The last few years were not a good time for arms control treaties. In 2023, Russia suspended its participation in New START and formally pulled out of the Treaty on Conventional Armed Forces in Europe. Attempts to restart the US-Russian dialogue on strategic stability have failed because Russia firmly insists on linking any progress in arms control to a broader normalization of relations with the United States—which, given the continuing war in Ukraine, is a very distant prospect.

In another disturbing development that year, Russia officially withdrew its ratification of the Comprehensive Nuclear Test Ban Treaty (CTBT) in November, prompting concerns that this was a step toward the resumption of full-scale nuclear tests. The fact that it happened in the context of the war in Ukraine added to the tension, especially as some Russian propagandists and experts close to the government were advocating for some kind of a nuclear gesture that would demonstrate Russia’s resolve to curtail western support of Ukraine.

The history of the issue, in fact, predated the war in Ukraine, and some signs suggested that the resumption of tests was not just a theoretical possibility. Russia has long maintained activity at its nuclear test site at Novaya Zemlya, as have the United States and China at their respective sites in Nevada and Lop Nur. Satellite images show that the level of activity at all sites has increased in the recent years (Lewis, 2023).  And while some of it appears to be civilian construction, there are also tunnels, the purposes of which are difficult to determine. China also appears to be building shafts that are more suitable for full-scale nuclear explosions than for experiments (Babiarz, 2023; Nilsen, 2023). All three states, along with France and the United Kingdom, have declared a testing moratorium, and all insist that they do not conduct any experiments that would violate the CTBT ban on nuclear explosions. But all three keep their test sites ready for a potential resumption of full-scale tests.

The concern has been that a nuclear test in Russia could prompt others to follow suit, ending the moratorium that has been in place for almost three decades. Russia has emphasized its reliance on nuclear weapons to keep the West out of the full-scale war that it started in February 2022. It has succeeded in preventing a direct intervention, but was unable to stem the flow of military assistance to Ukraine. It also discovered that its nuclear rhetoric was met with very strong pushback from the international community (Podvig, 2023). In this situation it was reasonable to suspect that the Kremlin might turn to a nuclear test as a way to show its determination. According to reporting from Russia, by the summer of 2023, discussions of various options related to nuclear testing were well underway in Moscow’s policy circles.

These discussions, however, were constrained by one major factor. While the CTBT has not yet entered into force, it has established a fairly strong norm against nuclear testing. North Korea is the only country to have conducted a nuclear test since 1998, and even it has not tested since 2017. Breaking a de facto norm that has been in place for more than 25 years is not an easy step, especially when it is clear that it will be met with almost universal disapproval by the international community. That disapproval would be even stronger if the state that goes first creates an excuse for others to follow.

The way Russia handled its relations with the CTBT in 2023 showed that this constraint is real. In his annual address to the parliament in February 2023, the Russian president stated that Russia “will not be the first to proceed with [nuclear] tests” (Putin, 2023). Russian diplomats  used this message to tell the internal audience that even friendly countries “will be upset if … Russia is the first to resume nuclear tests” (Ulyanov, 2023). Diplomats may not wield much influence in Moscow these days, but they have used what they have to push for the least damaging de-ratification option. Russia remained a signatory to the treaty, keeping intact all institutional structures that allow it to participate in the work of the CTBT Preparatory Commission. Indeed, shortly after the ratification was rescinded, Russia proudly inaugurated a seismic station that completed its national segment of the International Monitoring System. (See map below.)

This shows the power of institutions. It is very important that the norm against nuclear tests is institutionalized through the work of the CTBT Preparatory Committee and the operation of the International Monitoring System. Without these institutional structures, as well as those that were created in Russia to interact with them, Russia’s withdrawal from the treaty could have put its signatory status in question. Also, the Comprehensive Nuclear Test Ban Treaty Organization is a vital element of the nuclear test moratorium. Yes, there are limits to what it can do without the full legal power of the treaty, but it is almost certain that it will be able to detect any meaningful nuclear test.

All this suggests that the moratorium on full-scale nuclear tests will continue to hold. Russia is on the record saying that it will not start the tests unless the United States starts first. And the United States is on record that it has no intention—or need—to do so. While it is understandable that Russia’s official pronouncements cannot be taken at face value, one would expect at least a shift in rhetoric before any move to resume full-scale nuclear explosions. Similarly, despite all the activity at its nuclear test site at Lop Nur, it is highly unlikely that China will start nuclear tests out of the blue.

This does not mean that the moratorium is completely safe, though. The activity at the test sites creates considerable room for suspicion and misunderstanding. Russia, for example, may use the activity at the US test site in Nevada as a pretext for breaking the moratorium, claiming that these experiments cross the nuclear test threshold.

The good news is that the United States understands that the scale of activity in Nevada may indeed raise concerns and has undertaken steps to address them. Most importantly, the United States has made a commitment to the transparency of its experiments, declaring that it is “open to hosting international observers for monitoring and verification research and development on our subcritical experiments” (Hruby, 2023b) The full scope of this monitoring activity is not clear yet and it appears that the United States expects reciprocity from Russia and China before moving forward (Hruby, 2023a). While it is quite unlikely that Russia or China will let US inspectors into their test sites, this should not hold the US initiative back. For example, it can explore the possibility of hosting inspectors from other states, potentially in cooperation with the Comprehensive Test Ban Treaty Organization. This arrangement can become a very important element of the effort to keep the testing moratorium in place.

Another threat to the moratorium comes from a dispute about activities in Russia and China. Like the United States, these countries maintain that all their experiments are compliant with the central obligation of the CTBT “not to carry out any nuclear weapon test explosion or any other nuclear explosion.” However, since the treaty does not contain a definition of a nuclear explosion, there is some room for uncertainty here.

The United States maintains that its moratorium uses a strict “zero-yield” standard that requires that the experiments it conducts at its test site in Nevada are not supercritical, meaning that they do not produce a self-sustaining chain reaction. France and the United Kingdom appear to agree with this approach. As for China and Russia, they also agreed during the CTBT negotiations that the treaty prohibits nuclear explosions without any threshold. However, they have refrained from commenting on the US understanding of the zero-yield standard or suggesting their own.

In the compliance report issued by the US Department of State in June 2020, the United States stated that Russia “has conducted nuclear weapons experiments” that are not consistent with the US definition of zero yield (US Department of State, 2020). The report also stated that China’s activity raised concerns about adherence to the US standard, but it stopped short of saying that China had conducted supercritical experiments. The charges were repeated in 2021 and 2022, although the language of the reports indicates that if any supercritical experiments have taken place they were conducted before 2019. Other than the construction activity at the sites, no new causes for concerns were identified after that.

The extent to which these charges are substantiated is not clear, since the details are hidden in a classified annex. It can be something as convincing as an intercepted document that describes the experiments or something much less reliable. But these allegations, if sustained, can have serious consequences for the CTBT. Indeed, the United States did discuss the possibility of carrying out a nuclear test shortly after the publication of the Compliance Report in 2020 (Hudson and Sonne, 2020).

The argument put forward by the United States is that supercritical experiments allow Russia (and by extension China) to develop new and better nuclear weapons and eventually gain an advantage over the United States, which constrains itself to the strict “zero yield” definition. Moreover, the argument goes, since the treaty lacks a definition of a nuclear explosion and since these experiments are below the International Monitoring System’s detection threshold, they can be continued even if the treaty is in force.

The latter part of the argument is manifestly wrong. While certain activities indeed cannot be detected remotely by the International Monitoring System, the CTBT verification system is not limited to the International Monitoring System alone. Once the treaty is in force, it will allow its parties to use other tools, such as on-site inspections. This fact has been known all along, which is why the US allegations emphasize the “lack of any written, mutually agreed definition of a ‘nuclear weapon test explosion.’ ” This challenge is more difficult to address, as there is indeed no written definition of a nuclear explosion. It may appear that the CTBT prohibition would be impossible to verify even if the treaty were in force.

The treaty, however, does define a nuclear explosion, even though it does so in an indirect manner. It created a formal mechanism for raising and resolving compliance concerns, as well as the monitoring system and specific verification procedures. In effect, a nuclear explosion is defined as any activity that the parties determine to be non-compliant with the treaty’s central obligation after they follow the procedures included in the treaty. This way to define a nuclear explosion is flexible enough to resolve concerns that the parties may have—and, importantly, gives all states an opportunity to weigh in.

One problem with this approach is that it relies on the treaty being in force. Until the mechanism for reconciling various positions is available, states can be guided by their own understanding (Ifft, 2020). As we have seen from the allegations put forward by the United States, this creates room for misunderstandings and can endanger the nuclear test ban regime.

The idea that by applying a less stringent constraint to its nuclear test program a state can gain a significant advantage over other states is quite entrenched in the nuclear weapons debate. It is one of those notions that are extremely difficult to challenge from outside of the nuclear weapons complex without knowing classified details of the modern weapon design. But there is plenty of evidence that suggests that the alleged strategic advantage is an artificial construct.

For example, the Russian weapon designers reported that they developed, apparently from scratch, and delivered to the country’s strategic aviation a new high-yield gravity bomb between 2000 and 2010 (Andreev, 2010). Even if we assume that this development would not be possible without supercritical experiments, what kind of strategic advantage could this new weapon possibly bring? Indeed, it is quite likely that even the ability to conduct full-scale nuclear explosions would not significantly advance any of the existing nuclear weapon programs, probably not even the North Korean one at this point.

This argument, however, opens the CTBT to a criticism from a different direction. It is clear now that the nuclear test ban has not and probably cannot prevent modernization of the existing arsenals or development of new nuclear weapons. The Stockpile Stewardship Program in the United States, guided by its own strict definition of a nuclear explosion, includes a number of subcritical experiments that provide plenty of information about the behavior of nuclear weapon components. Similar programs certainly exist in other states and while they may be less advanced than the US program, they are likely adequate for these states’ needs. If this is the case, the question is: What is the value of the treaty, especially given that the prospect of it coming into force is rather remote? (Mian, 2016).

The answer to that is probably that the treaty has established a norm against nuclear testing, even without being in force. This norm withstood the challenge of Russia’s de-ratification of the treaty as well as that of US discussions in 2020 of a potential resumption of nuclear tests. This is not to say that the norm against full-scale nuclear testing will continue to be upheld automatically. Like any arms control and disarmament agreement, the CTBT is not immune to political pressure and interference.

The politics, however, are shaped by people, and the expert community can do its share to influence the debate. The value of the existing ban on nuclear explosions and the important role of the International Monitoring System are already widely recognized, yet these points are worth repeating: The debate about the precise definition of a nuclear explosion should not distract the international community from the existing capabilities of the system.

While US concerns about possible supercritical nuclear tests in Russia and China should not be ignored, they must be considered within an appropriate context. It would be helpful, of course, if the United States would release more details about the basis for its concerns. However, this is unlikely to happen. The accusations were clearly a political gesture, and the United States has every incentive to keep the details hidden if only to ensure that Russia and China have no realistic way to defend themselves in the court of international opinion. Neither they would be willing to do so, because of the largely political nature of the affair. Even if Russia and China agree to disclose more information about their experiments, the United States can always ask for more. Having said that, more transparency on their part would definitely help address the tensions, and Russia and China must be urged to reiterate their commitment to CTBT in the strongest possible terms.

The only reliable way to resolve the dispute about the definition of a nuclear explosion is to ensure that the treaty is in force and use its mechanisms to settle the differences. Today, the expert community should push back against the notion that these differences, whether real or not, provide any state with a significant advantage in developing nuclear weapons. While some experiments do provide more data than others, there is no reason to believe that this translates into a particularly significant gain.

Even if we accept that the disparity exists, it is clearly wrong to conclude that it will be able to persist once the treaty comes into force. Theoretically, while waiting for this to happen, nuclear weapon states could agree on an arrangement that includes transparency measures designed to demonstrate their adherence to a common zero yield standard. However, in the current political climate an agreement of this kind is all but impossible, and should the climate change, it is likely to become unnecessary. While any progress in this direction is welcome, the lack of progress should not be used as an argument against the treaty or undermine the existing moratorium.