In April, while most of the world was focused on defeating a devastating viral pandemic, the US State Department quietly released its annual compliance report, describing whether and how the United States and other countries have been abiding by various arms control agreements. The report is sober reading for those hoping that the coronavirus would usher in a new era of international collaboration.
The report made waves for raising “concerns” about China’s adherence to a “zero-yield” nuclear testing standard, as called for by the 1996 Comprehensive Nuclear Test Ban Treaty. Although neither the United States or China has ratified the treaty, both have signed it, and both claim to abide by a nuclear testing moratorium.
US allegations in this regard are nothing new; for decades, the United States has accused Russia of conducting low-yield nuclear tests. And in the 2020 report, those accusations are repeated. The report claims—without offering sources in its unclassified part—that “Russia has conducted nuclear weapons-related experiments that have created nuclear yield.” In its response, Russia categorically denied conducting “non-zero-yield tests.” It also made clear that it does “not intend to discuss compliance with the nuclear test ban” with the United States, which “has been delaying the ratification of the [treaty] for 25 years.”
Although US accusations are unlikely to be true, they could give a convenient pretext to officials who want to withdraw the US signature from the treaty, allowing the United States to resume its own nuclear testing. In fact, that may be the entire point.
Background on US accusations. The compliance report builds on concerns voiced by Lt. Gen. Robert P. Ashley, director of the Defense Intelligence Agency, in 2019. At a public event held in May, Ashley said that the United States “believes that Russia probably is not adhering to the nuclear testing moratorium.” Later in the seminar, he backtracked slightly, maintaining that “we believe [the Russians] have the capability to [test] the way they are set up.”
By August, when the State Department released the 2019 report, there was greater certainty: “The United States assesses that Russia has not adhered to its nuclear testing moratorium.”
It would appear—taking statements at face value—that the United States came across information in that three-month span, and perhaps even more over the last year, that would elevate its assessment from “believes that” to “assesses that” and from “probably” to “is.” In the language of estimative probability, this is quite a jump in assessment and would need to have been underpinned by some hard evidence. What this evidence might be, if it indeed exists, is subject to speculation.
Of course, US accusations about Russian nuclear testing go much further back than 2019. In 1997, the United States falsely accused Russia of conducting a nuclear test and summoned the Russian ambassador in Washington for an explanation. It turned out to be an earthquake. In the article “Russia: Of Truth and Testing,” published in 2002, Michael Jasinski, Christina Chuen, and Charles Ferguson go through similar compliance allegations raised at regular intervals throughout the late 1990s. In a sidebar, Ferguson casts doubt on the timing of the release of information, noting that accusations tended to happen ahead of critical decisions on nuclear test ban treaty negotiations.
Why it’s hard to verify the claims. If the US accusations were true, Russia’s actions would violate Article I of the treaty, which Russia ratified in June 2000. If it had conducted a test, however, it would not have been a large explosion, as this would have been picked up by the international monitoring system that was set up by the treaty’s verification organization. This worldwide network of sensors can now reliably detect explosions yielding between 12 and 40 metric tons of TNT equivalent throughout Russia. Coverage of Russia’s Arctic test site, in Novaya Zemlya, is sensitive enough to detect explosions of even lower yield. Given that the monitoring system’s data is shared with both member states and selected academic institutions, it is unlikely that a significant event would have happened on the island without anyone picking up on it.
Instead, what the United States is accusing Russia of conducting is called a hydronuclear test. These smaller experiments involve using explosives to compress a fissile material enough for it to behave as a fluid. Such explosions create a chain reaction and exceed the zero-yield standard, but, unlike a full-scale nuclear test, release only a negligible amount of fission energy. The yield of a hydronuclear test may be just a few kilograms or less of TNT equivalent. Test site data show that the Soviet Union conducted hydronuclear tests on a routine basis, but Russia claims to have ceased such activities when it signed the treaty in 1996.
Why test a device at such a small scale? There are several reasons, ranging from understanding the chain reaction to establishing the optimal moment to use a neutron initiator. There are also other uses, such as getting better data about the physical properties of the material or training test site personnel in the use of potent high explosives combined with toxic and radioactive heavy metals.
The treaty bans nuclear explosions, including hydronuclear testing, but it doesn’t prohibit all nuclear experiments. For example, it doesn’t prohibit preparations for nuclear tests, meaning that nuclear-weapon states can continue to maintain and staff their sites, and even place devices in boreholes or tunnels, provided that they don’t set them off. Both Russia and the United States conduct subcritical experiments (weapons-related work not involving an explosive chain reaction) and have been for decades. Since they are not generating yield, they are permissible under the treaty. In its rebuttal of the compliance report, Russia makes clear that it carries out “so-called subcritical tests, which,” it adds, “in no way run counter to our obligations in this area.”
Subcritical tests are tough to distinguish from hydrodynamic tests without an on-site presence. Because there is no international or US inspection team present at Russian sites, many independent analysts have cast doubt on the idea that the United States could claim with such certainty that Russia is violating the zero-yield standard.
Why now? Last year, Jeffrey Lewis, an arms control expert at the Middlebury Institute of International Studies at Monterey, pointed out on Twitter that the debate about whether Russia is conducting hydronuclear testing is subject to an “unstated but powerful bias.” The bias is a vestige of the 1990s, when the test ban treaty was being negotiated. At that time, there was concern among some US officials that the Stockpile Stewardship Program, which is designed to keep the US arsenal safe and reliable, could not be run without hydronuclear tests. Because the officials knew that the national laboratories wanted to continue these types of tests, they were inclined to believe that the Russians also wanted to continue them. Two decades later, however, the science underpinning the program has matured, and there is no such need. But old biases die hard.
What’s more, the Trump administration has been advised by serious arms control skeptics, Ambassador John Bolton being the most prominent example, who have seized on this concern to advocate for the US to remove its signature from the treaty. Whether the skeptics remain a dominant force within the White House following Bolton’s departure is unclear. Secretary of State Mike Pompeo has adopted the line held by the George W. Bush administration, stating in his confirmation hearing that the United States intends to abide by the testing moratorium while continuing to support the test ban treaty organization’s “development and operation of the International Monitoring System.”
Beggars can’t be choosers, and even partial US commitment to this vital treaty is something to hold on to. But this hold may be tenuous. The ongoing pandemic has diverted attention from virtually all other international issues, creating space for those that want to see the treaty’s undoing.
Why withdrawal would be lose-lose. If the United States is simply looking for a pretext to withdraw its signature so it can resume nuclear testing, that would be a lose-lose proposition.
First, it would give up a constraint on its strategic rivals, without receiving any clear benefits from its newfound freedom of action. While most nuclear weapon states have retained their capabilities to conduct tests by maintaining their tests sites and keeping staff on the books, they would all face different challenges to a resumption.
For China and the Russia, these obstacles can be overcome quickly, due to the nature of their political systems. Their test sites are maintained and appear to be in a state of readiness. For them, the main question would be whether they want to lose their diplomatic advantage by moving first to break the moratorium. If either judges that it does, in fact, need to conduct nuclear testing to modernize its arsenal, it might go ahead. Without the constraints of the treaty, Russia and China would be free to choose whatever means of testing they desire.
Meanwhile, if the United States judges that its own Stockpile Stewardship Program cannot ensure the reliability of its arsenal, a finding that would go against the established consensus in the nuclear laboratories, it would face domestic challenges to any resumption. US officials would probably need to seek the consent, for instance, of the relevant state governor, or face public backlash. Abrogating the test ban would seem to give only limited gains, in return for considerable challenges.
Second, the United States would have even fewer means by which to monitor or verify its rivals’ adherence to whatever testing norms remained. The US has formidable national technical means at its disposal, but these will not be able to clarify what is going on at the Chinese and Russian tests sites as effectively as an on-site presence would. Moreover, it would be difficult to come to a separate, post-treaty arrangement with the Chinese and the Russians. Recent attempts to rope the Chinese into a tripartite arms control effort illustrate some of the difficulties with such an approach. It is also unnecessary, as the treaty already gives scope for such arrangements on top of its existing verification measures.
Perhaps the worst consequence of withdrawal, though, is that the United States would give up leverage to prevent future North Koreas from trying to join the nuclear club.
What can be done? The United States should make clear that it possesses hard evidence of Russian noncompliance, and that it will share suitably sanitized information with its allies. If the Trump administration intends to make the case that Russian noncompliance gives it reasonable cause to withdraw its signature from the treaty (something that will require the US Senate to strike ratification off its agenda) its interests would be best served by making clear what information it holds. Should it withdraw on opaque grounds, it would open itself up to allegations that it is abrogating an international agreement for purely ideological and partisan reasons. It would also give rise to concerns amongst friends and rivals that it intends to resume nuclear testing.
Ironically, the whole case illustrates why the United States should not withdraw its signature but rather go in the opposite direction and fully ratify the treaty. Had the United States ratified the treaty previously, it could have started to resolve these concerns within the treaty framework.
Imagine that the United States and Russia had both ratified the treaty some 20 years ago and that this issue came up after many years of peaceful implementation. The two sides would have first been encouraged to resolve this concern among themselves. Treaty membership would also have come with the ability to request assistance from the technical secretariat of the treaty’s monitoring organization, or even the full weight of the executive council, the treaty’s highest decision-making body. International inspectors can’t stop a treaty violator, but they can make a violator lose face and standing, forcing it to unconvincingly obfuscate its transgressions.
If the hypothetical US attempt to resolve the compliance issue failed, the United States could have requested an on-site inspection. If Russia had nothing to hide, it would have allowed it. But if Russia did have something to hide and tried to foil the investigation, a breach of the treaty would have been established in the minds of many other governments. From the US perspective, this would be a much better diplomatic starting point than relying on its own—mostly unshareable—national assessments. It is much easier to marshal a sustained diplomatic offensive to rectify bad behavior when evidence is widely shared and neutral third parties can be convinced.
Additionally, state parties are free to reach further agreements under the treaty. For instance, the United States and Russia, if both were parties, could agree to mutual visits falling short of on-site inspections. They could decide on close monitoring of nuclear test sites. They could agree on the notification and monitoring of permitted activities, such as subcritical testing. Because the United States has not ratified, these options are not on the table. But it’s not too late.
Editor’s note: This article has been updated.
The Bulletin elevates expert voices above the noise. But as an independent, nonprofit media organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.