had written into the Democratic party platform. Rather than defend her, Pentagon leadership showed her the door, saying she was removed as part of a larger reorganization.15

The Biden administration began its NPR in July with the intention to release its report in January, along with the National Defense Strategy.16 Tomero’s removal meant there was no Pentagon political appointee in the NPR process who was prepared to implement the president’s sole purpose policy. Following Tomero’s departure, the NPR was led by Assistant Secretary of Defense Melissa Dalton and Richard Johnson, the deputy assistant secretary tasked with preventing the proliferation of weapons of mass destruction, who also served in an acting capacity in Tomero’s previous role, despite the apparent reorganization. Even capable officials such as Dalton and Johnson will have difficulty influencing the highly politicized and complex debates of nuclear weapons policy without experience with those arguments, without a portfolio that allows them to focus their full attention on the review, and without clear guidance from the president.

As a result, the administration has been unable to engage in a complete discussion on a sole purpose policy with allies, many of whom have been understandably apprehensive about potential shifts in an established U.S. policy. With firm guidance from the president and a concerted effort to adjust policy, U.S. officials might have engaged allies on their concerns about specific proposals. Without firm presidential guidance, allies have been left to fret about undefined concepts and rumors, allowing opponents of the president’s objectives in the Pentagon, Congress, and outside of government an opening to flood allies with misleading speculation. This mix of uncertainty and misinformation created an environment that made it easy for Pentagon political appointees to avoid serious consideration of the sole purpose issue altogether.

The administration also complicated the NPR process by folding nuclear weapons policy into a concept of “integrated deterrence.” The concept held considerable promise for Biden’s stated objectives for the review. If the United States was to reduce reliance on nuclear weapons, an integrated review could examine the utility and credibility of nuclear and non-nuclear options for performing specific missions and identify ways to safely reduce reliance on nuclear weapons.17 In principle, an integrated review could also communicate the benefits of Biden’s goals to allies, demonstrating how reduced reliance could lead to increased credibility in the overall U.S. deterrence posture.

A Soviet-era SS-23 missile is destroyed in 1989 under the now-defunct Intermediate-range Nuclear Forces Treaty. (Photo by TASS via Getty Images)The 2022 NPR and the National Defense Strategy did not undertake this assessment. There is no indication that the strategy was tasked with reducing reliance on nuclear weapons, and the bureaucratic silos that have divided the NPR and the broader National Defense Strategy remain intact. Instead, combining the documents could decrease the transparency of nuclear policy, concealing areas where the review failed to reach agreement or advance the president’s objectives. An integrated review that does not engage with the difficult questions of operational plans and posture might reduce the word count assigned to nuclear weapons policy, but not the missions assigned to the weapons.

Without firm presidential guidance, staff empowered to implement that guidance, and a detailed examination of the utility of nuclear and non-nuclear capabilities, the Pentagon is unlikely to produce a policy that significantly reduces reliance on nuclear weapons. The administration evidently acquiesced in a broad effort to undermine the president’s stated objectives and his ability to set policy. As it stands, the 2022 NPR will not only preserve the nation’s reliance on nuclear weapons, but if its practices are adopted by future administrations, will make it more difficult to accomplish the goal in coming years.

The Path Forward on Reducing Reliance

It is still possible that the administration could adjust declaratory policy through other means. The undersecretary of defense for policy or the national security adviser could choose to rewrite the NPR material or make significant amendments at the 11th hour, similar to the process that occurred in the 2010 NPR. Although such intervention might further Biden’s stated objectives, it would also underline that the NPR failed to perform that task and that the review process had to be circumvented to adjust policy. Furthermore, the administration should avoid last-minute changes that are simply cosmetic. Declaratory policy is consequential and credible to the extent that it reflects a strategy that shifts reliance away from nuclear weapons in operational plans.

Nevertheless, one option is to declare that the United States would use nuclear weapons only in the event of an “existential attack” against the United States or its allies.18 Although Colin Kahl, undersecretary of defense for policy, spoke approvingly about this possibility in the spring of 2021, the formulation raises its own questions. Could a cyberattack or chemical weapons attack ever threaten the existence of an ally? Would attacks that leave U.S. allies intact but exposed to subsequent attacks count as existential? These questions permit widely divergent interpretations by allies and adversaries and, depending on the exact language in the document and the statements of U.S. officials, might fail to raise the bar significantly for nuclear use.

The administration will have another opportunity to adjust nuclear weapons policy when it drafts its own nuclear employment guidance over the next year or two. This could serve as an opportunity to translate shifts in declaratory policy into operational plans and require planners to develop more credible, flexible nonnuclear options for specific contingencies. This would require a more active and directed employment guidance process than in previous years. Without fixing the decisions that constrained the NPR process, it will be even more difficult to affect the complex and parochial planning process. It will require that the president issues clear implementation guidance if he selects new declaratory language and empowers expert officials to create a significant change in strategy.

An administration committed to reducing reliance on nuclear weapons will have to learn three lessons from the 2022 NPR process if it is to succeed where its predecessors have failed. First, the president should issue clear guidance about what they want, including an explicit description of how to reduce nuclear reliance and what options should be developed and presented to the president for decisions. Second, the president will have to select and appoint expert officials to lead the NPR process who are ready to defend and implement that guidance. Third, civilian leaders must ensure that military officers, civil servants, and political appointees follow the president’s guidance and hold them accountable if they refuse to do so or attempt to subvert the review process, for example, if they mislead allies, undermine political appointees, or coordinate with the administration’s opponents in Congress. If Pentagon officials disagree with the president’s guidance, they have a duty to try to convince the president to change it, but they also have a duty to provide options requested by the president.

Whether or not Biden, confronted with political resistance or additional information, changed his mind on a sole purpose policy, the 2022 NPR demonstrates that the existing process for developing nuclear weapons policy is deeply flawed. Deputy National Security Advisor Jon Finer optimistically promised that “this is going to be the president’s posture review and the president’s posture.”19 It is also possible that, in his final days in office, Biden may find himself delivering another wistful speech lamenting that yet another administration has failed to establish a sole purpose policy as a guiding principle of U.S. nuclear policy or to significantly reduce reliance on nuclear weapons.


1. U.S. Department of Defense, “Nuclear Posture Review Report,” April 2010, p. 16, https://dod.defense.gov/Portals/1/features/defenseReviews/NPR/2010_Nuclear_Posture_Review_Report.pdf.

2. Ibid.

3. David E. Sanger and William J. Broad, “Obama Unlikely to Vow No First Use of Nuclear Weapons,” The New York Times, September 6, 2016.

4. Office of the Vice President, The White House, “Remarks by the Vice President on Nuclear Security,” January 12, 2017, https://obamawhitehouse.archives.gov/the-press-office/2017/01/12/remarks-vice-president-nuclear-security.

5. George Perkovich, “Really? We’re Gonna Nuke Russia for a Cyberattack?” Politico, January 18, 2018, http://politi.co/2Dpp28s; Scott D. Sagan and Allen S. Weiner, “The U.S. Says It Can Answer Cyberattacks With Nuclear Weapons. That’s Lunacy.” The Washington Post, July 9, 2021; Patrick Tucker, “No, the U.S. Won’t Respond to a Cyber Attack With Nukes,” Defense One, February 2, 2018.

6. Eric Edelman and Franklin Miller, “President Biden, Don’t Help Our Adversaries Break NATO,” The Washington Post, November 4, 2021; Jim Risch, “The U.S. Must Reject a ‘Sole Purpose’ Nuclear Policy,” Defense News, October 25, 2021. Those arguments were mirrored by editorials boards and some Democratic politicians. “Folding America’s Nuclear Umbrella,” The Wall Street Journal, November 12, 2021; Seth Moulton, “We Must Eliminate Nuclear Weapons, but a ‘No First Use’ Policy Is Not the Answer,” The Hill, November 29, 2021, https://thehill.com/blogs/congress-blog/politics/583433-we-must-eliminate-nuclear-weapons-but-a-no-first-use-policy-is. Reports on allied concerns about President Joe Biden’s objectives featured prominently in arguments made by opponents and in media accounts of the review. Demetri Sevastopulo and Henry Foy, “Allies Lobby Biden to Prevent Shift to ‘No First Use’ of Nuclear Arms,” Financial Times, October 30, 2021.

7. Patty-Jane Geller, “What Experts and Senior Officials Have Said About Adopting a No-First-Use or Sole-Purpose Nuclear Declaratory Policy,” Heritage Foundation Factsheet, No. 219 (October 20, 2021), https://www.heritage.org/sites/default/files/2021-10/FS219.pdf.

8. Robert Soofer and Matthew R. Costlow, “An Introduction to the 2020 Report on the Nuclear Employment Strategy of the United States,” Journal of Policy and Strategy, Vol. 1, No. 1 (Fall 2021): 2–8, https://nipp.org/wp-content/uploads/2021/10/1.1R.pdf.

9. U.S. Department of Defense, “Report on the Nuclear Employment Strategy of the United States - 2020,” 2020, p. 8, https://www.esd.whs.mil/Portals/54/Documents/FOID/Reading%20Room/NCB/21-F-0591_2020_Report_of_the_Nuclear_Employement_Strategy_of_the_United_States.pdf.

10. Charles R. Richard and Ronald R. Fritzmeier, Remarks to the Defense Writers Group, January 5, 2021, https://cpb-us-e1.wpmucdn.com/blogs.gwu.edu/dist/2/672/files/2021/01/DWG-Admiral-Charles-R.-Richard.pdf.

11. “To Receive Testimony on United States Strategic Command and United States Space Command in Review of the Defense Authorization Request for Fiscal Year 2022 and the Future Years Defense Program,” April 20, 2021, https://www.armed-services.senate.gov/download/transcript2042021.

12. Bryan Bender and Lara Seligman, “Biden’s Nuclear Agenda in Trouble as Pentagon Hawks Attack,” Politico, September 23, 2021, https://www.politico.com/news/2021/09/23/leonor-tomero-pentagon-nuclear-hawks-513974.

13. Kingston Reif and Shannon Bugos, “Biden’s Disappointing First Nuclear Weapons Budget,” Arms Control Association Issue Brief, Vol. 13, No. 4 (July 9, 2021), https://www.armscontrol.org/issue-briefs/2021-07/bidens-disappointing-first-nuclear-weapons-budget.

14. “To Receive Testimony on the Department of Defense Budget Posture for Nuclear Forces in Review of the Defense Authorization Request for Fiscal Year 2022 and the Future Years Defense Program,” May 12, 2021, https://www.armed-services.senate.gov/download/transcriptnuclear51221; Bryan Bender and Lara Seligman, “Biden’s Nuclear Agenda in Trouble as Pentagon Hawks Attack,” Politico, September 23, 2021, https://www.politico.com/news/2021/09/23/leonor-tomero-pentagon-nuclear-hawks-513974.

15. Lara Seligman, Alexander Ward, and Paul McLeary, “Pentagon’s Top Nuclear Policy Official Ousted in Reorganization,” Politico, September 21, 2021, https://www.politico.com/news/2021/09/21/pentagon-top-nuclear-official-ousted-reorganization-513502.

16. Kingston Reif, “Biden Administration Begins Nuclear Posture Review,” Arms Control Today, September 2021, pp. 26–27, https://www.armscontrol.org/act/2021-09/news/biden-administration-begins-nuclear-posture-review.

17. Adam Mount and Pranay Vaddi, “An Integrated Approach to Deterrence Posture,” Federation of American Scientists, January 2021, https://fas.org/wp-content/uploads/2020/12/An-Integrated-Approach-to-Deterrence-Posture.pdf; Brad Roberts, “It’s Time to Jettison Nuclear Posture Reviews,” Bulletin of the Atomic Scientists, Vol. 76, No. 1 (2020): 31–36.

18. This formulation would presumably apply to nuclear attacks and other existential attacks in order to maintain an option to employ U.S. nuclear forces to prevent any nuclear attack, not only existential nuclear attacks. For a recently proposed version of this option, see George Perkovich and Pranay Vaddi, “Proportionate Deterrence: A Model Nuclear Posture Review,” Carnegie Endowment for International Peace, January 21, 2021, https://carnegieendowment.org/2021/01/21/proportionate-deterrence-model-nuclear-posture-review-pub-83576. This version specifies that nuclear weapons could be used “only when no viable alternative exists to stop” an existential attack in order to confine potential use to preemption of an attack and to accommodate the “nuclear necessity principle.” Jeffrey G. Lewis and Scott D. Sagan, “The Nuclear Necessity Principle: Making U.S. Targeting Policy Conform With Ethics and the Laws of War,” American Academy of Arts and Sciences, Fall 2016, https://www.amacad.org/publication/nuclear-necessity-principle-making-us-targeting-policy-conform-ethics-laws-war.

19. Emma Belcher, “Press the Button,” podcast, Ploughshares Fund, November 2, 2021, https://soundcloud.com/user-954653529/the-white-houses-jon-finer-on-all-things-nuclear.


Adam Mount is director of the Defense Posture Project and a senior fellow at the Federation of American Scientists.

Public hints about the Nuclear Posture Review, to be released early this year, suggest President Joe Biden will not achieve his goal of reducing the U.S. reliance on nuclear weapons.

The Biden Nuclear Posture Review: Resetting the Requirements for Nuclear Deterrence

January/February 2022
By Sharon K. Weiner

As the Biden administration finalizes its Nuclear Posture Review (NPR), it faces the same challenges as the architects of the four earlier NPRs: how to make choices about nuclear deterrence and translate them into nuclear strategy and force structure. If it chooses to learn from the experience of its predecessors, the administration will confront two sets of requirements that are central to U.S. nuclear deterrence policy yet limit its freedom of action. The NPR managers would be wise not to just buy into those requirements but instead to be explicit and transparent about questioning them in order to enable choices that are based on a clear understanding of the trade-offs, as well as other possible options.

Airmen from the 90th Maintenance Group at F.E. Warren missile complex in Wyoming work on maintaining an intercontinental ballistic missile (ICBM), one leg of the nuclear triad, in December 2019. (Photo by U.S. Air Force)One set of deterrence requirements that is almost certainly being presented to President Joe Biden by the nuclear weapons establishment as strategic or military necessities are actually choices. A second set of requirements is the taken-for-granted assumptions that are often overlooked. All of these so-called requirements are presumed to be based on evidence and are never challenged in a way that would determine their actual validity. They are more aspirational than necessary. They are rooted in stories that strategists, policymakers, and the military tell themselves, each other, and the public about how they hope deterrence will work.

Discussions of nuclear strategy and force structure are full of references to things that are required. A modernized triad is a requirement for deterrence, and anything less will leave the United States vulnerable.1 The Ground-Based Strategic Deterrent (GBSD) system, a fleet of next-generation intercontinental ballistic missiles (ICBMs), meets U.S. Strategic Command’s requirements, but the current fleet of Minuteman III missiles does not.2 Operational requirements necessitate no fewer than four concurrent warhead life extension programs.3 Nuclear weapons plutonium pit production at a rate of least 80 pits per year by 2030 is a requirement, otherwise U.S. nuclear weapons will not work as intended.4 A national military uranium-enrichment plant is a requirement, otherwise there will be no way to make tritium for nuclear weapons or fuel for naval nuclear reactors.5 These and many other immutable positions held by the nuclear enterprise can make it seem as if everything is a requirement and that there can be no serious alternatives without a collapse of the whole deterrence structure.

Labeling something a requirement suggests it is necessary to avoid failure. In Pentagon jargon, however, a requirement is not required. It is the culmination of a decision-making process that found a particular outcome desirable, given other goals and constraints.6 In other words, something becomes required because it was the result of due process, not because it was the only option for achieving a national security goal. Requirements are, in fact, malleable bureaucratic constructions. They reflect and can change with the decision-making process and its inputs and constraints.

The unwillingness to confront the challenge of entrenched interests and ideas has led critics to judge that “all prior NPRs…have generally—and disappointingly—rubber-stamped the nuclear status quo.”7 This also underpins the broader observation by Admiral Charles Richard, the head of Strategic Command, that “this nation has had basically the same strategy dating back to the Kennedy administration. It’s been repeatedly validated through multiple administrations. It would be useful to do that again.”8 If the Biden NPR continues this trend, it should do so only after actively challenging the requirements and assumptions.

Choices, Not Requirements

The contextual nature of requirements can be seen in the shifting arguments in support of the GBSD program. Initially, the requirement for this weapons system was based on cost. Advocates argued that it is cheaper to design, develop, and build a new fleet of 659 ICBMs and to rebuild the command-and-control systems in the 450 missile silos and 45 missile launch control facilities than to sustain the existing Minuteman III fleet.9 When independent analysis suggested otherwise, the requirement argument shifted to technology: the GBSD program is required because the Minuteman III can no longer be maintained or upgraded indefinitely. Yet, numerous options to replace parts of the Minuteman system and keep it functioning for the foreseeable future have been offered. Today, arguments for the GBSD program increasingly focus on new threats that cannot be covered by the Minuteman III. Thus, the GBSD program now is a requirement for deterrence.

Nuclear-powered submarines constitute one leg of the U.S. nuclear triad.  (Photo by U.S. Navy)Before accepting that deterrence and also presumably U.S. national security rely unequivocally on the GBSD program, the administration should ask exactly what ICBMs in general and the GBSD system in particular contribute to deterrence that is necessary or unique and explore other choices to meet this requirement. For example, China’s nuclear modernization may create new targets or make existing targets more difficult to hold at risk. Yet, is deterring China somehow less effective if those targets are covered by submarine-launched nuclear weapons alone or in combination with Minuteman IIIs? More specifically, if the nuclear-armed submarines can hold at risk 95 percent of the targets in China, is it worth the estimated $264 billion life cycle cost of the GBSD program to increase that margin to 97 percent?

These are not rhetorical questions. The imbalance between the arsenal necessary to meet military requirements and the existing stockpile has been an enduring characteristic of U.S. nuclear decision-making. In the early 1960s, Secretary of Defense Robert McNamara argued that assured destruction of the Soviet Union would require the ability to destroy 20 to 25 percent of the Soviet population and half its industrial capacity.10 He calculated that this would necessitate 400 one-megaton warheads. At the time, the United States had just under 18,000 megatons in its arsenal. McNamara felt that he needed to translate deterrence into a precise requirement or it would be difficult to constrain spending on nuclear weapons.

In 2012 the military concluded it could meet all necessary military requirements with about 1,000 deployed strategic warheads rather than the approximately 1,550 deployed strategic warheads agreed under the New Strategic Arms Reduction Treaty.11 Almost 10 years later, in April 2021, Richard told the House Armed Services Committee that the triad is designed to meet all presidential requirements even if one leg is lost.12 Put simply, current nuclear deterrence goals could be met without ICBMs, either the Minuteman III or the GBSD program. There is plenty of additional evidence to suggest that the size of the arsenal is derived from something other than military requirements and that there is room for significant reductions without compromising deterrence.

Another requirement that is likely to be examined by the NPR is the production of pits, which are the hollow metal cores that enable the initial explosive reaction in a nuclear weapon. The nuclear establishment has asserted that large-scale pit production is vital because without it, nuclear weapons may not function as specified. If nuclear weapons do not work, then deterrence suffers because deterrence rests on the capability to inflict damage and on the ability to hold at risk things that the enemy values. The debate over pit production, however, is not about whether the weapons will work but how well they will work.

Military requirements for weapons performance are classified, but presumably the administration can be briefed on these requirements and on the degree to which they could suffer if pits do not function exactly as intended. For example, if the government has 90 percent confidence that a nuclear weapon will explode on target with 98 percent of its anticipated yield, does that deter less than a weapon in which there is 95 percent confidence? Given that the United States has about 1,550 deployed strategic warheads and perhaps twice that number in the arsenal of reserve warheads known as the hedge, how many of these weapons have to work at what level to deter? Does the country have enough redundant capability at least to call into question the need to spend $18 billion—a figure certain to increase, perhaps significantly—on the required pit production capability?

The GBSD system, pit production, and multiple other choices about force structure should be considered requirements only after they survive comparison to alternative means for achieving robust deterrence, including force structure trade-offs and possible changes to presidential guidance about targeting and the acceptable margin for error. To make such choices, the administration first needs to scrutinize the myriad requirements for deterrence that often go unexamined.

Requirements for Deterrence

Nuclear weapons are said to deter many things. In April 2021 testimony to the Senate Armed Services Committee, Richard outlined numerous threats facing the United States.13 The list was ominous. China is bent on establishing hegemony in East Asia and denying the United States the ability to project power in the region and to maintain stable relationships with traditional allies. Russia too is focused on expanding its sphere of influence, challenging U.S. leadership, and eroding international norms. North Korea threatens the stability of the Korean peninsula, and Iran is using proxy forces in an attempt to destabilize the Middle East.

In outlining these national security challenges, Richard is no different from other Department of Defense witnesses. Indeed, there seems to be a strong consensus that the United States faces multiple, growing threats, especially in East Asia. What makes Richard’s testimony stand out is not his assessment of the security situation but the nuclear arsenal that he is in charge of mustering in response. He made expansive claims about the power of nuclear deterrence, saying it is “the foundation of our national defense policy and enables every U.S. military operation around the world.”14 More explicitly, he said, nuclear weapons provide the “maneuver space” necessary for the United States “to project conventional military power strategically.”15

An aging Minuteman III missile, slated to be replaced by the new Ground-Based Strategic Deterrent system, is fitted with a new cable by members of the U.S. Air Force. (Photo by U.S. Air Force)The administration’s NPR should make clear its perspective on the expansive role for nuclear weapons and nuclear deterrence advocated by Strategic Command. Nuclear deterrence as cover for conventional military operations around the globe and as a requirement for nuclear forces and a posture able to “deter all countries, all the time”16 is a significant expansion of the original mission of these weapons, namely deterring existential threats against the United States. Twenty-plus years after the Cold War, a “bolt from the blue” surprise attack intended to destroy the United States is increasingly dismissed as unlikely.17 Deprived of the main raison d’être, one might expect nuclear weapons to be marginalized or at least relegated to a smaller role in U.S. strategy.

The review also needs to consider what next steps will be necessary if nuclear weapons fail to deter conventional or other aggressive actions. Most specifically, how will escalation be controlled? The experience of Strategic Command is that escalation control never works. “It ends the same way every time,” explained General John Hyten, the Strategic Command chief, in 2018 after the annual Global Thunder wargame. “It ends bad. And the bad meaning it ends with global nuclear war.”18

In the event that Russia uses a nuclear weapon for the first time, even on a limited basis, to what extent does Strategic Command planning and U.S. credibility dictate that the president respond not in kind but by escalating, by using just a bit more? This supposedly is the logic behind the escalate-to-deescalate doctrine, under which a country would threaten to ratchet up the violence to make an adversary back down. If Russia and the United States adopt this logic, then escalation is unlikely to be controlled, and the use of even low-yield nuclear options runs a significant risk that it will lead to mutually assured destruction.

From the perspective of deterrence, if the review endorses low-yield nuclear options, it means the administration has examined the requirements for escalation control, brinksmanship, and competitive risk-taking and has concluded that limited goals are worth the danger of total nuclear war.

At the deepest level, the most important requirement that the NPR should examine is that of rational decision-making, a concept fundamental to nuclear deterrence yet most often under-analyzed. Deterrence assumes leaders can weigh rationally the costs and benefits of their actions under any and all circumstances, if not completely then at least sufficiently to justify a final decision. Anyone who has been involved in a crisis understands, however, that this assumption is unrealistic. This is confirmed by a vast literature on foreign policy decision-making, behavioral economics, and behavioral psychology that shows people rely on a variety of less-than-rational shortcuts, especially in a crisis and when the stakes are high, information is missing or uncertain, and time is short.

Research has shown that people tend to assume the current situation is “just like” one they recently experienced or that they make a decision on the basis of a “gut feeling” rather than analyzing the available data or seeking additional relevant data. In a crisis, people tend to assume their motivations are clearly understood and assume that they are more in control of a situation than they actually are. Of particular concern is the tendency in crises for people to be biased toward risk taking rather than playing it safe. Given that in a nuclear crisis a U.S. president is likely to have 15 minutes or less to make a decision with unimaginably profound consequences, the NPR managers should ask themselves the degree to which they expect themselves and any adversary to behave rationally in a crisis and be prepared to explain the answer in detail and in public.

Choices and Assumptions

Given the experience of the past four NPRs, Biden can expect the review process to offer him few real options for nuclear policy reform; these options will likely allow, at best, only narrow deviations from the status quo. The nuclear weapons establishment will limit choice by presenting everything as an interlocking set of military requirements instead of multiple options for meeting deterrence goals.

As the administration weighs inputs into its review, managers could start by searching for and replacing every mention of “military requirement” with “presidential choice.” Biden can treat the requirements with which he is presented as choices that a president is entitled to make and seek new opportunities to satisfy national security needs with fewer nuclear weapons and with less reliance on the threat of their use. Biden finally could choose to reset the guidance to Strategic Command on nuclear deterrence goals. As Richard has recognized, for Strategic Command, “[T]here is a total amount of capability and capacity that's required to execute the responsibilities that I have been given.… We don't have capacity…to start to change that unless we change the guidance, right? And we can always do that.”19

The wisdom of developing new options for nuclear strategy and policy becomes even clearer if all questions of nuclear deterrence are seen not simply as questions of a calculus of nuclear forces and nuclear postures but as sets of unproven assumptions about the likely behavior of the United States and its potential adversaries under conditions of extraordinary uncertainty and stress with no basis for expecting a good outcome.


1. John E. Hyten, Statement before the U.S. Senate Armed Services Committee, February 26, 2019, https://www.armed-services.senate.gov/imo/media/doc/Hyten_02-26-19.pdf.

2. John A. Tirpak, “New GBSD Will Fly in 2023; No Margin Left for Minuteman,” Air Force Magazine, June 14, 2021.

3. Charles Richard, Testimony before the U.S. Senate Armed Services Committee, April 20, 2021, https://www.armed-services.senate.gov/imo/media/doc/21-22_04-20-2021.pdf (hereinafter Richard testimony).

4. Sharon K. Weiner, “Reconsidering U.S. Plutonium Pit Production Plans,” Arms Control Today, June 2020.

5. Frank N. von Hippel and Sharon K. Weiner, “No Rush to Enrich: Alternatives for Providing Uranium for U.S. National Security Needs,” Arms Control Today, July/August 2019, pp. 10–15.

6. For a discussion of this lexicon, see Mark Cancian, “Bad Idea: Using the Phrase ‘Military Requirements,’” Defense 360, December 6, 2018, https://defense360.csis.org/bad-idea-using-the-phrase-military-requirements.

7.  Stephen I. Schwartz, “Ready, Aim, Fired: Can Biden Rescue the Nuclear Posture Review?” Bulletin of the Atomic Scientists, September 30, 2021, https://thebulletin.org/2021/09/ready-aim-fired-can-biden-rescue-the-nuclear-posture-review.

8. Charles Richard, Remarks to the Defense Writers Group, Project for Media and National Security, George Washington School of Media and Public Affairs, January 5, 2021, https://cpb-us-e1.wpmucdn.com/blogs.gwu.edu/dist/2/672/files/2021/01/DWG-Admiral-Charles-R.-Richard.pdf.

9. Amy Woolf, “U.S. Strategic Nuclear Forces: Background, Developments, and Issues,” CRS Report, RL33640, July 13, 2021, https://sgp.fas.org/crs/nuke/RL33640.pdf.

10. John T. Correll, “The Making of MAD,” Air Force Magazine, July 27, 2018.

11. R. Jeffrey Smith, “Obama Administration Embraces Major New Nuclear Weapons Cut,” Center for Public Integrity, February 8, 2013, https://publicintegrity.org/2013/02/08/12156/obama-administration-embraces-major-new-nuclear-weapons-cut.

12. Richard testimony.

13. Ibid.

14. Ibid.

15. Charles Richard, “Forging 21st-Century Strategic Deterrence,” U.S. Naval Institute Proceedings, February 2021.

16. Richard used this phrase on April 22, 2021, when he told the press, “I don't have the luxury of deterring one country at a time, right? I have to deter all countries, all the time, in order to accomplish my mission sets.” “Admiral Charles Richard, Commander, U.S. Strategic Command, Holds a Press Briefing,” U.S. Department of Defense, April 22, 2021, https://www.defense.gov/Newsroom/Transcripts/Transcript/Article/2582171/admiralcharles-a-richard-commander-us-strategic-command-holds-a-press-briefing (hereinafter Richard press briefing).

17. For example, see “The Future of Strategic Deterrence and Nuclear Modernization: A Conversation With Admiral Charles Richard,” The Brookings Institution, May 7, 2021, https://www.brookings.edu/wpcontent/uploads/2021/05/fp_20210507_strategic_deterrence_richard_transcript.pdf.

18. John Hyten, Speech at the Mitchell Institute Triad Conference, July 17, 2018, https://www.stratcom.mil/Media/Speeches/Article/1577239/the-mitchell-institute-triad-conference/.

19. Richard press briefing.


Sharon K. Weiner is an associate professor at the School of International Service at American University. This article draws on and includes parts of testimony to the U.S. Senate Armed Forces strategic forces subcommittee on June 16, 2021.

What nuclear weapons are really needed to deter adversaries? The unwillingness to confront the challenge of entrenched interests and ideas has locked the United States into a decades-long status quo.

The Biden Nuclear Posture Review: Defense, Offense, and Avoiding Arms Races

January/February 2022
By Steven Pifer

President Joe Biden’s administration is conducting a missile defense review in parallel with its Nuclear Posture Review (NPR). Those reviews will determine whether to adjust the nuclear and missile defense programs that the administration inherited from its predecessor. They will also shape decisions on the contribution that negotiated arms control could make to meet the increasingly complex challenges of maintaining strategic stability and enhancing U.S. and allied security.

A ground-based interceptor (GBI) rocket is launched in May 2017 from Vandenberg Air Force Base, California. (Photo by Gene Blevins/AFP via Getty Images)One question the administration should consider is whether it can design a missile defense approach that would protect the homeland against limited attacks by rogue states such as North Korea while avoiding an offense-defense dynamic that would frustrate efforts to achieve nuclear arms reductions with Russia that go beyond the 2010 New Strategic Arms Reduction Treaty (New START) or to agree on any constraints on nuclear forces with China.

The Offense-Defense Relationship

In June, the administration launched a missile defense review, which should be completed early in 2022, about the same time as the NPR. The two documents produced by the reviews should be considered in tandem.

Washington and Moscow have long recognized the interrelationship between strategic offense and defense. The Strategic Arms Limitation Talks (SALT) in 1972 produced agreements addressing both sides of the equation. The Interim Offensive Agreement constrained intercontinental ballistic missile (ICBM) and submarine-launched ballistic missile (SLBM) launchers. The Anti-Ballistic Missile (ABM) Treaty limited strategic missile interceptors and prohibited a national missile defense. The two accords were seen as enhancing strategic stability, a situation in which incentives for the United States or the Soviet Union, and later Russia, to strike first with nuclear weapons were minimized.

U.S. President Ronald Reagan took a different approach in 1983 with the Strategic Defense Initiative. He sought to defend the United States against ballistic missile attacks of any size, although the limitations of technology and cost frustrated that goal. The National Missile Defense Act of 1999 set U.S. policy so as to “deploy as soon as is technologically possible an effective National Missile Defense system capable of defending the territory of the United States against limited ballistic missile attack.”1

The George W. Bush administration withdrew from the ABM Treaty in 2002 and subsequently began deploying ground-based interceptors (GBIs) to engage strategic ballistic missile warheads. The United States maintains 44 GBIs, with plans to add 20 more by 2030. The military also deploys Standard Missile-3 (SM-3) and Terminal High Altitude Area Defense (THAAD) interceptors, which are designed to engage short- and intermediate-range ballistic missile warheads.

The Trump administration’s Missile Defense Review affirmed the idea of defending against “a limited ICBM attack” mounted by a rogue state, although the president’s comment that the U.S. goal was to “ensure we can detect and destroy any missile launched against the United States—anywhere, anytime, anyplace” suggested that he might have something more ambitious in mind.2 As for Russian and Chinese strategic ballistic missiles, the review said the United States “relies on nuclear deterrence to prevent potential Russian or Chinese nuclear attacks.” The review spelled out, however, new technologies for exploration and possible development.3 Moreover, in 2020, the Pentagon successfully tested an SM-3 interceptor against an ICBM warhead-class target as part of an effort to develop a second layer of interceptors to supplement the GBI system.

Although Russian officials regularly voice concern about U.S. missile defenses, their fears appear overstated given Russia’s large ICBM and SLBM warhead numbers. Still, the Russian military has developed systems such as the Avangard boost-glide vehicle to penetrate missile defenses. In addition, Russia maintains its own missile defense systems, including the A-135 system protecting Moscow, the S-400, and the new S-500. The latter two systems are advertised as having capabilities similar to the SM-3 and THAAD systems.

Beijing too has expressed concern about U.S. missile defenses, including their ability to negate a retaliatory strike following a U.S. attack on China’s nuclear deterrent. That concern could explain the recent Chinese test of what appears to have been a hypersonic glide vehicle mounted on a fractional orbital bombardment system.4 Such a system could approach the United States from the south, thus potentially evading U.S. missile defense radars that are oriented toward threats from the north or the Pacific and Atlantic Oceans. At least for the near term, Beijing’s concerns have a stronger base than Moscow’s, given the significantly smaller number of Chinese strategic warheads.

Offense Wins and Arms Race Concerns

With existing missile defense capabilities, offense will win the strategic offense-defense competition, a point acknowledged in September by General John Hyten, vice chairman of the Joint Chiefs of Staff, who said, “The defensive capabilities that we have been building tend to be very cost prohibitive on us.… And when our interceptor costs more than the weapon attacking us, that’s a bad place to be.”5 This echoes the argument made by U.S. negotiator Paul Nitze in 1985 that missile defenses should be judged on, among other things, whether they are cost effective at the margin.6

Current U.S. missile defenses fall short of that criterion. The cost of a GBI missile and kill vehicle is $65–75 million. Although the Pentagon says GBI systems have succeeded in 55 percent of their tests, skeptics argue that, given the scripted nature of the tests, this record likely overstates their performance. If these interceptors could replicate their test performance in a real attack, it would take three GBIs, costing $195–225 million, to have a 91 percent chance of destroying an incoming warhead. Russia, China, and North Korea could each build many additional warheads and decoys for that same amount of money.7

This calculation could change, most probably if a missile defense technology based on directed energy were to prove feasible. For the foreseeable future, however, spending heavily on existing strategic missile defenses appears a losing game. An adversary can increase the number of its strategic warheads and decoys at far less cost. Although existing U.S. strategic missile defenses may not be that effective, the other side will assume they will improve and increase in number. That will affect the adversary’s calculation of what strategic offensive force it needs to be able to absorb a first strike and overcome U.S. missile defenses to inflict a powerful retaliatory blow.

If missile defenses remain unconstrained and grow in number, the other side may conclude that it must expand its strategic offensive forces. This may well be a factor behind China’s apparent effort to increase its strategic nuclear forces. Russian military planners, facing questions about the future of U.S. missile defenses, might question whether Russia can afford reductions below New START’s limits. The situation could devolve into something similar to the competition between the United States and Soviet Union in the 1960s with both sides increasing ICBM and SLBM forces in part to have confidence in defeating the other’s developing strategic missile defenses. In the worst case, Washington might find itself in offense-defense races with both Russia and China.

Missile Defense Review

The Biden administration’s Missile Defense Review should address several questions. First, should the objective of U.S. missile defense policy remain protecting against a limited ballistic missile attack on the United States? If so, are specific programs unnecessary for that goal, or do they suggest to potential adversaries a desire to defend against larger-scale attacks?

Looking out over the next 10 to 20 years, will U.S. GBIs and other interceptors improve their ability to destroy ICBM and SLBM warheads and come closer to meeting Nitze’s cost-effectiveness criterion? It is not just about a higher probability of hitting the target; the interceptor should not cost so much that an adversary could cheaply overwhelm the defense by adding warheads and decoys. A major factor affecting the answer to these questions will turn on the ability of radars, other sensors, and the interceptor itself to discriminate between warheads and decoys. GBI tests to date have not involved decoys and other countermeasures that are realistic.8

Another issue is whether there is some level of missile defense capability that the United States would consider adequate to deal with limited rogue-state attacks and, if so, what level that would be. Could that level be sufficiently low that it would not create incentives for Russia and China to increase their strategic offensive forces?

The review should also examine the pluses and minuses of giving SM-3 and THAAD missiles the capability to intercept ICBM- and SLBM-class targets. Creating a second layer to defend the United States may seem attractive. Yet, such capabilities may count for little if the warships carrying the SM-3 interceptors and the ground units equipped with THAAD missiles are deployed forward and thus not positioned to defend the U.S. homeland. Even so, these systems could still incentivize Russia and China to increase their ballistic missile numbers out of concern that the interceptors could be redeployed if needed to defend the homeland.

Missile Defense and Arms Control

Following up on the summit between Biden and Russian President Vladimir Putin in June, U.S. and Russian officials began a strategic stability dialogue. Senior U.S. officials have said they want to reduce reliance on nuclear arms and engage Russia in a negotiation to cover all U.S. and Russian nuclear warheads, not just the deployed strategic warheads constrained by New START. Russian officials, however, have different priorities, including missile defense and long-range conventional-strike weapons.9 Reconciling these competing priorities could pose a major challenge.

The Trump administration’s review stated repeatedly that it would not agree to any limits on missile defenses that are intended to protect against rogue-state ballistic missiles. The Obama administration resisted Russian efforts to bring missile defense into the New START negotiations. The treaty preamble notes the “interrelationship between strategic offensive arms and strategic defensive arms,” but contains just one limit related to missile defense: a prohibition on converting ICBM or SLBM launchers so that they could launch missile interceptors.

One option for Washington is to continue to reject any constraints on missile defense. Unlike its Russian and Chinese counterparts, the U.S. military seems relatively unconcerned about the ability of adversary missile defenses to prevent U.S. ICBMs and SLBMs from reaching their targets. Notably, the Pentagon has not sought limits to constrain Russian missile defenses.

A launcher for a Russian anti-ballistic missile system was on display at the Russian Army's 2021 International Military and Technical Forum near Moscow.  (Photo by Sergei Karpukhin\TASS via Getty Images)Russian officials, however, could continue to insist on addressing missile defense. In that case, the Biden administration would have to decide whether the U.S. interest in a negotiation covering all U.S. and Russian nuclear weapons is such that it justifies agreeing to confidence- and transparency-building measures on missile defense or even actual limits.

Washington has offered transparency measures in the past. In 2013 the Obama administration proposed an agreement mandating annual data exchanges with current numbers of certain missile defense systems, such as interceptors and radars, and projected numbers each year for the next 10 years. The Russians did not take up the idea.10 Other proposals have included exchanging notifications on missile launches through a Joint Data Exchange Center, an idea that was agreed by the United States and Russia in 1998 but never implemented. The idea was revived in 2011 in discussions regarding a NATO-Russian data fusion center, but the sides reached no agreement.

Washington might also consider actual limits. One idea would be to offer a time-limited ban on the testing or deployment of space-based interceptors. Such systems could pose stability concerns, but neither the U.S. military nor the Russian military has them at present.11 This could offer a way to defuse Moscow’s worst-case fears about U.S. plans.

Another approach would entail numerical constraints on missile defenses. Assuming that a successor to New START would have a duration of 10 to 15 years, with a provision for extension, Washington and Moscow might reach an agreement of similar duration on missile defenses. It appears possible to have a limit that would accommodate the U.S. desire for a capability to defend against limited rogue-state ballistic missile attacks while offering Russia and perhaps China assurance that their strategic ballistic missiles would not require a build-up.

For example, a limit of 100 to 125 strategic interceptors, along with transparency and verification measures, would permit the U.S. military to boost the number of GBIs beyond the 64 it plans to have in 2030. That would provide significant capability against North Korea, but should leave peer competitors with confidence that they could still hold at risk a large number of targets in the United States. Trying to include SM-3, THAAD, S-400, and S-500 missiles would significantly complicate this arrangement.

For the administration, negotiating such GBI limits would prove controversial politically, given the support among Republicans in Congress for strategic missile defense. It might also turn out to be only one of several issues the Russians try to link to a U.S.-desired limit on all U.S. and Russian nuclear weapons.

It appears, however, that a limit could be possible that would allow the U.S. military to maintain a capability to defend against a rogue-state ballistic missile attack while assuring Russia and China that their nuclear deterrents would not be rendered ineffective. That could enable further U.S.-Russian nuclear reductions, lower the likelihood of U.S.-Russian and U.S.-Chinese arms races, and perhaps open the door to a productive strategic stability discussion with Beijing. Hopefully, the Missile Defense Review will offer Biden such options.


1. Greg Thielmann, “The National Missile Defense Act of 1999,” Arms Control Today, July/August 2009, https://www.armscontrol.org/act/2009-07/national-missile-defense-act-1999.

2. Kingston Reif, “Trump Seeks Missile Defense Buildup,” Arms Control Today, March 2019, pp. 30–32, https://www.armscontrol.org/act/2019-03/news/trump-seeks-missile-defense-buildup.

3. Office of the U.S. Secretary of Defense, “2019 Missile Defense Review,” n.d., https://www.defense.gov/Portals/1/Interactive/2018/11-2019-Missile-Defense-Review/The%202019%20MDR_Executive%20Summary.pdf.

4. Cameron Tracy, “De-Hyping China’s Missile Test,” Union of Concerned Scientists, October 21, 2021, https://allthingsnuclear.org/guest-commentary/de-hyping-chinas-missile-test/.

5. “A Conversation With Vice Chairman of the Joint Chiefs of Staff General John E. Hyten,” The Brookings Institution, September 13, 2021, https://www.brookings.edu/wp-content/uploads/2021/09/fp_20210913_hyten_jcs_transcript.pdf.

6. Strobe Talbott, Master of the Game: Paul Nitze and the Nuclear Peace (New York: Alfred A. Knopf, 1988), p. 217.

7. Andrey Baklitskiy, James Cameron, and Steven Pifer, “Missile Defense and the Offense-Defense Relationship,” Deep Cuts Commission Working Paper, No. 14 (October 2021), pp. 23-24, https://deepcuts.org/images/PDF/DeepCuts_WP14.pdf.

8. David Wright, “Decoys Used in Missile Defense Intercept Tests, 1999–2018,” Union of Concerned Scientists, January 2019, https://www.ucsusa.org/sites/default/files/attach/2019/01/Missile-Defense-Intercept-Test-Decoys-white-paper.pdf.

9. Amy F. Woolf, “Nuclear Arms Control After the Biden-Putin Summit,” CRS Insight, IN11694, September 30, 2021, https://crsreports.congress.gov/product/pdf/IN/IN11694.

10. Steven Pifer, “Nuclear Arms Control Choices for the Next Administration,”
Brookings Arms Control and Non-Proliferation Series Paper, No. 13 (October 2016), p. 15, https://www.brookings.edu/wp-content/uploads/2016/10/acnpi_20161025_arms_control_choices_final.pdf.

11. James Timbie, “A Way Forward,” Daedalus, Vol. 149, No. 2 (Spring 2020): 190–204, https://doi.org/10.1162/DAED_a_01797.


Steven Pifer is a William Perry Fellow at the Center for International Security and Cooperation at Stanford University. This article is based on the Deep Cuts Commission working paper “Missile Defense and the Offense-Defense Relationship” co-authored with Andrey Baklitskiy and James Cameron.


As the United States and Russia contemplate new nuclear weapons reductions, the U.S. missile defense program stands as a complicating factor.

From Division to Constructive Engagement: Europe and the TPNW

December 2021
By Oliver Meier and Maren Vieluf

Europe remains deeply divided over the Treaty on the Prohibition of Nuclear Weapons (TPNW), with NATO membership the main political fault line between treaty critics and sympathizers. Since 2010, NATO has described itself as a nuclear alliance. In December 2020, all 30 of its members collectively stated their opposition to the ban treaty, but that appearance of unity is vanishing as the treaty picks up support in key allied nations.

Norwegian Prime Minister Jonas Gahr Store (R), shown hosting German President Frank-Walter Steinmeier in Oslo. Both countries recently announced that they will attend the first meeting of the states-parties of the Treaty on the Prohibition of Nuclear Weapons (TPNW) as observers in March 2022. (Photo by Håkon Mosvold Larsen / NTB / AFP via Getty Images)The 27 EU members, by contrast, have agreed to disagree on the TPNW. Three EU states are parties to the TPNW, while France, the only EU nuclear-weapon state, remains a staunch opponent. The other European states linger somewhere between these positions.

This disunity reduces European influence on the nuclear disarmament debate. In the past, when it acted jointly, the European Union was often able to facilitate global agreement on steps to reduce the role and number of nuclear weapons. The union has been a key force in shaping the multilateral arms control, disarmament and nonproliferation landscape. NATO, the EU, and Europeans could again assume such roles by constructively engaging with the ban treaty.

In this sense, the treaty is an opportunity, rather than an obstacle, to increase European agency on the way toward a world free of nuclear weapons.

So how can NATO, the EU, and European nations make the best use of the TPNW in striving toward nuclear disarmament? What options exist for EU and NATO members to reduce divisions over the treaty and build on the momentum it has created while respecting each other’s viewpoints? These questions are particularly acute as NATO revises its Strategic Concept and Europeans prepare for the 2022 Nuclear Nonproliferation Treaty (NPT) Review Conference in New York on January 4–28 and the first meeting of TPNW states-parties, which will take place in Vienna on March 22–24.

NATO’s Crumbling Unity

Until recently, NATO members collectively opposed the ban treaty. In the waning days of the Trump administration, the alliance argued that the TPNW “risks undermining the global non-proliferation and disarmament architecture” and called on “partners and all other countries to reflect realistically on the ban treaty’s impact on international peace and security.”1

In mid-October 2021, however, the newly elected Norwegian government, led by Prime Minister Jonas Gahr Støre, tore down the façade of unity by announcing its intention to attend the first meeting of the TPNW states-parties as an observer.2

In Germany, the parties in the new government pledged in their November 24 coalition agreement that Berlin would observe “in close consultation” with allies the first meeting of TPNW states-parties “in order to constructively accompany the intentions of the treaty.”

NATO’s hard-line stance against the treaty has stood on shaky ground for some time. In Spain, government coalition partners in 2018 informally agreed to join the treaty, although they subsequently failed to act on that commitment. The Dutch Parliament forced the government to take part in the 2017 UN General Assembly negotiations on the ban treaty. The Belgian government in 2020 pledged to explore how the treaty “can give new impetus to multilateral nuclear disarmament.”3 Thus, three of the five states that host U.S. nuclear weapons under NATO's nuclear sharing arrangements—Belgium, Germany, and the Netherlands—have struggled to stay faithful to the alliance position on the ban treaty.

French President Emmanuel Macron, shown at a NATO meeting in June in Brussels, has been leading the charge to keep the alliance from joining the Treaty on the Prohibition of Nuclear Weapons (TPNW). (Photo by THOMAS COEX/AFP via Getty Images)All of this is unlikely to sway treaty opponents. Paris is leading the charge. In a speech on French nuclear policy on February 7, 2020, French President Emmanuel Macron accused “advocates of abolition” of attacking the legitimacy of nuclear deterrence “where it is easiest, that is to say in…European democracies.”4 France is the only NATO state not participating in the alliance’s Nuclear Planning Group and its integrated nuclear policies. From its position of “splendid isolation” on nuclear issues within NATO, Paris has repeatedly vetoed progress on nuclear arms control and disarmament and is likely to remain adamantly opposed to any engagement with the ban treaty.5

As so often in NATO, in the end it will be the U.S. position that makes the difference. The Biden administration, however, has not indicated whether it intends to move away from the strong treaty opposition of his two immediate predecessors. A change of policy would require a step away from Biden’s past positions. In 2016, while Biden was vice president in the Obama administration, Washington had urged allies to reject any initiative to negotiate a treaty banning nuclear weapons. In an October 2016 nonpaper sent to the North Atlantic Council, NATO’s political decision-making body, even before treaty negotiations started in 2017, Washington warned that nuclear burden-sharing “could become untenable” and alliance consensus on the deterrence “could splinter” if any NATO member decided to join a ban treaty.6

The alliance is now struggling to square its rigid opposition to the treaty with changing political realities. When asked about his reaction to Oslo’s decision to observe the TPNW states-parties meeting, NATO Secretary-General Jens Stoltenberg, a Social Democrat like Støre and also a former Norwegian prime minister, replied that he expects allies to take NATO’s opposition to the treaty into account when addressing nuclear issues and that they should “consult closely with other NATO allies.”7

Stoltenberg made the remarks on the sidelines of a NATO defense ministers meeting in October that preceded a meeting of the Nuclear Planning Group, where ministers discussed the treaty issue. Stoltenberg’s chief of staff, Stian Jenssen, in an October 27 op-ed in the Norwegian newspaper VG warned that if Norway signs the TPNW, “it will endanger a wide range of defense and security policy cooperation with our closest allies.”8 So far, Oslo appears to be resisting peer pressure.

A Union Divided

Although NATO has worked hard to project unity on nuclear disarmament, the EU has effectively conceded failure.9 The breaking point came in 2015, two years before 122 states adopted the TPNW in the UN General Assembly. In 2013 and 2014, most EU member states had attended international conferences on the humanitarian and environmental impact of nuclear weapons in Oslo; Nayarit, Mexico; and Vienna. Nuclear-weapon state France shunned all three conferences. Reflecting this split, EU members failed to reach consensus on nuclear disarmament language for the 2015 NPT Review Conference. For the first time, in such a common decision, Europeans could only acknowledge their internal split by noting “the ongoing discussions on the consequences of nuclear weapons, in the course of which different views are being expressed, including at an international conference organized by Austria, in which not all EU Member States participated.”10

As discussions on a treaty prohibiting nuclear weapons moved ahead, this gap deepened, while support for the ban treaty globally as well as among Europeans grew broader. Many European non-nuclear-weapon states participated in a 2016 open-ended working group set up by the UN General Assembly to “take forward” multilateral nuclear disarmament negotiations. Yet, most EU and NATO members voted against a December 2016 UN resolution recommending full-fledged negotiations on a nuclear weapons prohibition treaty, even though the European Parliament had passed a resolution welcoming the ban treaty and calling on EU member states to support and “participate constructively” in the talks.11

In the end, many EU members stayed away from the 2017 UN General Assembly negotiations on the ban treaty. From Europe, only non-NATO members Austria, Cyprus, Ireland, Malta, and Sweden eventually voted for the legally binding TPNW, which comprehensively prohibits nuclear weapons. NATO member the Netherlands was the only state to vote against adoption of the ban treaty. Austria and Ireland, who are leading promoters of the treaty, and Malta are the only EU members among the 56 TPNW states-parties.

Another complication is that parliaments in many European countries and the European Parliament itself tend to be more open toward engagement with the TPNW than their governments. On October 21, the European Parliament issued a recommendation that referenced the security context and supported a step-by-step approach to nuclear disarmament but also recognized the TPNW as an expression of discontent by a majority of states with international disarmament efforts.12 Because the EU position remains fragmented along multiple lines, reaching a unified European position on the TPNW is likely to remain difficult.

Action Out of Diversity

Any future European policy on the treaty will have to take into account the broad spectrum of views on nuclear weapons that NATO and EU member states represent, with seemingly irreconcilable differences between the extremes. Although the TPNW has brought these divergences to light, it is not their root cause. Rather, they can be traced to different interests, cultures, historical experiences, and alliance relationships among European states. Differences also exist at the domestic level, as demonstrated by ongoing debates on national nuclear disarmament policies even in countries whose governments have charted a clear course against the TPNW.

Efforts to address European divisions on the TPNW and on nuclear weapons more generally will be arduous, but they are necessary to mitigate the counterproductive consequences of the European disputes and paralysis. For example, European engagement with a growing number of states supportive of the TPNW and the new ban treaty regime is a key to preserving Europe’s global role as an advocate of multilateralism. Europe is the region most opposed to the nuclear weapons ban. Of the 42 countries that reject the ban treaty, 31 are European nations. Europe’s legitimacy as a disarmament advocate depends on its ability to speak to ban treaty supporters and particularly the 86 countries that have signed or ratified the ban treaty.13

Europeans also need to find a more productive position toward the treaty if they want to reduce the risk of a widening rift among NPT member states on nuclear disarmament, which is a declared goal of the EU.

Meanwhile, NATO’s denunciation of the treaty is at odds with domestic developments in a number of key European countries. As an alliance of democracies, NATO should be responsive to such political shifts among its members and accommodate the momentum toward the TPNW. Unity has never been and cannot be based on demands to adhere to previously agreed positions when the context has changed.

Finally, European engagement is important for the treaty’s evolution. Many Europeans and NATO members criticize the TPNW’s verification provisions as too weak. They also argue that the treaty could undermine the NPT by establishing a competing legal and normative framework. Yet, the TPNW entered into force on January 22, 2021, and it is here to stay. If European critics want their concerns about the treaty to be addressed, they need to sit at the table. Staying on the political sidelines takes Europeans out of the nuclear disarmament game, so
how can Europeans engage constructively with the treaty through NATO, the EU, and individually?

Adjusting NATO's Policy

To maintain NATO unity, allies should agree on a more sustainable and forward-looking policy on the treaty. The current approach of completely rejecting the TPNW confronts members who are willing to recognize the treaty as a useful addition to the disarmament toolbox with an impossible choice. States such as Norway either can bow to pressures from the nuclear-weapon states and grudgingly step back into line or would have to break ranks with the alliance. For NATO, which is built on solidarity and rightly prides itself on being a political institution as well as a military alliance, such strongman tactics would be counterproductive.

At least implicitly, the alliance should acknowledge that engagement with the treaty, including attendance at TPNW meetings, does not weaken NATO coherence on nuclear matters, just like Dutch participation in the treaty negotiations had no discernible impact on perceptions of NATO unity or the effectiveness of NATO’s nuclear posture.

One opportunity to adjust NATO’s nuclear disarmament policies to political realities are discussions on a new strategic concept, to be agreed by the next alliance summit in Spain in the summer of 2022. Allies launched that process at their June 2021 summit in Brussels, when they confirmed their intention “to take all necessary steps” to maintain a credible deterrence posture but also committed to “support and strengthen arms control, disarmament, and non-proliferation.”14

In the new strategic concept, NATO could acknowledge the ban treaty as a good faith effort by the majority of states to eliminate the nuclear danger and build up the legal framework for the elimination of nuclear weapons.15 Such a statement would stop well short of support for the nuclear weapons prohibition as a generally accepted international norm.

Collectively and individually, NATO allies have persistently objected to the “argument that the TPNW reflects or in any way contributes to the development of customary international law.”16 Engagement with the treaty, including observing meetings of TPNW states-parties, can be consistent with such a policy if the allies continue to point out the substantive reasons for not acceding to the TPNW. For example, Israel, which possesses nuclear weapons, has observed several NPT review conferences and in 2015 submitted a working paper even though it does not want to join that treaty and nobody would expect that.

NATO could also engage with the TPNW by being more transparent about how it wants to ensure its nuclear planning is consistent with international humanitarian law. Scott Sagan and Allen Weiner have analyzed how the United States goes to great lengths to ensure that its nuclear war plans are consistent with these laws.17 They describe the procedures by which U.S. governments seeks legal advice so they follow international legal guidelines when developing nuclear war-fighting scenarios and preparing rules for nuclear weapons use. Naturally, nobody would expect NATO to discuss nuclear targeting plans. Yet, the alliance could describe the processes by which it implements its international humanitarian law obligations, including at which points and through which institutions allies individually can have an impact on such plans and their possible implementation in case of war.


Getting the EU’s Act Together

The 27 EU member states have been tiptoeing around their TPNW differences. Their foreign ministers on November 15 adopted a common position for the NPT review conference. The conclusions of the Council of the European Union do not even mention the ban treaty and merely note “the very severe consequences associated with nuclear weapons use”, while emphasizing “that all States share the responsibility to prevent such an occurrence from happening.”18

This lowest common denominator agreement does not put Europe in a position to shape discussions on nuclear disarmament. High-level engagement by European leaders will likely be needed to develop a more nuanced position. Such a stance could identify areas of agreement, along with those areas where Europeans continue to disagree or where they would push ban treaty supporters for further clarification. The difficult quest for greater convergence on nuclear disarmament is also mandated by the Treaty on the European Union, which calls on member states to “coordinate their action in international organizations and at international conferences.”19 More importantly, such a graded position on nuclear disarmament would help improve the credentials of the EU as a bridge-builder within the global nonproliferation regime.

At the review conference, the EU and member states should seek to reduce polarization over the TPNW by promoting language that recognizes the entry into force of the treaty and acknowledges the different perspectives on it. As Spanish analyst Clara Portela has emphasized, “Recognising the legitimacy of the treaty objectives does not equate to sympathising with the treaty, let alone to adhering to it.”20

The EU could also be a champion for those issues that are widely perceived as common ground among the nuclear disarmament camps. Thus, on nuclear risk reduction, Europeans could endorse a restatement by NPT states-parties of the Reagan-Gorbachev formula that “a nuclear war cannot be won and must never be fought.” Highlighting the importance of transparency as a method of fostering nuclear disarmament is another way in which Europeans can bolster the disarmament agenda, while enlarging the middle ground among NPT states-parties.

Looking beyond the review conference, the EU could seek to link the intersessional processes under the NPT and TPNW. Politically, this should be unproblematic as long as TPNW membership remains a subset of NPT membership. For example, NPT states-parties could invite the chairs of TPNW meetings of states-parties to issue formal reports at subsequent NPT preparatory committee meetings and at the 2025 NPT Review Conference. The EU could also propose that NPT states-parties establish an open-ended working group on ways and means to better integrate humanitarian law in nuclear weapons-related security concepts.21

The choice of Europe as the venue for the first TPNW meeting of states-parties places a special responsibility on Europeans to make the meeting a success. The EU itself could apply to observe the meeting as a relevant international or regional organization. The fact that the ban treaty remains contested must not be an insurmountable hurdle. The European Commission and the EU have observed the first meetings of states-parties to other treaties, even though at the time only some EU member states were parties.22

The European Parliament, which remains divided on the ban treaty, has not decided whether to send a delegation to the TPNW meeting, but the parliament’s Left and Green parliamentary groups will be represented. They also have asked the EU special envoy for nonproliferation and disarmament to represent the EU at the meeting as an observer regional organization.23 EU members Germany, Finland, and Sweden have already stated their intention to observe the meeting,24 and others might and should follow their lead.

Looking beyond the Vienna meeting, the EU could facilitate implementation of TPNW provisions on victim assistance and environmental remediation. France and the United Kingdom conducted all their nuclear weapons tests outside their European home territories and have a special responsibility to address the consequences of their nuclear weapons programs for humans and the environment. Such support would flow from the EU’s stated goal to “strongly support” the full implementation of multilateral disarmament, nonproliferation, and arms control treaties.25

The European Parliament could strengthen its role as a forum for diverse viewpoints on nuclear disarmament. It could seek to address the tension between deterrence and disarmament in a series of hearings involving both TPNW supporters and critics. The parliament could invite France and other NATO allies to explain how the alliance will ensure that nuclear planning takes into account international humanitarian law requirements to reduce human suffering in conflict. This could involve further discussion on whether any use of nuclear weapon is incompatible with international humanitarian law, as TPNW supporters argue.

Furthermore, the parliament could build on its traditional strength of using parliamentary diplomacy to influence the EU foreign and security policy agenda. For example, the parliament’s delegation to the NATO Parliamentary Assembly could use the results of hearings on the TPNW to initiate debate within the assembly. Such interaction could also help to broaden NATO-EU relations, which are now focused on defense issues, to include a common strategic culture on arms control, disarmament, and nonproliferation.

Participation of individual countries as observers in TPNW meetings is key. According to the Treaty on the European Union, states participating in international organizations and at international conferences shall uphold EU positions. A common EU approach toward the TPNW, as outlined above, would thus strengthen Brussels’ voice at the Vienna conference and other ban treaty gatherings.

In the end, Europeans need to change their perspective. Ban treaty supporters need to acknowledge the position of those who support nuclear deterrence. TPNW critics should view the ban treaty not as a problem, but as an opportunity to make progress toward reducing the role and number of nuclear weapons. Through a policy of constructive engagement with the TPNW, Europe can address its long-standing ambiguity on the role of nuclear weapons and strengthen European agency in nuclear disarmament. That would be good for Europe and, more importantly, for nuclear disarmament in general.



1. NATO, “North Atlantic Council Statement as the Treaty on the Prohibition of Nuclear Weapons Enters Into Force,” press release (2020) 131, December 15, 2017, https://www.nato.int/cps/en/natohq/news_180087.htm.

2. Alf Bjarne Johnsen and Gisle Oddstad, “Støre-Regjeringen Reiser Til Atomforbud-Konferanse: USA Stiller Spørsmål” [Støre government travels to nuclear ban conference: U.S. asks questions], VG, October 14, 2021, https://www.vg.no/nyheter/innenriks/i/mr33pE/stoere-regjeringen-reiser-til-atomforbud-konferanse-usa-stiller-spoersmaal.

3. Government of Belgium, “Accord de Gouvernement” [Government agreement], September 30, 2020, p. 78, https://www.belgium.be/sites/default/files/Accord_de_gouvernement_2020.pdf.

4. Emmanuel Macron, “Speech of the President of the Republic on the Defense and Deterrence Strategy,” February 7, 2020, https://www.elysee.fr/en/emmanuel-macron/2020/02/07/speech-of-the-president-of-the-republic-on-the-defense-and-deterrence-strategy.

5. Oliver Meier and Simon Lunn, “Trapped: NATO, Russia, and the Problem of Tactical Nuclear Weapons,” Arms Control Today, January 2014, pp. 18–24, https://www.armscontrol.org/act/2014_01-02/Trapped-NATO-Russia-and-the-Problem-of-Tactical-Nuclear-Weapons.

6. NATO, “United States Non-Paper: ‘Defense Impacts of Potential United Nations General Assembly Nuclear Weapons Ban Treaty; Note by the Secretary,’” AC/333-N(2016)0029 (INV), October 17, 2016.

7. NATO, “Doorstep Statement by NATO Secretary General Jens Stoltenberg Ahead of the Meetings of NATO Defence Ministers on 21 and 22 October at NATO Headquarters,” October 21, 2021, https://www.nato.int/cps/en/natohq/opinions_187624.htm.

8. Stian Jenssen, “Veien Mot En Verden Uten Atomvåpen” [The road to a world without nuclear weapons], VG, October 27, 2021, https://www.vg.no/nyheter/meninger/i/RrKzjd/veien-mot-en-verden-uten-atomvaapen.

9. The EU-related section of this article draws heavily on Tytti Erästö et al., “A Fresh Breeze for Nuclear Disarmament in Europe? Making Best Use of the Treaty on the Prohibition of Nuclear Weapons,” The Greens/EFA in the European Parliament, June 2021, https://europeecologie.eu/wp-content/uploads/2021/07/Nuclear-Disarmament_A-Fresh-Breeze.pdf.

10. Council of the European Union, “Council Conclusions on the Ninth Review Conference of the Parties to the Treaty on the Non-Proliferation of Nuclear Weapons,” No. 8079/15, April 20, 2015.

11. European Parliament, “European Parliament Resolution of 27 October 2016 on Nuclear Security and Non-Proliferation (2016/2936(RSP)),” October 27, 2016.

12. European Parliament, “Recommendation to the VPC/HR and to the Council in Preparation of the 10th Non-Proliferation of Nuclear Weapons Treaty (NPT) Review Process, Nuclear Arms Control and Nuclear Disarmament Options (P9_TA(2020)028),” October 21, 2020.

13. On the current status of support of the TPNW, see Nuclear Weapons Ban Monitor, https://banmonitor.org/ (accessed November 13, 2021).

14. NATO, “Brussels Summit Communiqué Issued by the Heads of State and Government Participating in the Meeting of the North Atlantic Council in Brussels 14 June 2021,” June 14, 2021, https://www.nato.int/cps/en/natohq/news_185000.htm (hereinafter “NATO Brussels summit communiqué”).

15. Daryl G. Kimball, “The Nuclear Ban Treaty: A Much-Needed Wake-Up Call,” Arms Control Today, November 2020, p. 3, https://www.armscontrol.org/act/2020-11/focus/nuclear-ban-treaty-much-needed-wake-up-call.

16. NATO Brussels summit communiqué.

17. Scott D. Sagan and Allen S. Weiner, “The Rule of Law and the Role of Strategy in U.S. Nuclear Doctrine,” International Security, Vol. 45, No. 4 (Spring 2021): 126–166.

18. Council of the European Union, “Council Conclusions on the 10th Review Conference of the Parties to the Treaty on the Non-Proliferation of Nuclear Weapons (NPT),” 13243/21, November 15, 2021.

19. Treaty of the European Union, art. 34.

20. Clara Portela, “The EU’s Arms Control Challenge: Bridging Nuclear Divides,” Chaillot Paper, No. 166, April 2021, https://www.iss.europa.eu/sites/default/files/EUISSFiles/CP_166.pdf.

21. See Thilo Marauhn, “Reducing the Role of Nuclear Weapons: A Role for International Law,” in Meeting in the Middle: Opportunities for Progress on Disarmament in the NPT, King’s College London Centre for Science and Security Studies, December 2019, pp. 34–37, https://www.kcl.ac.uk/csss/assets/meeting-in-the-middle.pdf.

22. See Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on Their Destruction, 2056 U.N.T.S. 211, September 18, 1997; Diplomatic Conference for the Adoption of a Convention on Cluster Munitions, “Convention on Cluster Munitions,” CCM/77, May 30, 2008.

23. Mounir Satouri (@MounirSatouri), “Together with fellow MEPs @Oezlemademirel @Brandobenifei and @Lukasmandl we will attend the First meeting of the #TPNW We Call on the @EUCOUNCIL to mandate the EU special envoy to attend #1MSP #nuclearban,” Twitter, September 7, 2021, 12:05 p.m., https://twitter.com/MounirSatouri/status/1435273050245697538.

24. Ann Linde, “Statement of Foreign Policy 2021,” Swedish Ministry of Foreign Affairs, February 24, 2021, https://www.government.se/speeches/2021/02/statement-of-foreign-policy/; Pekka Haavisto, “Speech by Pekka Haavisto, Minister for Foreign Affairs, at the Meeting of Heads of Mission, 23 August 2021,” Ministry of Foreign Affairs of Finland, August 23, 2021, https://um.fi/speeches/-/asset_publisher/up7ecZeXFRAS/content/ulkoministeri-pekka-haaviston-puhe-suurlahettilaskokouksessa-23-8-2021: German Coalition Agreement, 2021.

25. European Union, “Shared Vision, Common Action: A Stronger Europe; A Global Strategy for the European Union’s Foreign and Security Policy,” June 2016, pp. 41–42, https://eeas.europa.eu/sites/eeas/files/eugs_review_web_0.pdf.

Oliver Meier is a senior researcher in the Berlin office of the Institute for Peace Research and Security Policy at the University of Hamburg (IFSH). Maren Vieluf is a researcher in the Challenges to Deep Cuts Project in the Berlin office of IFSH.

Deep divisions over the Treaty on the Prohibition of Nuclear Weapons are hampering the ability of Europe
to influence the goal of nuclear disarmament.

Revitalizing the Missile Technology Control Regime

December 2021
By William Alberque and Timothy Wright

More states than ever have cruise and ballistic missiles in their arsenals. In 1987, for instance, only three states—the United Kingdom, the United States, and the Soviet Union—possessed cruise missiles. Twenty-three states and one nonstate actor possess them today. The demand for these systems is partly driven by their increasing utility, resulting from exponential improvements in survivability, accuracy, and speed. Missile proliferation has been accelerated by the spread of enabling technologies that have allowed more actors to overcome previous structural hurdles.

India unveils a second generation rocket, called the Geo Synchronous Launch Vehicle (GSLV). It is capable of placing 2,000 kilogram satellites in orbit roughly 36,000 kilometers above Earth. In 2001, the project was sanctioned by the United States under the Missile Technology Control Regime (MTCR). (Photo by Pallava Bagla/Corbis via Getty Images)These and other threats present significant challenges to those who seek to maintain global governance and restrictions on missiles. The landscape has altered dramatically since the end of the Cold War, when the most significant effort to stop the spread of missile technology through multilateral export controls, known as the Missile Technology Control Regime (MTCR), was established. As a result, there is an urgent need for innovative thinking about how to reform the existing governance structure. The responsibility to do so needs to be global, and the choices will be difficult. Political will at the highest level is essential.

The Regime

The MTCR is a technology-focused export control regime, comprising a voluntary association of 35 member states that apply agreed standards, known as the Guidelines for Sensitive Missile-Relevant Transfers, to limit the export of technology that can be used to deliver weapons of mass destruction (WMD). It differs from most other export control regimes by creating a presumption of exportation denial of longer-range ballistic and cruise missiles and unmanned aerial vehicles (UAVs), as well as the most worrisome technology related to their manufacture.1 The MTCR puts the burden for compliance on the sellers rather than the buyers, but it contains no verification or enforcement requirements. The guidelines include a detailed technical annex that defines complete delivery systems and production facilities, known as Category I items, as well as supporting equipment, software, and technologies that could contribute to building delivery systems, known as Category II items.2

The scope of the MTCR has expanded considerably over the past three decades. The regime was secretly negotiated by the Group of Seven (G7) nations and announced on April 16, 1987, to address what was seen as an urgent proliferation crisis driven by detected or suspected sales of nuclear-capable missiles by the Soviet Union and China, as well as emerging and advanced missile programs in countries such as Argentina, Brazil, India, Israel, Pakistan, and South Africa. The MTCR also was influenced by a series of breakthroughs in bilateral negotiations between the Soviet Union and the United States, including a decision after the 1986 Reagan-Gorbachev summit in Reykjavik to conclude the Intermediate-Range Nuclear Forces (INF) Treaty, which banned both countries from possessing ground-launched cruise and ballistic missiles with ranges between 500 and 5,500 kilometers. The INF Treaty drafters intended for the MTCR to reinforce the treaty by providing global limits on the spread of missiles, missile components, and related technology and by focusing on ballistic missiles capable of delivering a 500-kilogram warhead a distance of 300 kilometers. This range was based on the performance of the Soviet-designed Scud short-range ballistic missile, which at the time was thought to be the technological threshold for a missile capable of delivering a nuclear warhead.

In 1992, the regime was altered to include all potential systems capable of delivering chemical, biological, and nuclear weapons, except manned aircraft, thus expanding the regime’s scope to include UAVs, target drones, and cruise missiles. Members also began issuing regular appeals to nonmember states to apply the MTCR guidelines on a voluntary basis and declare themselves “universal adherents.”3 Following the September 11 attacks, the MTCR’s goals were expanded to address missile transfers to nonstate actors, as reinforced by UN Security Council Resolution 1540.

Regime Strengths

The MTCR is a supply-side arrangement that has evolved into a global norm, increasing security and helping to curb missile proliferation. This occurred despite the absence of a universal, legally binding treaty within the UN framework. The MTCR is not a panacea and has many faults, but it also has strengths and demonstrable successes.

The regime’s membership incorporates many of the most important state possessors and producers of missiles and related technologies. The roster grew from the original G7 states in 1987 to 32 members in just more than a decade. Further growth has since slowed, with only three additional states—Bulgaria, India, and South Korea—joining since 1998. Given that New Delhi possesses nuclear weapons, Seoul historically has employed a nuclear hedging strategy, and both states operate diverse missile programs, their inclusion in the regime marks an important effort to control potential missile proliferation.4

Furthermore, by limiting membership to the most significant states that possess delivery systems and related technologies, the MTCR’s three expert working groups can operate at the highest levels of detail.5 The resulting MTCR annex is an impressive achievement, with an agreed scope and a set of definitions of the highest technical complexity. The expertise provided by MTCR working group members also means the annex is continually updated by some of the world’s most knowledgeable missile technology experts. The annex has such as an extraordinary level of detail, relevance, and scope that it forms the basis for stricter missile technology controls enacted under other UN Security Council resolutions.6

One weakness of the MTCR is that China, which has a rapidly growing arsenal of nuclear weapons, like the missiles shown here, does not belong to the regime.  (Photo by Feng Li/Getty Images)The adaption of a fully elaborated and well-defined list of WMD-capable delivery systems creates a nonproliferation norm through the presumption of denial standard for Category I items and establishes the principle that states pursuing missile exportation or acquisition will be scrutinized. Paragraph 2 of the MTCR guidelines notes that “particular restraint will be exercised in the consideration of Category I transfers regardless of their purpose, and there will be a strong presumption to deny such transfers.”7 The particular restraint and presumption to deny standards for the most concerning systems and their enabling technologies therefore put pressure on countries, whether regime members or not, to adhere to the guidelines.

Finally, the MTCR has a program of regular outreach that seeks to persuade nonmember states to adopt MTCR controls through partnership and the “adherent” system. This is an important evolution because it has allowed nonmember states to adhere not simply rhetorically but through a formal mechanism.8 Estonia, Kazakhstan, and Latvia are recognized adherents while China and Israel have declared themselves as self-adherents.

Together, these factors have played a role in slowing or stopping several significant missile programs, including the joint Argentine-Egyptian-Iraqi Condor II ballistic missile program; the missile stockpiles of former Warsaw Pact states that had aspirations to join the EU, NATO, and the nuclear Nonproliferation Treaty; the missile programs of India and South Korea; and Libya’s Category I missiles.9 Although other factors, including diplomatic and economic pressures and incentives, encouraged these positive outcomes, the MTCR regime played an important contributory role.

Unilateral and Institutional Constraints

Despite these MTCR strengths, if a state decides to acquire advanced missile technology and is willing to pay the costs, it likely will succeed. Although the MTCR can delay, complicate, and raise the political and economic costs of such a decision, the regime is unable to prevent or reverse missile proliferation unless member states wield an array of other separate security, diplomatic, or economic incentives or punishments. Moreover, the regime has an innate discriminatory nature that has been exploited as MTCR members trade among themselves and sometimes seek special exemptions or allowances for trade or other forms of support with states inside and outside the regime.

The MTCR’s powerlessness to unilaterally prevent missile proliferation reflects its voluntary character, the lack of formal linkages to the UN system, and the facts that adherence to the guidelines is entirely self-enforced and that no sanctions or penalties can be imposed on states that circumvent the regime.10 The regime also has shortcomings in membership because it does not include states with significant ballistic and cruise missile programs, including China, Iran, and North Korea. Relatedly, the MTCR’s lack of commitment to halt missile development, production, or trade among members has led some nonmembers states to criticize its legitimacy. The impression that the MTCR is an exclusive club of haves and have-nots has resulted in some nonmembers labeling it a cartel.11

There are other institutional weaknesses that affect efficacy. The regime requires consensus for decisions, thereby limiting its ability to quickly resolve political issues or adapt to change. The consensus rule has prevented expanding membership to key states due to unrelated political considerations, such as Italy blocking India’s membership bid due to a dispute following New Delhi’s arrest of two Italian servicemen.12 An area where advancements in capabilities have outstripped the agreed MTCR guidance is UAV technology, whose utility has improved substantially since MTCR restrictions on exporting these systems were adopted in 1992. These restrictions were implemented due to concerns that UAVs might be used for delivering chemical, biological, and nuclear weapons. Given the growing global demand for acquiring these systems for conventional purposes by non-MTCR members, such as China, the United States proposed changes to allow the exportation of UAVs that cannot exceed an airspeed of 800 kilometers per hour.13 After its proposals failed to gain consensus, Washington took unilateral action.14 In this case, consensus changes to the regime were hampered by a combination of problems, including the disconnect between technical experts and political representatives and competing economic and nonproliferation interests between members.15

The regime is also weakened by its lack of a permanent secretariat. This places a strain on France, which serves as the MTCR host and “point of contact,” and prevents the development of a dedicated, professional, international staff that can carry forward the regime’s work in a consistent, focused way. The lack of a permanent secretariat can also impede faith in the implementation of the agreement, especially if a nonmember perceives itself to have differing interests from the point of contact.

Although the MTCR’s rotating chair demonstrates a lack of institutional bias and a sharing of leadership, this feature creates significant risks, including the possibility that a chair may lack the political clout or will to achieve anything during their term and that institutional knowledge will be lost and the momentum behind projects will dissipate each time the chair changes. The rotating chair has created uncertainty when countries have hesitated to volunteer for the chairmanship, requiring last-minute lobbying to persuade somebody to fill the position. This issue was temporarily resolved during the 2019 plenary meeting, as members agreed to three successive chairs until 2023, but no additional chairmanship appointments have been publicly agreed beyond that.16

The lack of technical expertise among some member and adherent states and the disconnect between technical experts and political representatives can burden MTCR chairs and stretch limited national resources and capabilities. Encumbrances in carrying out routine tasks can limit progress. For instance, in some member states, export control and arms control functions are housed in different departments; and economic and trade policies are considered separately from security policy, often to the detriment of all these equities.

Finally, the MTCR does not address vertical proliferation among missile producers, thus allowing these member states to develop and expand existing domestic programs. The regime also allows for continued cooperation among states with preexisting programs, such as the Trident II submarine-launched ballistic missile, which the UK and the United States both operate as part of their respective sea-based nuclear deterrents. In other instances, some states, such as South Korea, have received waivers and been able to expand their domestic ballistic and cruise missile production capability despite earlier efforts to limit their ambitions. Earlier this year, the United States agreed to lift all limits on South Korean missile development due to political pressure from Seoul following increased North Korean nuclear and missile threats.17

Incremental Reform of the MTCR

Advocates for incremental reform of the MTCR argue that the work needs to focus on achievable, “low-hanging fruit,” which is often described but somehow never harvested. Incremental approaches should likely be informed by Track 1.5 meetings, a UN Security Council Resolution 1540 comprehensive review, and possibly a convention of a high-level UN group of governmental experts with a clear mandate to identify specific proposals, backed by substantial political will to use upcoming MTCR plenaries to agree to changes. Although the previous UN expert group in 2008 was unable to agree to substantial changes, the current global environment of accelerating missile proliferation may provide a new impetus for future agreement.

Given the institutional weaknesses of the MTCR, establishing a permanent secretariat to support and improve national expert training, implementation of the regime, and information sharing should be a priority. It could help overcome the distrust of nonmember states who sometimes view the point of contact as a biased actor. Other aims should be opening internal processes and decision-making to greater scrutiny and increasing MTCR engagement with other global and regional initiatives, such as the Missile Dialogue Initiative and the Warsaw Process Missile Proliferation Working Group.18

Such changes would initially require coordination and agreement in small groups followed by sequencing and leadership of outreach to members and nonmembers alike to build support for any changes. Progress would require sustained engagement within and among governments to bring together the correct mix of diplomats and experts on arms control, export control, missile and UAV technology, security policy, and intelligence. Even so, such an incremental process would take years and would have a low likelihood of success based on the failure of all recent efforts toward systemic reform. That calculation is only likely to change if there is some catalytic event that is so broadly destructive or destabilizing that the major powers begin spending the political capital necessary to make significant changes.

Russian mobile launchers carrying Yars nuclear armed intercontinental ballistic missiles are among the systems limited by the Missile Technology Control Regime (MTCR). (Photo by Sergei Fadeichev\TASS via Getty Images)There is hope for a more radical approach. Russia, as it took over the rotating MTCR chair in 2021, expressed interest in radical reforms to global missile governance. Deputy Foreign Minister Sergei Ryabkov, in his address to the MTCR plenary on October 6, called for an experts group to begin talks on a legally binding global missile treaty.19 So far, the membership has not expressed enthusiasm for this proposal, but it may be an opening for more ambitious ideas. A radical approach could also be timelier, insofar as it may channel Russia’s enthusiasm for change and make some incremental changes more achievable than they are today.

Radical Approaches to Change

More drastic approaches to MTCR reform should be designed so that even if such efforts fall short, they change the terms of the debate and help advance the goals anticipated under the incremental reform rubric.20 Support for a broad package of radical changes does not require abandoning the existing MTCR framework, but rather reconsidering the scope, scale, membership, and impact of the regime. This could include, for instance, setting limits on the spread of UAVs capable of launching missiles above a certain speed and establishing a code of conduct for their use that would incorporate the standards of international humanitarian law.

Rather than focusing singly on the MTCR, states could deliver more useful benefits and provide a radical change to global missile controls by strengthening and increasing the MTCR’s complementarity with other frameworks, including with UN Security Council Resolution 1540, which is aimed at preventing nonstate actors from acquiring weapons of mass destruction and the means of delivery, and the Hague Code of Conduct Against Ballistic Missile Proliferation, a voluntary transparency and confidence-building instrument that aims to establish norms around ballistic missile proliferation.21 This initiative could include having the UN General Assembly pass a resolution to form an experts group to design a global agreement within the United Nations to subsume the MTCR and the Hague Code of Conduct. The eventual structure could include transforming the MTCR into a technical body focused on the MTCR annex and its expert working groups, which would present options to the Hague Code of Conduct, which would act as a plenary body for decisions and promulgation.

A larger change could involve expanding the Resolution 1540 infrastructure into a joint secretariat with the MTCR and Hague Code of Conduct in Vienna, with a secretary-general appointed by the UN and an ad hoc national chair rotating every two years, along with regular monthly technical and political meetings and an annual decision-making plenary. Considering the dual capability of civilian space technology, a possible agreement could also integrate the UN Office for Outer Space Affairs, the MTCR, and the Hague Code of Conduct, with the proposed secretariat providing a common operating framework that encompasses national policy, launches, and situational awareness in outer space under one umbrella. Finally, incorporating the Proliferation Security Initiative into the expanded organization would enable broader cooperation on interdiction.

Other functions could include the development of new confidence-building measures modeled after U.S.-Russian missile launch and strategic exercise notifications and the establishment of national risk reduction centers and a central data-fusion center to increase transparency and minimize risk.22 Sharing the benefits of the peaceful pursuit of space launch capabilities, including communications and navigation, with less capable member states would reduce the conceptions of have and have-not states.

Setting forth obligations regarding transparency and responsibility for corporations and other entities seeking or using satellite launch vehicle capabilities would ensure that commercial firms are not overlooked and minimize disguised civil-military fusions. Incorporating targeted UN implementation, enforcement, and reporting measures, based on those used to limit the missile programs of Iran and North Korea, into the agreement would provide codified penalties for violations. This could be taken one step further by codifying all of the above into a unified agreement, which could be named the Missile and Space Technology Transparency and Control Treaty (MASTTRACT), or the Missile Nonproliferation Treaty (MNT).

Limited Time to Act

The MTCR’s original purpose was to limit the proliferation of missiles that could deliver a 500 kilogram warhead a distance of 300 kilometers. This was later expanded to limit the spread of all nonpiloted means of delivering chemical, biological, and nuclear weapons. The regime was also meant to establish a norm that would restrict activity by nonmembers, some of whom were active proliferators of missile systems.

The MTCR has successfully facilitated cooperation by like-minded states to limit the proliferation of missiles and associated technologies, and it has bought time by delaying and increasing the cost of these practices. Despite increases in membership and the broad application of the regime’s guidelines, the number of states that ignore the guidelines while possessing or producing systems capable of accurately delivering large payloads over long distances continues to increase. The regime has also allowed members to trade freely among themselves and to seek special exemptions or allowances for trade with states as needed, such as South Korea. These practices have allowed for exploitation, thereby weakening the efficacy of the regime and its relevance.

Meanwhile, many more countries have proven that they can overcome the technological hurdles that previously limited missile proliferation, including the ability to manufacture solid fuel propellants, miniaturized turbofan engines, complex rocket motors, precision guidance systems, and special-property materials for reentry vehicles.23 Most damningly, these capabilities are now within the grasp of nonstate actors, ranging from commercial firms such as Space X to insurgent groups such as Ansarullah.

Despite its successes, the MTCR has failed to substantially resolve the concerns that it was created to address. Further proliferation of missile technology, as demonstrated by Iran, North Korea, and nonstate actors such as Ansarullah and Hezbollah, is only a matter of time despite the likelihood of the UN Security Council adopting additional resolutions tailored specifically to prevent this from happening.

Reform of the MTCR’s scope and functions is the only viable path forward, and an incremental approach could deliver tremendous benefits. Achieving this will require long-term focus and significant political, intellectual, and financial commitments. Although a more radical approach is unlikely to be agreed among members and adherent states, if such radical ideas are pushed within an unofficial Track 1.5 format, they could shock the system toward incremental reform.

Recent decisions on waivers by the MTCR and unilateral reinterpretations of the regime’s scope are likely the beginning of a broader shift by member states away from the presumption of denial that is at the core of the MTCR. Even if no reform action is taken, it is likely that the UN Security Council will still look to the MTCR for guidance on national export controls, but this will occur as an increasing number of technologically advanced states outside the regime ignore or take advantage of MTCR limitations on member states by selling restricted items to interested buyers.24 Advanced missile and space launch technology will spread in the meantime as the world continues down a dangerous path. The world therefore must decide not only if it wants reforms that strengthen the MTCR, but whether it is really committed to making that reform happen.



1. Missile Technology Control Regime (MTCR), “MTCR Guidelines and the Equipment, Software and Technology Annex,” n.d., https://mtcr.info/mtcr-guidelines/ (accessed November 20, 2021).

2. MTCR, “Missile Technology Control Regime Equipment, Software and Technology Annex,” MTCR/TEM/2019/Annex, October 11, 2019.

3. MTCR, “Plenary Meeting of the Missile Technology Control Regime - Oslo, Norway — 29 June – 2 July 1992,” July 2, 1992, https://mtcr.info/plenary-meeting-of-the-missile-technology-control-regime-oslo-norway-29-june-2-july-1992/ (hereinafter 1992 plenary meeting).

4. Mark Fitzpatrick, Asia’s Latent Nuclear Powers: Japan, South Korea and Taiwan (New York: International Institute for Strategic Studies, 2016), intro.

5. There are three annual meetings of experts from Missile Technology Control Regime (MTCR) partners in support of the annual plenary meeting: the Technical Experts Meeting, which seeks agreement on updates to the MTCR Equipment, Software and Technology Annex; the Information Exchange; and the Licensing and Enforcement Experts Meeting.

6. Brendan Murphy, “Public Statement From the Plenary Meeting of the Missile Technology Control Regime, Sochi, 8 October 2021,” MTCR, October 26, 2021, https://mtcr.info/public-statement-from-the-plenary-meeting-of-the-missile-technology-control-regime-sochi-8-october-2021/ (hereinafter 2021 MTCR public statement).

7. MTCR, “Guidelines for Sensitive Military Missile-Relevant Transfers,” n.d., https://mtcr.info/guidelines-for-sensitive-missile-relevant-transfers/ (accessed November 20, 2021).

8. 1992 plenary meeting.

9. For further details on the specifics of some of these actions, see Nuclear Threat Initiative, “Argentina Missile Facilities,” December 7, 2011, https://www.nti.org/analysis/articles/argentina-missile-facilities/; “Bulgaria, Slovakia Still Hold SS 23s,” Arms Control Today, September 1997, https://www.armscontrol.org/act/1997-09/arms-control-today/bulgaria-slovakia-still-hold-ss-23s; Kelsey Davenport, “Chronology of Libya's Disarmament and Relations With the United States,” Arms Control Association, March 2021, https://www.armscontrol.org/factsheets/LibyaChronology.

10. Jeffrey Lewis, “Storm Shadow, Saudi and the MTCR,” Arms Control Wonk, May 31, 2011, https://www.armscontrolwonk.com/archive/204051/saudi-arabia-storm-shadow-the-mtcr/.

11. Baqir Sajjad Syed, “Why Pakistan Doesn’t Want to Join the MTCR,” Dawn, June 30, 2016.

12. Tom Kington and Vivek Raghuvanshi, “Italy Blocks Indian Application to MTCR,” Defense News, October 17, 2015.

13. Valerie Insinna and Aaron Mehta, “Here’s How the Trump Administration Could Make It Easier to Sell Military Drones,” Defense News, December 19, 2017.

14. Daryl G. Kimball, “U.S. Reinterprets MTCR Rules,” Arms Control Today, September 2020, p. 32.

15. Mike Stone, “U.S. Relaxes Rules to Export More Aerial Drones,” Reuters, July 24, 2020.

16. Jeffrey Taylor, “Public Statement From the Plenary Meeting of the Missile Technology Control Regime, Auckland, 11 October 2019,” MTCR, October 18, 2019, https://mtcr.info/public-statement-from-the-plenary-meeting-of-the-missile-technology-control-regime-auckland-11-october-2019/; 2021 MTCR public statement.

17. Timothy Wright, "U.S. and South Korea Scrap Ballistic Missile Range Limits," IISS, June 2, 2021, https://www.iiss.org/blogs/analysis/2021/06/us-south-korea-ballistic-missile-range-limit.

18. IISS, “Missile Dialogue Initiative,” n.d., https://www.iiss.org/research/defence-and-military-analysis/missile-dialogue-initiative (accessed November 20, 2021); U.S. Department of State, “Missile Proliferation Working Group Summary Statement,” November 15, 2019, https://2017-2021.state.gov/missile-proliferation-working-group-summary-statement/index.html.

19. 2021 MTCR public statement.

20. William Alberque, “Revitalising Arms Control: The Missile Technology Control Regime (MTCR) and the Hague Code of Conduct Against Ballistic Missile Proliferation (HCoC),” IISS, November 2021, https://www.iiss.org/-/media/files/research-papers/iiss_revitalising-arms-control-the-mtcr-and-the-hcoc_mdi-02112021.pdf.

21. Kolja Brockmann, “Controlling Ballistic Missile Proliferation: Assessing Complementarity Between the HCoC, MTCR, and UNSCR 1540,” HCoC Research Papers, No. 7 (June 2020), https://www.nonproliferation.eu/hcoc/wp-hcoc/uploads/2020/06/Assessing-the-complementarity-vf.pdf.

22. U.S. Department of State Bureau of Arms Control, Verification, and Compliance, “Memorandum of Understanding on Notifications of Missile Launches (PLNS MOU),” December 16, 2000, https://2009-2017.state.gov/t/avc/trty/187152.htm.

23. Amy Nelson and T.X. Hammes, “Inevitable Bedfellows? Cooperation on Military Technology for the Development of UAVs and Cruise Missiles in the Asia-Pacific,” IISS, July 28, 2020, https://www.iiss.org/-/media/files/research-papers/cooperation-on-military-technology-for-the-development-of-uavs-and-cruise-missiles-in-the-asiapacifi.pdf.

24. Bradley Bowman, Jared Thompson, and Ryan Brobst, “China’s Surprising Drone Sales in the Middle East,” Defense News, April 23, 2021.

William Alberque is director of strategy, technology and arms control at the International Institute for Strategic Studies (IISS). Timothy Wright is a research analyst and program administrator for defense and military analysis at the IISS.

The 34-year-old MTCR, an international export control regime, has slowed or stopped some significant missile programs but needs reform to deal with the increase in states adding ballistic and cruise missiles to their arsenals.

The Australia-UK-U.S. Submarine Deal: Submarines and Safeguards

December 2021
By Laura Rockwood

If the AUKUS deal, under which the United States and the United Kingdom agreed to sell nuclear-powered submarines to Australia, goes forward, it will have a precedent-setting effect on the International Atomic Energy Agency (IAEA) safeguards implemented pursuant to the nuclear Nonproliferation Treaty (NPT), which is formally known as the Treaty on the Non-Proliferation of Nuclear Weapons.

Rafael Mariano Grossi, director-general of the International Atomic Energy Agency, has said that the special arrangement that Australia must negotiate to acquire nuclear-powered submarines promised by the United Kingdom and the United States will require the agency “to dot the I’s and cross the T’s, which has never been done before, and it’s a very, very demanding process.” (Photo by Mandel Ngan/POOL/AFP via Getty Images)These technical measures are at the heart of international efforts to prevent the spread of nuclear weapons. They are accepted by non-nuclear-weapon state-parties to the NPT through the conclusion of comprehensive safeguards agreements. Under these agreements, the agency is empowered to verify independently that such states are complying with their obligation not to divert nuclear material for use in nuclear weapons or other nuclear explosive devices. Australia concluded such an agreement in 1974.1

The NPT does not prohibit the use of nuclear material for certain non-explosive military uses, such as nuclear naval propulsion. To accommodate that possibility, the comprehensive safeguards agreement includes a provision that allows a state to request the withdrawal of nuclear material from safeguards for use in such a nonprohibited activity. Prior to doing so, the state must conclude a separate, specific arrangement with the IAEA.

In the 40 years since the IAEA has been implementing comprehensive safeguards agreements, it has never concluded such an arrangement. If one results from the AUKUS deal, Australia, the IAEA, and the other two participant countries will have to take great care to avoid creating a loophole in such agreements that could provide cover for the diversion of nuclear material for use in a nuclear weapons program.

As IAEA Director-General Rafael Mariano Grossi stressed to reporters on October 21, “There has to be a specific arrangement with the IAEA” that will require the agency “to dot the I’s and cross the T’s, which has never been done before, and it’s a very, very demanding process.”

Legal Framework

All comprehensive safeguards agreements are based on INFCIRC/153, a document negotiated in the 1970s by a committee of the IAEA Board of Governors that was open to all members of the agency.2 As regards submarines, the most important provision is paragraph 14, entitled “Non-application of safeguards to nuclear material to be used in non-peaceful activities.” It contains the procedures to be followed in the event that a state wishes to “exercise its discretion to use nuclear material required to be safeguarded under the agreement in a nuclear activity which does not require the application of safeguards.” This is often referenced as the “withdrawal” of nuclear material from safeguards, distinguishing it from provisions related to the termination of safeguards on nuclear material, for example, if the material has become practicably irrecoverable, and to the exemption of nuclear material from safeguards, such as for use in a non-nuclear activity.

Pursuant to paragraph 14, before any nuclear material may be withdrawn from safeguards for use in such an activity, the IAEA and the state concerned must “make an arrangement” so that safeguards will not be applied only while the nuclear material is being used in that activity.

The drafters of INFCIRC/153 insisted on this as a means of balancing a state’s interest in protecting militarily sensitive information while ensuring, to the extent possible, that the withdrawal of nuclear material from safeguards would not create opportunities for the state to divert the material for prohibited purposes.

The withdrawal of nuclear material under paragraph 14 is not automatic and is not intended as a blanket exemption for nuclear material, facilities, or activities due to their military nature. A state may not withdraw nuclear material from safeguards without invoking paragraph 14 and concluding an arrangement with the IAEA.

Only Canada has ever invoked paragraph 14. Although discussions on a possible arrangement with the IAEA were initiated in the late 1980s, Canada decided not to pursue the project, and no arrangement was ever concluded.

Brazil is the only non-nuclear-weapon state with an active nuclear naval propulsion program. Some experts worry that the AUKUS deal, in which Australia plans to buy nuclear-powered submarines from the UK and the United States, will encourage other countries to pursue this technology. (Photo by Mauro Pimentel / AFP via Getty Images)Brazil, the only non-nuclear-weapon state with an active nuclear naval propulsion program, has already announced its plan to build a land-based prototype for a submarine reactor and has provided facility design information to the IAEA. Unlike the AUKUS project, the Brazilian project is based on a domestic fuel cycle, including conversion, enrichment of nuclear material using low-enriched uranium, fabrication of the fuel, and assembly of the fuel into a reactor core. The IAEA has not received a request from Brazil to conclude a relevant arrangement.

In 2012, Iran also announced its intention to produce nuclear-powered submarines3 and recently started producing uranium enriched to 60 percent in uranium-235.4 Although Iran also recently alluded to the possible use of 60 percent-enriched uranium for a submarine program, no formal announcements have been made. South Korea has also expressed interest in acquiring nuclear-powered submarines and commissioned a feasibility study in 2017.5 Neither Iran nor South Korea has raised with the IAEA the prospect of concluding a paragraph 14 arrangement.

In the 1960s and 1970s, Germany and Japan developed nuclear naval propulsion but for civilian application to surface ships.6 No special safeguards arrangements were concluded with either country.

Questions to Answer

The key process-related question is whether approval by the IAEA Board of Governors is required for a paragraph 14 arrangement. Paragraph 14 speaks only of the IAEA agreeing on the arrangement, not whether such an arrangement would be subject to board approval.

An exchange of letters between the IAEA and Australia in 1978 concerning the secretariat’s handling of paragraph 14 requests is likewise ambiguous, but suggests that the board would determine the appropriate action.7 Should the matter be presented to the board by the director-general, the board itself could decide whether approval is necessary.8 Given the divergent interests of the states represented on the board, the results of such deliberations are far from predetermined and could become quite political and contentious.

There are also substantive issues associated with the use of nuclear material in a nonproscribed military activity and the conclusion of an arrangement under paragraph 14.

Details of the arrangement. Paragraph 14 requires that “only while the nuclear material is in such an activity, the safeguards provided for in the Agreement will not be applied” and that safeguards are to apply again “as soon as the nuclear material is reintroduced into a peaceful nuclear activity.” The arrangement must identify, to the extent possible, the period or circumstances during which safeguards will not be applied. It also must provide for the IAEA to be kept informed of the total quantity and composition of the material withdrawn from safeguards, whether in the country or exported.9

In addition, the arrangement may only relate to “such matters as” temporal and procedural requirements and reporting arrangements. The list in INFCIRC/153 of what the arrangement may include is not exhaustive and the IAEA and the states involved must work out the details. In agreeing to such an arrangement, however, the IAEA has no authority to require information deemed by the state to be classified or to approve or disapprove the activity in question.10

Withdrawal of material from safeguards and safeguards reapplication. The material should spend as little time as possible outside safeguards. As agreed by the drafters of INFCIRC/153, “[S]uch peaceful nuclear activities as transport and storage, and activities or processes which merely change the chemical or isotopic composition of nuclear material, such as enrichment or reprocessing, are not intrinsically military and, therefore, [are] not entitled to exclusion from safeguards under paragraph 14.”11 Insofar as Australia could be receiving submarines already equipped with the reactors,12 there inevitably will be some transport and storage of the submarines, which will have to be addressed. As Australia will not be engaged in enrichment or reprocessing of the reactor fuel, that could simplify the negotiation process. Yet, there needs to be clarity regarding when the nuclear material in the reactor would have to be brought back under safeguards.

This Australian Collins Class submarine is among those to be replaced under a new defense agreement among Australia, the United Kingdom and the United States. (Photo by POIS Yuri Ramsey/Australian Defence Force via Getty Images)Implications for additional protocols. The implementation of an additional protocol could mitigate the possible negative impacts on safeguards of a paragraph 14 arrangement. An additional protocol offers the IAEA expanded access to information and locations, which increases the agency’s ability to detect indications of undeclared nuclear material and activities. Many measures contained in an additional protocol, such as the IAEA’s right to request access to and information about nuclear fuel-cycle-related research and development activities not involving nuclear material, could be relevant to a nuclear naval propulsion program, depending on what activities are actually carried out by the non-nuclear-weapon state.

Australia concluded an additional protocol in December 1997.13 Would some of these provisions be suspended as well? If so, how would that impact the IAEA’s ability to determine whether a state is pursuing undeclared nuclear activities?

More complicated still is how such a suspension might impact the IAEA’s drawing of “safeguards conclusions,” in particular the drawing of the “broader conclusion.” Each year, the IAEA draws a safeguards conclusion for each state in which safeguards were implemented during the previous year. If a state with a comprehensive safeguards agreement has an additional protocol in force and the IAEA sees no indications of the diversion of declared nuclear material and no indications of undeclared nuclear material or activities in the state, the agency draws a “broader conclusion” that “all nuclear material remained in peaceful activities.” Would the IAEA still be able to draw such a conclusion? Would there have to be a reformulation of the broader conclusion?

Although it is tempting to suggest that no paragraph 14 arrangement should be approved unless a state has an additional protocol in force, such a position is likely to meet resistance among NPT states-parties, given the insistence by many states on the voluntary nature of such protocols.

Military-to-military transfers and paragraph 14 arrangements. The question has been posed whether a transfer of nuclear material from a military program in a nuclear-weapon state to a military program in a non-nuclear weapon state—a so-called military-to-military transfer—would fall outside the requirements of the NPT or INFCIRC/153. The answer is no. Any efforts to circumvent the paragraph 14 mechanism should be resoundingly rejected, not just from a policy perspective but from a legal perspective. This approach was actually raised in the context of the Canadian project and rejected by the secretariat.

Paragraph 1 of INFCIRC/153 tracks the language of the NPT in requiring the application of safeguards “on all source or special fissionable material in all peaceful nuclear activities.” The reference to peaceful nuclear activities was intended to accommodate the interest in nuclear-powered submarines among some non-nuclear-weapon states in the late 1960s. It was not intended as a means for securing an exclusion of nuclear material from safeguards due its use in a military activity.

Paragraph 34(c) requires that nuclear material of a composition and purity suitable for fuel fabrication or isotopic enrichment or produced later in the nuclear fuel cycle, as the nuclear material in a reactor core would be, become subject to all safeguards procedures upon its import into a state with a comprehensive safeguards agreement. Unlike other provisions in paragraph 34, subparagraph (c) is not limited to the import of such material for particular purposes. Thus, the nuclear material contained in a reactor would become subject to safeguards upon its import, regardless of the purpose for which it is imported.

Finally, pursuant to paragraphs 92 to 96, a state must provide advance notification to the IAEA of the expected transfer into the state of nuclear material in an amount greater than one effective kilogram, as would be the nuclear material in a reactor core, and in any case not later than the date on which the recipient state assumes responsibility for the material. Likewise, the state would be obliged to report the export of such material. None of these provisions has an exclusion for nuclear material used in or transferred for a military purpose.

The general rule under customary international law for interpreting a treaty is that a treaty should be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of an agreement in their context and in light of its object and purpose. From a plain reading of INFCIRC/153, safeguards agreements must be interpreted as states having committed themselves to notifying the IAEA of the production and import of nuclear material even if the material is intended for use in a nonproscribed military nuclear activity. Such states also are committed to abiding by the provisions of paragraph 14 should they wish to use nuclear material in a nonproscribed military nuclear activity.

Under customary international law, supplementary means of interpretation, such as the negotiating history of a treaty, may be used to confirm the ordinary meaning resulting from the application of this general rule. The negotiating history also can be used to determine the meaning of a treaty when interpretation in accordance with the general rule results in an ambiguous or obscure interpretation or leads to a result that is manifestly absurd or unreasonable. In this instance, the interpretation resulting from the application of the general rule does not result in an ambiguous, obscure, or manifestly absurd or unreasonable interpretation of INFCIRC/153. Nevertheless, it is useful to review the negotiating history, which clearly confirms that interpretation.

The negotiators agreed that the IAEA “should be consulted and satisfactory administrative arrangement[s] reached concerning the use of any nuclear material for a military purpose permitted under [the NPT], whether or not the material was initially under safeguards.”14 They also noted that “[t]he provision should thus be applied to all material which was either actually under safeguards and to be withdrawn or which had never been placed under safeguards and which was intended to be used in a permitted nuclear activity.”15 Finally, the negotiators made a change to the secretariat’s proposed draft of paragraph 14 to avoid any ambiguity that might suggest that nuclear material would not be subject to safeguards if it were assigned at the moment of production to a non-explosive military activity.16

Thus, to interpret paragraph 1 as providing what would be tantamount to an automatic exclusion from safeguards of nuclear material because it is used in or was produced for use in a military activity would be manifestly absurd and unreasonable. It would create an enormous loophole in safeguards, thereby defeating the very object and purpose of comprehensive safeguards agreements, contrary to international treaty law.17

Although some will argue that Australia’s sterling nonproliferation credentials should allow for greater flexibility, any arrangement will inevitably be invoked as a precedent by other states.

In fact, as Grossi told reporters, it “cannot be excluded” that other countries would use the AUKUS precedent to pursue their own nuclear submarine plans. General Sir Nicholas Carter, the departing UK chief of the defense staff, recently suggested that Japan, Canada, and New Zealand could eventually join the AUKUS partnership.

To that end, whatever the arrangement, it must be designed as fit for purpose, which is to say, it cannot be used to defeat safeguards regardless of who the partner states might be. Ultimately, the acceptability of any given arrangement should be judged on its nonproliferation merits and be able to survive the following test: if the names of the parties involved are changed, is it still acceptable?



1. International Atomic Energy Agency (IAEA), “The Text of the Agreement Between Australia and the Agency for the Application of Safeguards in Connection With the Treaty on the Non-proliferation of Nuclear Weapons,” INFCIRC/217, December 13, 1974.

2. IAEA, “The Structure and Content of Agreements Between the Agency and States Required in Connection With the Treaty on the Non-proliferation of Nuclear Weapons,” INFCIRC/153 (Corr.), June 1972 (hereafter INFCIRC/153). The paragraph numbers in INFCIRC/153 correspond, by and large, to article numbers in the actual agreements. To avoid confusion, this article will refer to paragraphs as reflected in INFCIRC/153.

3. Olli Heinonen, “Nuclear Submarine Program Surfaces in Iran,” Harvard Kennedy School Belfer Center for Science and International Affairs, July 23, 2021, https://www.belfercenter.org/publication/nuclear-submarine-program-surfaces-iran; Tom O’Connor, “Iran Says It Wants Nuclear Submarines to Power Up Fleet After Confrontation With U.S. Navy,” Newsweek, April 17, 2020.

4. Uranium enriched to a level of 20 percent or higher uranium-235 is considered to be highly enriched uranium. Uranium that is enriched below 20 percent U-235 is referenced as low-enriched uranium.

5. Nuclear Threat Initiative, “South Korea Submarine Capabilities,” February 17, 2021, https://www.nti.org/analysis/articles/south-korea-submarine-capabilities/.

6. In the case of Japan, its surface vehicle Mutsu (1970–1992) never carried commercial cargo and was converted to diesel engine power in 1996. Germany’s Otto Hahn (1968–1979) was converted to diesel engine power in 1979.

7. IAEA, “Exchange of Letters Between the Resident Representative of Australia and the Director General,” GOV/INF/347, November 27, 1987. See Laura Rockwood, “Naval Nuclear Propulsion and IAEA Safeguards,” Federation of American Scientists Issue Brief, August 2017, https://uploads.fas.org/media/Naval-Nuclear-Propulsion-and-IAEA-Safeguards.pdf.

8. See Rockwood, “Naval Nuclear Propulsion and IAEA Safeguards,” p. 11.

9. INFCIRC/153, para. 14(b).

10. INFCIRC/153, para. 14(c).

11. GOV/COM.22/53/Mod.1.; GOV/COM.22/OR.76, paras. 47–53.

12. It is not certain whether construction of the actual submarines will be in Australia, the United Kingdom or the United States.

13. IAEA, “Protocol Additional to the Agreement Between Australia and the International Atomic Energy Agency for the Application of Safeguards in Connection With the Treaty on the Non-proliferation of Nuclear Weapons,” INFCIRC/217/Add.1, February 9, 1998.

14. GOV/COM.22/OR.11, para. 40.

15. GOV/COM.22/OR.13, para. 11.

16. GOV/COM.22/53/Mod.1; GOV/COM.22/OR.76, paras. 47–53.

17. In 1993 the IAEA advised North Korea that there was no automatic exclusion for IAEA access to information or locations simply by virtue of such information or locations being associated with military activities.

Laura Rockwood is the director of Open Nuclear Network in Vienna. She retired from the International Atomic Energy Agency (IAEA) in November 2013 as the section head for nonproliferation and policy in the Office of Legal Affairs after 28 years of service.

For Australia to buy nuclear-powered submarines from the UK and the U.S. as planned, it will have to negotiate an unprecedented safeguards arrangement with the International Atomic Energy Agency.

The Australia-UK-U.S. Submarine Deal: Not Necessarily a Sure or a Good Thing

November 2021
By Trevor Findlay

In June 1987, Canada announced that it intended to build 10 to 12 nuclear-powered submarines, based on a French or UK design and fueled with highly enriched uranium (HEU) possibly of Canadian origin. Faced with insurmountable strategic, political, financial, logistical, and nonproliferation obstacles, the idea sank without trace within two years.1 Although the Australian nuclear-powered submarine proposal, announced 34 years later on September 16, is different in several respects, it faces equally strong headwinds that may deliver the same result.

A Royal Australian Navy diesel and electric-powered Collins Class submarine sits in Sydney Harbour in 2016. That naval weapon is to be replaced by nuclear-powered submarines that the United Kingdom and the United States recently agreed to provide Australia as part of the new AUKUS defense cooperation announcement. (Photo by Peter Parks/AFP via Getty Images)Much about the Australian project is speculative. Australian Prime Minister Scott Morrison, U.S. President Joe Biden, and UK Prime Minister Boris Johnson simply released a one-page statement launching “an enhanced trilateral security partnership” called AUKUS aimed at fostering “deeper integration of…security and defense-related science, technology, industrial bases, and supply chains.”2 The headline-grabbing item was the announcement of a trilateral effort to support Australia in acquiring nuclear-powered submarines, beginning with an 18-month study to seek “an optimal pathway to deliver this capability.” No numbers were announced, no likely design was suggested, and no nuclear fuel type or acquisition plan was outlined. Although all three partners committed themselves to “the highest standards for safeguards, transparency, verification, and accountancy measures to ensure the non-proliferation, safety and security of nuclear material and technology,” the length of this list alone suggests that complex and profound questions arise not just for the three governments, but for the international community, particularly the global regime governing the use of nuclear energy.

Knowns and Unknowns

At this stage, the unknowns of the project are Rumsfeld-esque in their tortuousness and interrelatedness. Yet, there are some knowns or likely knowns to guide preliminary analysis.

First, for parochial political reasons, the submarines must be built in Australia for the most part, specifically in Adelaide, in the state of South Australia. Australia’s conventionally powered submarines have been built there for decades, resulting in a skilled, specialized workforce. One of the smallest and economically challenged of Australia’s states but electorally important, South Australia has relied on government-funded projects to boost employment and capacity in its industrial sector. The joint project with France to produce conventionally powered submarines that was unceremoniously cancelled seemingly minutes before the AUKUS announcement, required 50 percent Australian “content,” down from the originally expected 75 percent. Australia’s purchase of U.S. or UK submarines off the shelf, as some have suggested, would seem politically untenable.

A second known factor is that Australia cannot produce enriched uranium itself whether low-enriched or highly enriched, for submarine propulsion or any other purpose, despite having among the largest deposits of uranium in the world. It does not have the industrial, technical, or financial capacity or political license to build and operate a standard gas-centrifuge plant. Australia sold off its domestically invented SILEX laser-enrichment technology to the United States two decades ago.3 In any case, Australian federal law prohibits uranium enrichment in the country. Enriched uranium for submarines would need to be imported.

A third certainty is that Australia does not have any current or likely future capacity to build a nuclear reactor, especially for submarine propulsion. Unlike Canada, which developed and operates CANDU reactors, Australia has no experience with nuclear reactors beyond research units based at Lucas Heights in Sydney. The latest model, devoted largely to producing medical radioisotopes, was imported from Argentina. Therefore, Australia would need to buy the reactor and its fuel from the United Kingdom or the United States. If HEU is chosen, the reactors will contain “lifetime cores,” which will operate for around 30 years and require no refueling, a much prized characteristic of HEU-powered vessels. The sealed reactors would presumably be transported by ship to Adelaide to be encased in the submarine hulls and returned to the provider at the end of the submarines’ lifetime for dismantlement and disposition of the spent fuel.

A final important area of clarity is that Australia is seeking to arm its submarines with conventional weapons, presumably sea-launched cruise missiles, not nuclear weapons. For decades, Australia has been a dedicated supporter of the nuclear nonproliferation regime and more recently of the global nuclear security architecture. After initial reservations, Australia signed the nuclear Nonproliferation Treaty (NPT) in February 1970, just before it entered into force, and ratified it in 1973. It has subsequently become one of the strongest champions of the treaty and of the International Atomic Energy Agency (IAEA) and its safeguards system.

Australia not only has a comprehensive safeguards agreement as required by the NPT, but also imposes bilateral safeguards on Australian-origin uranium exports. It was the first country to sign an additional protocol to its safeguards agreement and was the first to receive the so-called broader conclusion, indicating that it has accounted for all nuclear material subject to safeguards in its territory. Australia was instrumental in negotiations on the 1985 Treaty of Rarotonga, which created a nuclear-weapon-free zone in the South Pacific. It is also an active member of the Nuclear Suppliers Group and other export control arrangements.

In the nuclear security realm, Australia’s track record is also impressive. It is party to the Convention on the Physical Protection of Nuclear Material and its 2005 amendment, along with all other nuclear governance conventions. It has consistently been rated number one by the Nuclear Threat Initiative in the annual Nuclear Security Index and has enthusiastically contributed to continuing efforts to strengthen nuclear security resulting from the four nuclear security summits between 2010 and 2016.

One might imagine, then, that if any country were to become the first non-nuclear-weapon state to acquire nuclear-powered submarines, Australia’s would be the safest pair of hands. Indeed, some have argued that Australia could use its submarine acquisition plan to strengthen global nuclear governance. Better Canberra than Brasilia or, at worst, Tehran. Even so, the implications for the nonproliferation regime are far-reaching, overlapping, and complex.

Disturbing the Nonproliferation Zeitgeist

The NPT and the collection of other treaties, arrangements, and organizations that compose the nonproliferation regime do not exist in a vacuum, but are profoundly affected by states’ attitudes, perceptions, and actions. As a nonproliferation “white knight,” Australia’s announcement that it is considering acquiring nuclear-powered submarines in partnership with two nuclear-weapon states portends a further roiling of the political atmosphere around a regime that is already being buffeted by numerous gales. The worst of those include the ongoing noncompliance cases of Iran and North Korea; the absence of India, Israel, and Pakistan from the NPT; the continuing nonfulfillment of undertakings by the nuclear-weapon states-parties to the NPT to achieve nuclear disarmament; the modernization and expansion programs of almost all of the states with nuclear weapons; the decades-long lack of progress at the Conference on Disarmament, especially in negotiating a fissile material cutoff treaty; and the non-entry into force of the Comprehensive Test Ban Treaty. The AUKUS submarine proposal will undoubtedly be added to this litany of woes at the 10th NPT review conference, originally scheduled for 2020 but now deferred to 2022 due to the COVID-19 pandemic.

It is not that anyone suspects Australia of seeking nuclear weapons through the backdoor of nuclear submarine propulsion, but rather that the idea reeks of the hypocrisy that has always plagued a regime built on the premise of a more or less eternal divide between nuclear haves and have-nots. Unlike the IAEA Statute, which envisaged no military use of nuclear material, the NPT carved out an exception for non-explosive military use, apparently at the suggestion of Italy, with U.S. and Soviet acquiescence.

The United States nonetheless has consistently refused to provide nuclear-propulsion technology to non-nuclear-weapon states, including to allies such as Canada, South Korea, and reportedly Japan, due to proliferation concerns. It has now made an exception for Australia as an exclusive member of the “Anglosphere,” whatever that means for three increasingly multicultural societies. Australia itself carved out an exception to its policy of not supplying uranium to non-NPT parties by doing a deal with India, a state with nuclear weapons that from the outset sought to undermine the treaty. The constant chipping away at the fundamentals of the nonproliferation regime, especially by erstwhile champions, can only increase cynicism and undermine confidence in its longevity.

Setting Unsettling Precedents

If the AUKUS project is realized and assuming that Brazil, which is building its own nuclear-powered submarines, does not get there first, Australia will become the first non-nuclear-weapon state to acquire a nuclear-powered submarine. The precedent will be set, paving the way for other states to demand similar capability, either as a legitimate defense asset or as cover for more alarming nuclear ambitions, such as nuclear weapons development. Unlike Australia, some of the states that have expressed interest in nuclear-powered submarines, including Brazil and South Korea, also wish to enrich their own fuel. Exhibit A on this list is Iran, which has long argued implausibly that it needs to enrich its own uranium for peaceful purposes, notably its Tehran Research Reactor and Bushehr nuclear power plant, currently supplied by Russia, but has now added nuclear-powered submarines to its list.

Australia would set another precedent by becoming the first state to take advantage of the “loophole” in comprehensive safeguards agreements that permits nuclear material for a non-explosive military purpose to be removed from safeguards for the duration of that use. If Australia chooses the military-to-military option whereby the reactor and its HEU fuel are supplied by the U.S. or UK navies and returned to their control when the submarine is decommissioned, it might be assumed that there will be no requirement for removal or reapplication of safeguards because the material will originate from, remain in, and return to military use. Yet, allowing a non-nuclear-weapon state to import HEU outside of safeguards in this manner would make a mockery of the entire nonproliferation regime.

Fortunately, Australia’s safeguards agreement, like all others, requires that it notify the IAEA of its intention to acquire nuclear material for a non-explosive military purpose and help devise suitable verification arrangements with the IAEA to ensure that the material is not diverted to nuclear weapons. In working with the IAEA on this challenging task, Australia would be setting a precedent, for good or ill, that other states will be able to exploit. The sensitivity of the technology and the inaccessibility of the reactor to inspectors preclude a traditional approach. Instead, new approaches and methods will have to be devised to satisfy the IAEA that no diversion of nuclear material to weapons purposes takes place, while protecting confidential, proliferation-sensitive information .

Australia has already notified IAEA Director-General Rafael Grossi of its intentions and signaled its willingness to work with the agency, presumably along with the United States and the UK, to craft suitable arrangements. Grossi has responded publicly by noting that verification will be “very tricky.”4 For Australia itself, the situation may become even trickier. Under the strengthened safeguards system that Australia has long championed, the IAEA accords a state the broader conclusion when it is able to certify that, based on the information available to it, it has accounted for all nuclear material within the state. Just how this conclusion could be reached after Australia’s nuclear-powered submarines have begun operating, especially at sea, is unknown. Australia has insisted that the IAEA should not automatically reissue the broader conclusion for states without reassessing their current circumstances, as occurred for Libya when civil war prevented the agency from ensuring the continuity of safeguards in its territory. Australian officials will undoubtedly work in good faith with the IAEA to craft an effective arrangement to ensure verifiability to the extent possible, but there is an element of moral hazard for Australia. It may succeed in making the world “safe” for the proliferation of nuclear-powered submarines in the hands of non-nuclear-weapon states.

Australian Prime Minister Malcolm Turnbull (fourth from left), French President Emmanuel Macron (second left) and other officials visit the Australian submarine HMAS Waller in Sydney in May 2018 when France was still planning to sell submarines to Australia. That deal has now been upended by the AUKUS arrangement. (Photo by Brendan Esposito - Pool/Getty Images)A final precedent relates to nuclear security. The Australian project would see the acquisition of HEU by a non-nuclear-weapon state at a time when the United States and others, including Australia, are attempting to minimize global holdings of HEU, including by converting reactors to using low-enriched uranium (LEU) and repatriating HEU to the United States or Russia for disposition. Although the nuclear material in submarine reactors is relatively secure, albeit nonstationary, the use of HEU for naval propulsion by a country that has been HEU free goes against the grain of the impressive efforts in recent years to ensure that nuclear material does not fall into the hands of terrorists or other nonstate actors. Some observers have suggested that Australia use LEU for its submarines, perhaps in collaboration with France, which uses such fuel. This may assuage French fury at the cancellation of its contract to build Australia’s conventional submarines, whose design paradoxically was to be based on French nuclear-powered submarines at Canberra’s insistence. IAEA verification, however, would become more challenging because LEU-fueled submarines, at least those using existing technology, require periodic refueling.

Going Quietly Into the Deep?

Despite an opinion poll indicating immediate domestic support for the AUKUS announcement, there remains significant public skepticism in Australia about the use of nuclear energy for any purpose. It remains to be seen whether this will shift as the 18-month study proceeds, details emerge, and the political, diplomatic, military, economic, nonproliferation, security, and opportunity costs become clearer. Although the opposition Labor Party has felt it politically expedient to support the AUKUS announcement, this is conditional on nonproliferation concerns being assuaged. A general election is due within a year. The Australian nuclear-powered submarines could be destined to go the way of Canada’s. In the meantime, the AUKUS partners need to explain how they propose to deliver the gold standard safeguards, transparency, verification, and accountancy measures they have promised.



1. See Tariq Rauf and Marie-France Desjardins, “Opening Pandora’s Box? Nuclear Powered Submarines and the Spread of Nuclear Weapons,” Aurora Papers, no. 8 (1988).

2. “Joint Leaders Statement on AUKUS,” Prime Minister of Australia, September 10, 2021, https://www.pm.gov.au/media/joint-leaders-statement-aukus.

3. “Message to the Congress Transmitting the Australia-United States Peaceful Nuclear Technology Transfer Agreement,” Public Papers of the Presidents of the United States, William J. Clinton, 1999, Vol. 2 (Washington: U.S. Office of the Federal Register, 1999), pp. 1963–1965.

4. See John Carlson, “IAEA Safeguards, the Naval ‘Loophole’ and the AUKUS Proposal,” Vienna Center for Disarmament and Non-Proliferation, October 8, 2021, https://vcdnp.org/wp-content/uploads/2021/10/Safeguards-and-naval-fuel-JC-211008.pdf; Laura Rockwood, “Naval Nuclear Propulsion and IAEA Safeguards,” Federation of American Scientists Issue Brief, August 2017, https://uploads.fas.org/media/Naval-Nuclear-Propulsion-and-IAEA-Safeguards.pdf. Francois Murphy, “AUKUS Submarine Deal ‘Very Tricky’ for Nuclear Inspectors—IAEA Chief,” Reuters, September 28, 2021.

Trevor Findlay is a principal fellow in the School of Social and Political Sciences at the University of Melbourne. His next book, Transforming IAEA Safeguards Culture: The IAEA, Iraq, and the Future of Nonproliferation, will be published in early 2022.

From an Australian perspective, there are lots of questions to be answered about the Australia-UK-U.S. submarine deal.

The Australia-UK-U.S. Submarine Deal: Mitigating Proliferation Concerns

November 2021
By Frank N. von Hippel

On September 15, U.S. President Joe Biden joined UK Prime Minister Boris Johnson and Australian Prime Minister Scott Morrison to announce an Australian-UK-U.S. security pact (AUKUS) under which the United States and the United Kingdom will assist Australia in building at least eight nuclear-powered attack submarines. The purpose is to strengthen the alliance trying to contain a growing Chinese navy. The first submarine is
not expected to be operational before 2040.

Brazil, a non-nuclear-weapon state with a program to develop nuclear-powered attack submarines, plans to power its first submarine with LEU fuel but has not forgone the right to use HEU. Photo from 2019 shows ceremony in Rio de Janeiro celebrating Brazil's French-designed, Brazilian-built Humaita submarine, which runs on diesel-electric propulsion. (Photo by Mauro Pimentel/AFP via Getty Images)The AUKUS countries said that it would take 18 months to work out the specifics of the deal, but obvious candidates for the submarine designs to be provided to Australia are the U.S. Virginia-class attack submarine and the UK Astute-class submarine, in production since 1999 and 2001, respectively.

Both submarine classes are fueled with U.S. weapons-grade uranium enriched to more than 90 percent uranium-235 that was declared excess to weapons needs following the drastic downsizing of the U.S. Cold War nuclear warhead stockpile. Both submarine types have life-of-ship cores, which means they should not have to be refueled during their design lives of approximately three decades.

The deal replaces one that Australia reached with France in 2016 under which Australia would have received 12 French Suffren-class submarines equipped with conventional propulsion rather than the nuclear propulsion used by France. In 2016, Australia did not wish to develop the infrastructure required to supply fuel for a nuclear-powered ship.1 France refuels its nuclear submarines every 10 years.

Life-of-ship cores could enable Australia to avoid having to produce its own nuclear fuel, refuel its submarine reactors, and dispose of the spent fuel. The United States or UK could simply sell Australia the reactor cores and then take them back for disposal when the submarines are decommissioned.

A Troublesome Precedent

The proposed AUKUS submarine plan, however, would set an important precedent of a nuclear-weapon state selling nuclear submarines to a non-weapon state. The use of highly enriched uranium (HEU) fuel makes the AUKUS precedent especially troublesome from a nonproliferation perspective.

HEU can be used directly by nations to make nuclear weapons. It also could be used by terrorists to make a simple gun-type nuclear weapon like the Hiroshima bomb.

Because HEU is so easily weaponized, the United States has spent $2 billion since the September 11 terrorist attacks to eliminate it as a research reactor fuel and replace it with low-enriched uranium (LEU) fuel, containing less than 20 percent U-235, which cannot be used to make a nuclear explosive.2 As part of this effort, the United States has converted most of its own research reactors to LEU use and has cleared HEU from 33 of 55 countries down to a level of less than one kilogram, a small fraction of the amount required to make a nuclear weapon.3

At the same time, U.S. and UK naval reactors are the world’s largest consumers of HEU. Annually, about three tons of weapons-grade uranium, enough for more than 100 nuclear weapons, are being fed into their naval reactors. In contrast, Chinese and French submarines are fueled with LEU, while India and Russia are believed to use HEU enriched to 21–45 percent U-235.4 HEU in this enrichment range is considered weapons usable, but has a critical mass much larger than weapons-grade uranium.

The United States and UK should be designing their future naval reactors to use LEU fuel. They certainly should not be setting the precedent of spreading HEU-fueled naval reactors to non-nuclear-weapon states such as Australia, especially when Iran and a few other non-nuclear-weapon countries are considering fielding their own nuclear-fueled submarines. Whether to fuel research reactors or naval reactors, expanding the use of HEU increases the risk of this material being diverted to nuclear weapons use.

Nuclear Submarines and Non-Nuclear-Weapon States

Nuclear-powered attack submarines have been of interest to non-nuclear-weapon states for some time. In the late 1980s, Canada explored buying some from France or the UK to reinforce its sovereignty in its northern waters, but with the end of the Cold War, abandoned the project as too costly.5

Brazil has a program to develop nuclear-powered attack submarines that dates to the 1970s.6 It is learning from France how to build conventional submarines and is assembling a land-based prototype reactor inside a mockup of a hull section of a future nuclear submarine. The Brazilian navy developed and controls Brazil’s uranium-enrichment plants. This was a major proliferation concern for the United States when Brazil was ruled by a military dictatorship in 1964–1985 and before it entered a nuclear transparency agreement with Argentina in 1991 and joined the nuclear Nonproliferation Treaty (NPT) in 1998. Brazil plans to fuel its first submarine with LEU, but has not forgone the right to use HEU fuel if that proves advantageous.

In South Korea, President Moon Jae-in and his administration have expressed a sustained interest in developing nuclear-powered submarines.7 The United States has refused to change the two countries’ nuclear cooperation agreement to allow South Korea to enrich uranium. Therefore, South Korea may look to Russia, which has offered Seoul an icebreaker propulsion reactor design that can be fueled with 19.75 percent-enriched LEU.8 Russia’s existing nuclear agreement with South Korea covers only “peaceful uses of atomic energy.”9 If the United States can change its agreement with Australia,10 however, Russia can change its agreement with South Korea.

In the past, Japan has not expressed an interest in nuclear submarines. After the AUKUS deal was announced, however, two of the four candidates for prime minister declared their interest,11 although not Fumio Kishida, who won the Liberal Democratic Party’s support and was sworn in as prime minister in October.

There is also the case of Iran. In 2013, during the hard-line administration of President Mahmoud Ahmadinejad, the head of the Atomic Energy Organization of Iran suggested Tehran might require uranium enriched to 45–56 percent U-235 for a nuclear submarine program.12 In April, as U.S.-Iranian negotiations stalled on reviving the Joint Comprehensive Plan of Action, Iran began producing 60 percent-enriched HEU.13

After the AUKUS deal was announced, two journalists from The New York Times interviewed aides accompanying Iran’s new hard-line foreign minister to the United Nations and reported that the aides noted that HEU “could be used in naval reactors, suggesting they might want to use it for that purpose. And they cited Mr. Biden’s new deal with Australia, which calls for [the United States and the UK] to supply Australia with the technology for nuclear-propelled submarines,” which use HEU.14

HEU and Naval Fuel

The United States and UK are creating a dangerous precedent by proposing to export HEU naval fuel to a non-nuclear-weapon state. Other countries are likely to see the deal as creating a more permissive environment to acquire their own HEU-fueled nuclear submarines and, in the absence of a willing supplier, make HEU fuel themselves as Iran threatens to do. The world does not need HEU in more places and more being produced in more countries.

IrFour torpedo tubes in the bow of a Suffren-class nuclear attack submarine, under construction in north-western France in 2017, during a visit by French Defence Minister Florence Parly and Australian Prime Minister Malcolm Turnbull. Australia would have bought 12 Suffrens equipped with conventional propulsion from France under a deal Australia abrogated in favor of buying nuclear submarines from the United Kingdom and the United States. (Photo by Charly Triballeau/AFP via Getty Imges)onically, the nuclear version of France’s Suffren-class attack submarine, which Australian leadership insisted in 2016 should be converted to diesel-electric power, is fueled with LEU containing an average of only 6 percent U-235.15

To make LEU weapons usable, a country would have to run it through an enrichment plant to produce HEU. In a non-nuclear-weapon state, especially one that has an additional protocol to its safeguard agreement, the International Atomic Energy Agency would have a good chance of detecting such an activity. Therefore, if non-nuclear-weapon states feel they need nuclear submarines and to have their own enrichment plants to fuel them, the fuel should be LEU.


Congressional Interest in LEU Fuel

Since 1994, reducing the risk of proliferation of naval HEU fuel has been the primary driver behind efforts by some members of Congress to require the National Nuclear Security Administration (NNSA) Office of Naval Reactors to develop LEU fuel for future U.S. submarine and aircraft carrier propulsion reactors.

As the office has made clear, however, its priority has been to achieve life-of-ship cores. In fact, it believes it has done so for the Virginia-class attack submarine, which began production in 1999, and for the Columbia-class ballistic missile submarine, the first of which began construction in 2020. The design lives of these submarines are 33 and 42 years, respectively. This means that, after the older classes of U.S. submarines have had their midlife refueling, there will be no need for routine refueling of submarines. Refueling equipment and capabilities will be retained only on a standby basis for core repair or replacement following potential fuel-element failure.

The Office of Naval Reactors’ first report in response to Congressional interest in LEU fuel was in 1995.16 It stated that because the U-235 chain reaction provides almost all of the fission energy from the fuel, if the U-235 were diluted to just below 20 percent U-235, which is the top of the LEU enrichment range, it would be necessary to increase the volume of the core threefold to achieve the same core life. This would require a larger, heavier pressure vessel and a bigger hull.

The Virginia-class attack submarine Minnesota (SSN-783), shown under construction in 2012, is among the class of submarine that could be sold to Australia. (U.S. Navy Photo)For Virginia-class submarines, the Office of Naval Reactors found that a life-of-ship core would require the diameter of the submarine to be increased from 10 to 11 meters. The office did not expect a significant impact on the sizes of the larger ballistic missile submarine and aircraft carrier. If, as reported,17 the SSN(X) next-generation U.S. attack submarines are to have hull diameters significantly larger than the Virginia-class, they too could accommodate larger LEU cores.

In 2012, Congress asked for an update and this time, the response was more encouraging. The Office of Naval Reactors reported it was developing a new higher uranium-density fuel that might not require as large a core volume increase for an LEU life-of-ship core.18 Yet, it was testing the new fuel design with weapons-grade uranium to pack more U-235 into the core and to increase U.S. submarines’ lifetime energy budgets for higher-speed transits across the Pacific Ocean and other uses. The energy requirement for potential LEU cores was therefore a moving target.

Congress asked for a research and development plan for developing and testing the new fuel design with LEU.

The Office of Naval Reactors delivered the outline of a plan in July 2016. The report emphasized that the R&D would cost about $1 billion and take “at least 15 years” and that “success is not assured.” It also said that providing LEU cores for aircraft carriers would cost an additional “several billion dollars,” including the cost of a land-based prototype aircraft carrier propulsion reactor.19 This would be comparable to the cost of an additional nuclear-powered submarine.

The Office of Naval Reactors also asked JASON, an elite technical group of mostly academic consultants for the Department of Defense, the NNSA, and other agencies, to review its proposed program for developing LEU fuel. The JASON report, which was partially declassified three years later was supportive. It emphasized, however, that there is only a limited opportunity to make sure that the follow-on to the Virginia-class submarine, the not-yet-named SSN(X) that is scheduled to be procured starting in the early 2030s,20 will be able to accommodate an LEU core. “If the reactor compartment is not designed to accommodate a life-of-ship LEU core, and if later re-design to accommodate such an LEU core is impractical, then HEU cores will be required for all [SSN(X)s], the last of which will launch in the 2060s. On the other hand, if design parameters and fuel development allow an LEU reactor…then it is possible that the Navy's final HEU core will be built in the 2040s.”21

Unfortunately, the Navy came to oppose even conducting that R&D. The simplest explanation is that the Navy does not view minimizing HEU use as its responsibility. Members of Congress sympathetic to this perspective inserted into the 2016 National Defense Authorization Act (NDAA) the requirement that “the Secretary of Energy and the Secretary of the Navy shall jointly submit to the congressional defense committees the determination of the Secretaries as to whether the United States should continue to pursue research and development of an advanced naval nuclear fuel system based on low-enriched uranium.”

At the beginning of 2018, the Trump administration responded that it saw no benefits to the Navy incurring the cost of shifting to LEU fuel use.22

Despite this opposition, Congress has appropriated funding for Navy LEU fuel development every year since fiscal year 2016, starting at $5 million and rising to $20 million in fiscal year 2021.23 Given the Office of Naval Reactors’ resistance, congressional advocates of LEU fuel for naval reactors shifted the funding for LEU fuel development to the NNSA Office of Defense Nuclear Nonproliferation.

The executive branch, however, has never requested funding for this program. For fiscal year 2022, the House of Representatives voted to appropriate another $20 million, but the Senate Armed Services Committee recommended in the 2022 NDAA a provision that would “prohibit the obligation or expenditure of any fiscal year 2022 funds [by the NNSA] to conduct research and development of an advanced naval nuclear fuel system based on low-enriched uranium unless the Secretary of Defense, the Secretary of Energy, and the Secretary of the Navy communicate certain determinations to the congressional defense committees.”24

What Is Next

Recently, a group of nonproliferation experts, including the author, wrote to the Biden administration stressing the importance of designing future U.S. naval reactors to use LEU fuel.25 The AUKUS deal highlights the fact that the United States and UK are undermining the nuclear nonproliferation and anti-terrorism regimes by fueling their naval reactors with weapons-grade uranium. Now they propose to export these reactors to a non-nuclear-weapon state.

If the United States does not switch to using LEU naval fuel by about 2060, when its excess stock of weapons-grade uranium is projected to run out, it will have to restart production of weapons-grade uranium for the first time since the end of the Cold War.

The United States and UK should instead exploit the opportunity created by the furor over the AUKUS deal to commit to design their future naval propulsion reactors to use LEU fuel. They also should use the planned 18-month period of study and evaluation of the technical and policy details of the proposed AUKUS submarine deal to do their utmost to design any submarines built by or leased to Australia to use LEU-fueled propulsion reactors rather than the more problematic HEU-fueled option. Otherwise, the three AUKUS countries, long-time leaders in efforts to limit the spread of nuclear weapons, may well find themselves on a path that would undermine global nonproliferation norms and long-standing nonproliferation objectives.



1. Malcolm Turnbull, “Address to the National Press Club,” 29 September 2021, https://www.malcolmturnbull.com.au/media/address-to-the-national-press-club-september-2021.

2. U.S. Department of Energy, “Budget & Performance,” https://www.energy.gov/budget-performance (accessed October 22, 2021).

3. National Nuclear Security Administration (NNSA), “NNSA Removes All Highly Enriched Uranium From Nigeria,” December 7, 2018, https://www.energy.gov/nnsa/articles/nnsa-removes-all-highly-enriched-uranium-nigeria.

4. Frank von Hippel, “Banning the Production of Highly Enriched Uranium,” International Panel on Fissile Materials Research Report, no. 15 (March 2016), https://fissilematerials.org/library/rr15.pdf.

5. Adam Lajeunesse, “Sovereignty, Security and the Canadian Nuclear Submarine Program,” Canadian Military Journal, Winter 2007–2008, pp. 74–82.

6. Andrea de Sá, “Brazil’s Nuclear Submarine Program: A Historical Perspective,” Nonproliferation Review, Vol. 22, No. 1 (2015): 3.

7. Jun Ji-hye, “South Korea Moving to Build Nuclear-Powered Submarines,” The Korea Times, September 5, 2017.

8. “Russia May Help South Korea to Build Nuclear Reactor for Maritime Vessels,” Sputnik International, August 7, 2018; Atomenergomash JSC, “Solutions for the Shipbuilding Industry,” n.d., https://aem-group.ru/static/images/buklety/2020/Booklet_sudostroenie_en.pdf.

9. Agreement Between the Government of the Republic of Korea and the Government of the Russian Federation on the Cooperation on the Peaceful Uses of Atomic Energy, May 28, 1999, 2396 U.N.T.S. 43273.

10. Agreement Between the Government of the United States of America and the Government of Australia Concerning Peaceful Uses of Nuclear Energy, May 4, 2010, T.I.A.S. no. 10-1222, https://www.state.gov/wp-content/uploads/2019/02/10-1222-Australia-Atomic-Energy-Peaceful-Uses.pdf.

11. Steven Stashwick, “Japan’s Kono Says He Supports Building Nuclear Submarines,” The Diplomat, September 28, 2021.

12. “Iran May Need Highly Enriched Uranium in Future, Official Says,” Reuters, April 16, 2013.

13. International Atomic Energy Agency Board of Directors, “Verification and Monitoring in the Islamic Republic of Iran in Light of United Nations Security Council Resolution 2231 (2015),” GOV/2021/39, September 7, 2021, p. 6.

14. David Sanger, Michael Crowley, and Rick Gladstone, “Rebuking Biden, Iran’s Chief Diplomat Demands More Sanctions Relief,” The New York Times, September 24, 2021.

15. Sébastien Philippe and Frank von Hippel, “The Feasibility of Ending HEU Fuel Use in the U.S. Navy,” Arms Control Today, November 2016, p. 15.

16. Director, Naval Nuclear Propulsion, U.S. Navy, “Report on Use of Low Enriched Uranium in Naval Nuclear Propulsion,” June 1995, https://fissilematerials.org/library/onnp95.pdf.

17. Sam LaGrone, “BWXT CEO: Navy’s Next-Generation SSN(X) Attack Boat Will Build Off Columbia Class,” USNI News, November 2, 2020, https://news.usni.org/2020/11/02/bwxt-ceo-navys-next-generation-ssnx-attack-boat-will-build-off-columbia-class.

18. Office of Naval Reactors, U.S. Department of Energy, “Report on Low Enriched Uranium for Naval Reactor Cores: Report to Congress,” January 2014.

19. NNSA, “Conceptual Research and Development Plan for Low-Enriched Uranium Naval Fuel: Report to Congress,” July 2016.

20. Ronald O’Rourke, “Navy Next-Generation Attack Submarine (SSN[X]) Program: Background and Issues for Congress,” Congressional Research Service In Focus, IF11826, September 15, 2021.

21. JASON, “Low-Enriched Uranium for Potential Naval Nuclear Propulsion Applications,” JSR-16-Task-013 (November 2016), https://irp.fas.org/agency/dod/jason/leu-naval.pdf (declassified portions).

22. Richard V. Spencer and Rick Perry to Deb Fischer, March 25, 2018, https://fissilematerials.org/library/usn18b.pdf. The same letter went out to the ranking Democratic senator and to the chair and ranking member of the counterpart House of Representatives subcommittee.

23. “Navy LEU Fuel R&D,” Nuclear Proliferation Prevention Project, January 2, 2021, https://sites.utexas.edu/nppp/files/2021/02/Navy-LEU-Fuel-RD-2021-Jan-2.pdf.

24. Senate Committee on Armed Services, National Defense Authorization Act for Fiscal Year 2022, S. Rep. No. 117–39 at 354.

25. Joe Biden from Robert Gallucci et al., “Mitigate the Proliferation Impact of Offering Submarines Fueled With Weapon-Grade-Uranium to a Non-Nuclear-Weapon State by Committing to Design Future US Naval Reactors to Use Low-Enriched-Uranium Fuel,” October 6, 2021, https://sgs.princeton.edu/sites/default/files/2021-10/AUKUS-Letter-2021.pdf.

Frank N. von Hippel is a senior research physicist and professor of public and international affairs emeritus with the Program on Science and Global Security at Princeton University.

The proposed AUKUS submarine plan would set a precedent of a nuclear-weapon state selling nuclear submarines to a non-weapon state. The use of highly enriched uranium (HEU) fuel is especially troublesome.

New Iran Leadership Complicates Negotiations

October 2021
By Sina Azodi

The election of Ebrahim Raisi as Iran’s new president represents a consolidation of power by hard-liners who generally oppose engagement with the West. These forces, who previously worked to undermine President Hassan Rouhani’s engagement agenda, are now in control of all three branches of the Iranian government. Meanwhile, Raisi is grappling with several other major challenges, including a crumbling economy battered by U.S. and international sanctions, high unemployment, and the COVID-19 pandemic, all of which have put the country on its heels.

Ebrahim Raisi speaks during the swearing-in ceremony for the new Iranian President on August 5, 2021 in Tehran, Iran.  (Photo by Meghdad Madadi/ATPImages/Getty Images)Although Raisi has expressed a desire to revive the nuclear deal, formally known as the Joint Comprehensive Plan of Action (JCPOA), and to achieve a lifting of U.S. sanctions, he has repeatedly refused any modifications in Iran’s ballistic missile program and its regional activities, two other areas on which the United States and its partners in the nuclear deal—China, France, Germany, Russia and the United Kingdom—have demanded action. Historical factors and a complicated geostrategic environment are also driving Iranian decision-making, thus making compromise with the West even more unlikely. The United States still has some policy options for dealing with Iran’s regional activities and missile program, but they are likely to fall far short of what was once envisioned.

The JCPOA, signed in 2015, was a diplomatic achievement that ended decades of tensions over Iran’s controversial nuclear program. From the onset, however, critics undermined the deal by claiming it did not cover such critical issues as Iran’s ballistic missile program and regional involvement in places such as Lebanon, Syria, and Yemen. Ultimately, President Donald Trump used these shortcomings as a pretext to withdraw the United States from the agreement, reinstate previously lifted sanctions, and impose even tougher new ones on the Iranian economy, all in an attempt to force Tehran to submit to a “better” agreement. This “maximum pressure” campaign failed miserably as Iran responded first by exercising restraint, then by expanding its nuclear program. Today, Iran is enriching uranium to a level of 60 percent uranium-235 and has much more advanced centrifuge machines compared to where things stood when the JCPOA was being fully implemented by all signatories.

The Consequences of Choices

Critics ignore that exclusion of Tehran’s missile program and regional activities from the nuclear agreement was a deliberate choice. Both sides preferred to focus attention on the more dangerous issue—the nuclear program—and neither was ready to accept a compromise on the ancillary issues. In January 2021, Iranian Foreign Minister Mohammad Javad Zarif contended during the negotiations that “[w]e agreed from the beginning [of nuclear negotiations] that regional and missile issues will not be negotiated in the JCPOA…. This [missile] issue was raised, but we refused to negotiate over it, and we paid a price for not talking [about it]."1

After he took office, Trump cited the agreement’s “near total silence on Iran's missile programs”2 as a pretext for the U.S. withdrawal from the JCPOA. The Biden administration, although committed to reviving the agreement, has expressed its intention to eventually seek follow-up talks with Tehran on the missiles and regional topics. As U.S. Secretary of State Antony Blinken observed last February, the United States, working with its allies and partners, “will also seek to lengthen and strengthen the JCPOA and address other areas of concern, including Iran's destabilizing regional behavior and ballistic missile development and proliferation.”3 Similarly, in March, he told members of Congress that “[w]e have fundamental problems with Iran’s actions across a whole series of things, whether it is support for terrorism, whether it is a ballistic missile program.”4

The View in Washington

Given that ballistic missiles are a primary method of delivering nuclear weapons, Iran’s large and diverse inventory of short- and medium-range missiles, in conjunction with its quest for nuclear capability, has raised many concerns among U.S. officials, intelligence analysts, and think tank experts. Shortly after the nuclear deal was implemented in April 2016, President Barack Obama criticized Iran for undermining the “spirit” of the agreement by testing ballistic missiles.5 Two successive intelligence directors also raised alarms: James Clapper argued in 2016 that Iran’s ballistic missiles are “inherently capable of delivering” weapons of mass destruction,6 and three years later, Daniel Coats warned that Iran’s missile program continues to pose a threat to the countries of the Middle East.7

Such comments reflect a strong consensus in Washington that because Iran’s ballistic missile program jeopardizes the national security interests of the United States and its allies, the United States must somehow contain the program.

Meanwhile, in 2018, Federica Mogherini, the European Union foreign policy chief who played a pivotal role in the JCPOA negotiations, made clear that the EU shares some of the U.S. concerns over Iran’s ballistic missiles.8 Similarly, in June 2021, the foreign ministers of the Gulf Cooperation Council states urged that the revived nuclear negotiations also encompass Iran’s “sponsorship of terrorist and sectarian militias” and missile program.9

The View From Tehran

For the Iranians, however, ballistic missiles are the backbone of the country’s national defense strategy and a symbol of its power projection capabilities in a hostile and unstable neighborhood. Although much attention has been given to Iran’s missile development, it is noteworthy that the country’s quest to acquire indigenous ballistic missile technology dates back to the time of Shah Mohammed Reza Palavi, who was then a close ally of the United States. After Washington refused to sell nuclear-capable Lance surface-to-surface missiles to Iran, the shah joined Israel in a secret multibillion-dollar project code-named Project Flower to develop missiles capable of carrying 1,650-pound warheads a range of up to 300 miles.10 Although the project was abandoned after the 1979 Islamic Revolution, the backbone of Iran’s defense strategy remained its U.S.-supplied air force with state-of-the-art fighter aircraft.

The fall of the shah, the subsequent taking of U.S. diplomats hostage by Iranian student radicals, and the outbreak of the Iran-Iraq war in 1980 fundamentally reshaped Iran’s national defense strategy. The hostage crisis destroyed the U.S.-Iranian relationship and deprived Tehran of its primary source of weapons. Iraq’s invasion of Iran in September 1980 and the systematic use of chemical weapons on Iranian troops and population centers taught bitter and important lessons about the nature of regional threats that left an indelible mark on the Iranian political psyche.

Official UN documents reveal that the Iraqi army began systematically using chemical agents against Iran as early as October 1983,11 and by the end of the war, up to 100,000 Iranian civilians and soldiers had been exposed to these weapons.12 These atrocities were largely ignored by international organizations and world powers, some of whom actively supported Saddam Hussein’s war machine. The United States, for instance, reportedly gave Iraq intelligence on Iranian positions.13

These memories are still raw. As Zarif stated in 2016, “We really wish and hope for the day when nobody spends any money on weapons…. [W]e spend a fraction of others’ expenditure. We are entitled to the rudimentary means of defense, which we need to prevent another Saddam Hussein around the corner to attack us with chemical weapons.”14 In 2018, Abbas Araghchi, Iran’s chief nuclear negotiator, recalled that “[w]e still remember the French Super Etendards [fighter jets], British Chieftain tanks, German chemical weapons, U.S. AWACS planes and Saudi dollars…[which aided Iraq during the war]. Our missile program is defensive.”15

The Value of Missiles

The brutal eight-year conflict also taught Iranians an important lesson on the strategic value of ballistic missiles and their retaliatory function against an adversary’s population centers. Similar to World War II tactics, Iraq during the conflict with Iran launched a variety of ballistic missiles on Iranian population centers, including Shiraz, Isfahan, and Tehran, with the aim of breaking the Iranian will to fight. Over the course of the conflict, Iraqi military units reportedly launched 533 ballistic missiles on Iranian cities, resulting in nearly 14,000 deaths and injuries among Iran’s civilian population.16 Iran initially lacked a ballistic missile capability, but illicitly acquired a small number of Soviet-made Scud missiles from Libya, North Korea, and Syria.17 These missiles set the foundation for Iran’s ballistic missile program.

Equally important for Iran’s security calculations is the country’s current strategic environment and the ongoing military imbalance in the region. To the west, Iran faces an existential threat in nuclear-armed Israel. Because of the decades-long international arms embargo, Iran’s conventional military has been unable to modernize and procure new weapons systems, but its Arab neighbors are among the top customers of advanced U.S. and European military equipment. Iran’s estimated defense spending in 2020 was $12 billion, while Saudi Arabia, Iran’s main regional rival, spent $55 billion dollars in that same period.18 Iran has compensated for its lack of access to an array of modern weapons systems by heavily investing in an arsenal of short- and medium-range ballistic missiles, which can target large population centers and, with improved accuracy, can conduct precision strikes almost anywhere in the Middle East.

In addition to its defensive qualities, the missile program symbolizes Tehran’s power projection capabilities in the Middle East. After a terrorist attack by the Islamic State group in Ahvaz in October 2018, Iran’s Islamic Revolutionary Guard Corps (IRGC), which controls the country’s ballistic missile arsenal, showcased its capabilities by launching six ballistic missiles into Syria targeting Islamic State bases. More significantly, in January 2020, after the United States assassinated General Qassem Soleimani, the commander of the IRGC Quds Force, the IRGC launched a dozen ballistic missiles at the Al-Assad base where U.S. forces were stationed. This strike marked the first state-sponsored attack on U.S. military bases in decades. Although no U.S. personnel died, the attack sent a strong political message that Tehran is willing and capable of directly targeting U.S. military in the region.

These factors can explain the widespread domestic popularity of Iran’s missile program. An Iranian public opinion survey in October 2019 found that 92 percent of respondents believed it is important for Iran to develop its missile program, while 60 percent of respondents view the program as an effective deterrent.19 By February 2021, that number had increased to 66 percent, demonstrating steady support.20

Raisi and the Future of Talks

For the moment, the talks to revive the JCPOA are stalled, primarily due to the transition of power in Tehran and the new administration’s apparent ambivalence about resuming them. Although Raisi and Iranian Supreme Leader Ayatollah Ali Khamenei have spoken in favor of the nuclear negotiations and the lifting of sanctions, several factors have chilled Tehran’s appetite for follow-up talks over Iran’s missile program and its regional activities, as the United States and its European partners are demanding.

In 2016, Hossein Amirabdollahian, then Iran's deputy foreign minister for Arab and African affairs, met in Tehran with UN Envoy to Syria Steffan de Mistura to discuss Syria peace negotiations. Amirabdollahian was just promoted to foreign minister by new Iranian President Ebrahim Raisi. (Photo by STRINGER/AFP via Getty Images)Unlike the Rouhani administration, Raisi and his cabinet are more aligned with Iran’s deep state which works in parallel with the elected bodies, often undermining their efforts to engage the West.21 One example is newly approved Foreign Minister Hossein Amirabdollahian, a career diplomat with ties to the IRGC. Because of his support for Iran’s regional activities, including Iran’s intervention in Syria and support for Houthi rebels in Yemen, he is often referenced as Diplomat-e Movaghemat, or the Resistance Diplomat, a reference to Iran’s “axis of resistance” in the Middle East.22

Amirabdollahian has an academic background in regional affairs and previously served as deputy foreign minister for Arab and African affairs. He speaks fluent Arabic and halting English, meaning that, unlike his predecessor, he likely will find it more difficult to effectively communicate with officials in Washington and Europe. His linguistic skills, regional expertise, and close relationship with the IRGC could enable him to focus on improving Iran’s relations with its neighbors, especially Saudi Arabia.

Regarding ballistic missiles, the Raisi administration’s approach is quite similar to its predecessor’s. The new president has stated that Iran will not negotiate on the program or its regional activities.23 His foreign minister has asserted that “American dreams for negotiations over Iran’s missile program will never come to realization…. Iran’s missile capability is a strategic asset for regional stability.”24 It bears noting that Amirabdollahian, who holds a Ph.D. in international relations, shares the view with “realist” scholars that the essence of foreign policy and international relations is “power.”25 As a result, one should expect Iran’s new chief diplomat to be even more hawkish than Zarif in support of the country’s ballistic missile program.

More importantly, Khamenei, who has the final say on Iran’s national security decisions, deeply distrusts the West and has repeatedly rejected any negotiations beyond the nuclear program. He reinforced this point in July when he said, “In this government, it became clear that trust in the West does not work and they do not help, and they strike a blow wherever they can, and if they do not strike somewhere, it is because they cannot.”26 Two years earlier, he warned that Iran “will not negotiate over the issues related to the honor of our revolution. We will not negotiate over our military capabilities. Negotiations means a deal, meaning that you need to compromise over your defense capabilities.”27 In short, Iran’s key national security decision-makers all favor the country’s missile program and regional interventions, which they perceive to be in the vital interests of the state.

Nevertheless, there may be some wiggle room. Notwithstanding his strong opposition to negotiations over the missile program, Khamenei claimed in June 2021 that he ordered the IRGC to limit the range of Iran’s ballistic missiles to 2,000 kilometers. “At a time we could only produce two types of artillery shells, now we have ballistic missiles with the range of 2,000 kilometers; they [the military] wanted to go to 5,000 kilometers, but I didn’t allow it…. [T]hese precision-strike capabilities are notable,” he said.28 This view has been echoed by Iranian military commanders and reflects the leadership’s threat perception. Brigadier General Amir Ali Hajizadeh, commander of IRGC Aerospace Force, noted in December 2018 that although Iran has no technical limitation on increasing the range of its missiles, the current range satisfies Iran’s existing security needs.29

Reaching a Consensus

The U.S. decision to unilaterally renege on its JCPOA commitments in May 2018 has deepened Iran’s distrust of Western countries. Iran is unlikely to participate in any negotiations that would jeopardize the backbone of its national defense strategy; no sensible country would. Meanwhile, credible reports have indicated that the United States plans to impose new sanctions on Iran’s drone and precision missile capabilities.30 Sanctions alone, however, will not prevent Tehran from advancing its national defense, as security concerns always trump other issues.

An Iranian medium range missile passes by the official reviewing stand in Tehran during the annual military parade in September 2017, marking the anniversary of the outbreak of Iran's devastating 1980–1988 war with Iraq. Iran's diverse and growing missile arsenal concerns the United States and its allies. (Photo by STR/AFP via Getty Images)Nevertheless, a face-saving missile compromise might be achievable. Under favorable circumstances, the United States and Iran could agree to codify Tehran’s self-imposed 2,000-kilometer-range into a formal agreement. That is far less than what the United States has advocated, but at least it would restrain the program somewhat. Washington must be willing, however, to reciprocate Tehran’s concessions and recognize its legitimate security concerns. In the words of U.S. Secretary of State Henry Kissinger, “[E]very agreement generally reflects reciprocal concessions rather than unilateral satisfaction.”31 Washington has a number of options in its foreign policy toolbox. These concessions could include a U.S. commitment not to prevent other countries from selling conventional weaponry to Iran or a commitment to lift sanctions on Iran’s missile program, if such a framework is reached.

With regard to Iran’s regional activities, the United States should take a hands-off approach and instead throw its diplomatic and political support behind a regional dialogue that offers the possibility of a favorable outcome for all regional powers, including Iran. Washington, for example, could support the ongoing talks between Riyadh and Tehran, which aim to mend relations between the two regional powers. Recently, Amirabdollahian attended the Iraqi Neighboring Countries Conference, which was aimed at supporting Iraq.32 He also met with a number of Arab leaders, including Kuwait’s foreign minister and Sheikh Mohammed bin Rashid Al Maktoum, emir of Dubai. “What we need more than ever today is sustainable regional security…. Security can only be achieved through mutual trust between the countries of the region…strengthening communication and good neighborliness,”33 Amirabdollahian observed. Such initiatives can create a platform for regional leaders to meet and discuss their outstanding issues, including the devastating war in Yemen. A framework for considering the interests of all parties could advance regional peace and stability and enable the United States to focus more of its attention on the rise of China.

Raisi’s inauguration marks a hostile takeover by hard-liners in all three branches of Iran’s government. Given the alignment of views among Raisi, Amirabdollahian, and Khamenei, in addition to the IRGC, the resulting synergy is certain to create a more homogenous and effective decision-making environment within national security circles, potentially leading to a more assertive Iran. In other words, Raisi’s tenure fills the gap between what Zarif once dubbed “diplomacy and field,”34 a reference to the struggle between the Foreign Ministry and the IRGC in determining and executing Iran’s foreign policies in the region.

To produce the economic results so vital to Iran’s survival, the Raisi administration is certainly interested in and requires a revival of the JCPOA and the termination of what has been effectively the economic strangulation of Iran. Even so, the U.S. withdrawal from the JCPOA has left very little appetite or political capital in Tehran to negotiate with Washington and European capitals. Perhaps the best approach for the Biden administration is first to revive and then implement the nuclear agreement in good faith, allowing Iran to see the benefits of negotiations. Only after that is Tehran likely to be amenable to follow-on negotiations to reach a broader framework agreement with Washington.


1. “Iran’s Missile Program Not Subject to Negotiations, Zarif Says,” Tehran Times, January 20, 2021.

2. “Iran Nuclear Deal: Trump’s Speech in Full,” BBC, October 10, 2017, https://www.bbc.com/news/world-us-canada-41617488.

3. “Antony Blinken on Iran,” The Iran Primer, June 25, 2021, https://iranprimer.usip.org/blog/2021/jan/21/antony-blinken-iran.

4. Rachel Oswald, “Blinken Tells House Panel to Expect Firmness Toward China,” MSN, March 10, 2021.

5. Julian Hattem, “Obama: Iran Not Following the Spirit of the Deal,” The Hill, April 1, 2016.

6. James R. Clapper, Statement on the worldwide threat assessment of the U.S. intelligence community, February 9, 2016, https://www.armed-services.senate.gov/imo/media/doc/Clapper_02-09-16.pdf.

7. Daniel R. Coats, Statement on the worldwide threat assessment of the U.S. intelligence community, January 29, 2019, https://www.dni.gov/files/ODNI/documents/2019-ATA-SFR---SSCI.pdf.

8. “Iran Deal: EU United in Keeping Iran Nuclear Deal in Place for European Security,” European Union External Action Service, May 29, 2018, https://eeas.europa.eu/topics/multilateral-relations/45352/iran-deal-eu-united-keeping-iran-nuclear-deal-place-european-security_en.

9. “Gulf States Want Iran Deal Talks to Address Tehran’s Missiles Program, Support for Proxy Groups,” Al-Monitor, June 16, 2021, https://www.al-monitor.com/originals/2021/06/gulf-states-want-iran-deal-talks-address-tehrans-missiles-program-support-proxy.

10. Elaine Sciolino, “Documents Detail Israeli Missile Deal With the Shah,” The New York Times, April 1, 1986.

11. UN Security Council, “Letter Dated 9 November 1983 From the Permanent Representative of the Islamic Republic of Iran to the United Nations Addressed to the Secretary-General,” S/16140, November 10, 1983.

12. Marcus George, “Insight: After Syria, Iran Laments Its Own Chemical Weapons Victims,” Reuters, September 13, 2013.

13. Shane Harris and Matthew M. Aid, “CIA Files Prove America Helped Saddam as He Gassed Iran,” Foreign Policy, August 26, 2013, https://foreignpolicy.com/2013/08/26/exclusive-cia-files-prove-america-helped-saddam-as-he-gassed-iran/.

14. “Iran FM Javad Zarif Responds to a Reporter's Question Regarding Ballistic Missiles,” YouTube, April 20, 2016, https://www.youtube.com/watch?v=ejudkZgs5Vg.

15. Sina Azodi, “U.S. Should Offer Incentives for Iran Missile Testing Moratorium,” Atlantic Council, February 20, 2018, https://www.atlanticcouncil.org/blogs/iransource/us-should-offer-incentives-for-iran-missile-testing-moratorium/.

16. Ali Khaji, Shoadin Fallahdoost, and Mohammad Reza Sorush, “Civilian Casualties of Iranian Cities by Ballistic Missile Attacks During Iran-Iraq War (1980-1988),” Chinese Journal of Traumatology, Vol. 13, No. 2 (April 1, 2010).

17. “Shahab-1 (Scud B-Variant),” Center for Strategic and International Studies, July 31, 2021, https://missilethreat.csis.org/missile/shahab-1/.

18. “SIPRI Military Expenditure Database,” Stockholm International Peace Research Institute, https://www.sipri.org/databases/milex (accessed September 14, 2021).

19. Center for International and Security Studies at Maryland (CISSM), “Iranian Public Opinion Under ‘Maximum Pressure,’” October 2019, https://cissm.umd.edu/sites/default/files/2019-10/Iranian%20PO%20under%20Maximum%20Pressure_101819_full.pdf.

20. CISSM, “Iranian Public Opinion, at the Start of the Biden Administration,” February 2021, p. 28, https://cissm.umd.edu/sites/default/files/2021-02/CISSM%20Iran%20PO%20full%20report%20-02242021_0.pdf.

21. Mohammad Ayatollahi Tabaar, “Iran’s War Within,” Foreign Affairs, September/October 2021, https://www.foreignaffairs.com/articles/iran/2021-08-05/irans-war-within-ebrahim-raisi.

22. “Naagofte-haye Diplomat-e Moghavemat az Nabard-e Shaam,” Islamic Republic News Agency, January 7, 2017, https://www.irna.ir/news/82787902/.

23. Erin Cunningham and Kareem Fahim, “Raisi Says Iran’s Ballistic Missiles Are Not Negotiable, and He Doesn’t Want to Meet Biden,” The Washington Post, June 21, 2021.

24. “Ro’yaye kelid Khordan e Moazekereh Moushaki Iran Hargez Ta’bir Nemisahavd,” Iranian Students News Agency, September 21, 2018, https://www.isna.ir/news/97063014707/.

25. Iran Documentary, “Hossein Amir-Abdollahian Interview With Dast-Khat Documentary,” YouTube, June 24, 2020, https://www.youtube.com/watch?app=desktop&v=fUWe8g3F1X4.

26. Rick Gladstone, “Khamenei Adds to Doubts in Iran Nuclear Deal Talk,” The New York Times, July 28, 2021.

27. “Ali Khamenei: We Will Not Negotiate Over Issues Related to the Honor of Revolution,” Radio Free Europe, May 29, 2019, https://www.radiofarda.com/a/f4_ali_khamenei_statement_iran/29970492.html (in Farsi).

28. Ali Javid, “Iran Ayatollah Khamenei: Missile and Range,” YouTube, June 17, 2021, https://www.youtube.com/watch?v=Cdk0VqG7Ni4.

29. “Iranian General Says Nation Can Extend Missile Range Beyond 2,000 kilometers,” The Times of Israel, December 10, 2018.

30. Ian Talley and Benoit Faucon, “U.S. Plans Sanctions Against Iran’s Drones and Guided Missiles,” The Wall Street Journal, July 29, 2021.

31. Henry Kissinger, Diplomacy (New York: Simon and Shuster, 1994), p. 740.

32. Sara Masoumi, Twitter, August 14, 2021, https://twitter.com/SaraMassoumi/status/1430200197972381702?s=20.

33. “Deepening Ties With Neighbors a Priority of Raisi’s Foreign Policy,” Tehran Times, September 6, 2021.

34. Parisa Hafezi, “In Leaked Recording, Iran’s Zarif Criticises Guards’ Influence in Diplomacy,” Reuters, April 26, 2021.


Sina Azodi is a nonresident fellow at the Atlantic Council and a lecturer of international affairs at the Institute for Middle East Studies in the Elliott School of International Affairs at the George Washington University. He is also a Ph.D. candidate in international relations at the University of South Florida, where he studies Iran’s nuclear program.

New leadership in Iran, historical factors and a complicated geostrategic environment are driving
Iranian decision-making, thus making compromise with the West on the nuclear deal unlikely.


Subscribe to RSS - Features