David Logan on China's Nuclear Forces, Nuclear Signaling, and a Potential Taiwan Conflict

Have there been any interesting events in China's nuclear force development in the year since you and Phillip Saunders published this monograph on the subject last year? If so, have these events affected your perception of the trajectory of China’s nuclear force development (or is it too soon to tell?)

​

We published our monograph in July 2023. There have been a few developments related to China’s nuclear forces since then, but I’m not sure any of them provide significantly new information that should affect our perceptions of the trajectory of China’s nuclear force development. I’ll focus on three here. First, we have new data which shows that, unsurprisingly, the buildup has continued. The 2023 China Military Power Report from the U.S. Department of Defense assesses that China “possessed more than 500 operational nuclear warheads as of May 2023.” The 2023 version of the CMPR also maintained the prediction that China will “probably have over 1,000 operational nuclear warheads by 20230.” However, it did not include the prediction mentioned in previous versions of the report that, if China’s buildup continued at its current pace, the country would have 1,500 warheads by 2035. I’m not sure whether that indicates a revision to expectations about the eventual extent of the expansion. We should hopefully have a new China Military Power Report soon, which may give us more insights into China’s ongoing nuclear buildup.

 

Second, media reports in early 2024 claimed that corruption within the PLA Rocket Force, the military organization that operates China’s land-based ballistic and cruise missiles, had seriously undermined the force’s readiness. The reports, based on anonymous sourcing from U.S. officials, alleged that some of the missiles operated by the Rocket Force were filled with water, not fuel, and that some of the lids on China’s ICBM silos were malfunctioning. This comes after several senior leaders with the Rocket Force and other parts of the PLA have been removed. This indicates continued challenges in rooting out corruption from the PLA but also that this corruption might undermine the readiness of even the most sensitive and politically significant elements of the Chinese military.

 

Third, and most recently, China conducted a test of an ICBM over the Pacific. This was the first time the country conducted such a launch over international waters since 1980. It’s not clear why the launch was conducted. It may have been part of a signaling effort to demonstrate to the world that China’s strategic nuclear forces are credible. It may have been to send a similar signal to domestic audiences. It may have been to bolster morale within the Rocket Force, especially after reports of rampant corruption. One aspect of the test that may provide some grounds for cautious optimism is that China reportedly informed the United States of the launch in advance.

 

But we still lack official acknowledgement from the Chinese government of the country’s ongoing buildup, which means that we are left to speculate about its drivers, its goals, and the ultimate endpoint.

 

I have often gotten the sense that public discussion of a potential Taiwan conflict features insufficient consideration of the nuclear dangers therein. Is this your feeling? Why or why not? 

 

There would certainly be nuclear dangers in a potential conflict over Taiwan, and I think it’s a good question about to what extent these dangers are incorporated into discussions about such a conflict. A conflict over Taiwan could well see conventional strikes by one nuclear weapon state (the United States) against the territory of another nuclear weapon state (China). Add in concerns about conventional-nuclear entanglement, the high stakes for such a conflict, and domestic political pressures for escalation and the dangers may be even higher. 

 

We might see some of this in the decision of wargamers to exclude from their simulations of a possible Taiwan conflict the possibility of nuclear use. This may be a very reasonable methodological decision to try to better model certain kinds of conventional operations. And even when the wargames are designed to exclude players using nuclear weapons, we still see the players in these games acting as if nuclear escalation is a concern, such as U.S. players limiting their strikes against Chinese territory. But this is certainly a very difficult thing to model.

 

I haven’t seen evidence about the U.S. public’s awareness of the nuclear risks in a Taiwan conflict, but I wouldn’t be surprised if it is relatively low and that this low awareness influences public policy preferences. Using survey experiments of the U.S. public and a subset of U.S. national security practitioners, I find evidence that those with higher levels of knowledge about nuclear weapons are less likely to support nuclear use. I also find that, even in a scenario envisioning “limited” nuclear use against Russia, a nuclear weapon state, U.S. respondents are quite willing to support those strikes. Nuclear weapons are not a particularly salient issue today and certainly less than during the Cold War. This may also influence people’s beliefs about the nuclear dangers in a potential Taiwan conflict. 

​

For those who do not have an advanced understanding of nuclear theory, could you briefly explain what nuclear signaling is? How is it defined, and what are some ways it has tended to occur in the past? Is signaling seen as effective in accomplishing military or political goals? Does signaling carry any risks for the signaler?

​

I’m not sure that I’ve seen a formal definition of nuclear signaling, but you might think of it as actions taken to increase the perceived risks of nuclear escalation in order to gain some kind of coercive advantage. When both sides of a conflict are perceived to possess sufficiently survivable strategic nuclear forces, there are inherent challenges to using or threatening to use nuclear weapons. Nuclear use and, by extension, nuclear threats are seen as not credible under conditions of mutually assured destruction.

 

We might think of nuclear signaling as a set of behaviors that intend to raise the credibility of those threats to either deter or compel a target. The goal here would be to make the target believe that nuclear dangers are higher. This might be done by simply referencing the nuclear dimension of the conflict and emphasizing to the target that nuclear escalation is a possibility. It might be done by taking steps which suggest that you are willing to deliberately escalate to nuclear use under certain circumstances. Or it might be done to make the target believe that inadvertent or accidental nuclear escalation could happen even if neither side considers it rational.

 

Nuclear signals could take many forms, as we can see with Russia’s actions as part of its invasion of Ukraine. They could include statements by senior leaders or non-governmental observers alluding to Russia’s nuclear forces or the risks of nuclear escalation. It could be mobilization of nuclear-capable delivery systems or increasing their alert status. It could be changes to the country’s nuclear posture. It could be even non-nuclear signals that increase the overall perceived risks of escalation or the perception that you are willing to run high risks and pay high costs generally, which the target might use to infer your willingness to run nuclear risks.

 

There certainly are dangers for the one sending these signals. Of course, if they increase the risk of escalation and you face the prospects of paying the costs of that potential escalation, then this is a danger. They could, in certain circumstances, reduce domestic support for the conflict among either the public or elites if they view that nuclear signals are inappropriately raising the risks of escalation. Sending nuclear signals could also result in diplomatic costs to the extent that international audiences view you as violating a “nuclear taboo.”

 

The challenge is that crises and conflicts can be very messy, and one side may misread the actions of the other side by either over- or underestimating the credibility of signals. This can be dangerous.

 

Do you have a personal opinion on the likelihood of China engaging in nuclear signaling in a conflict with the United States (and perhaps Japan) over Taiwan? How do you think such signaling would transpire? Do you think this signaling would be sufficient to shape the outcome of a conflict?

 

It is difficult to say whether China might engage in nuclear signaling in a crisis or conflict involving the United States over Taiwan.

 

It might depend on what one means by nuclear signaling. If Chinese leaders believe that the likelihood of a conflict is sufficiently high and they are worried about the survivability of their nuclear forces, they might raise the alert level of those forces. This could mean dispatching mobile missiles from their garrisons and flushing SSBNs from their ports. These might be viewed in the United States and elsewhere as nuclear signals even if that is not Beijing’s intent. Perhaps one of the silver linings of China’s nuclear buildup is that it may face fewer pressures to alert and mobilize its nuclear forces out of concerns about their survivability. If that is the case, then it might help reduce the possibility of inadvertently sending potentially escalatory signals.

 

There are several aspects which might discourage Chinese nuclear signaling and recent developments which might encourage such signaling. I’ll start with the aspects that might discourage it. First, most of our frameworks argue that states are more likely to resort to nuclear use when they are facing a conventionally superior adversary. Regardless of whether, in the aggregate, China or the United States possess conventional superiority, China’s military modernization and reforms mean that, in the Western Pacific, the local conventional balance has certainly shifted in China’s favor. This means that China should perceive fewer incentives to nuclear escalation (including, perhaps, nuclear signaling) than it did in the past. Indeed, some Chinese writings worry that nuclear first use in a conflict with the United States would be initiated by Washington in order to overcome Chinese conventional advantages.

 

Second, Chinese strategists, to the extent that they write about Russia’s invasion of Ukraine, appear concerned about nuclear escalation there. This matches traditional Chinese strategic beliefs that nuclear escalation cannot be controlled and that there is no such thing as a “limited nuclear strike” or “limited nuclear wars.”

 

Third, Chinese decisionmakers might believe that their no-first-use (NFU) policy prohibits nuclear signaling. Since its first nuclear test in 1964, Chinese declaratory policy has included a NFU policy under which China pledges not to use nuclear weapons first. There have long been debates within the United States about the sincerity or credibility of this policy. I’ve characterized it as sincere but not credible. China’s NFU policy may not be credible because, as declaratory policy, it could quickly shift in a crisis or conflict (and, indeed, the ongoing buildup has reduced the inherent credibility of the NFU policy by reducing the technical constraints on China’s nuclear forces). However, this may depend on how Chinese strategists interpret NFU and whether they create space for things like nuclear threats including against non-nuclear states backed by nuclear allies (such as Japan) or against what China claims is Chinese territory (Taiwan). There are indications that Chinese strategists have been trying to interpret launch-on-warning as permissible under the country’s NFU policy, indicating that there may be some ambiguity in what exactly it does and does not cover.

 

At the same time, other developments might increase the likelihood of Chinese nuclear signaling. First, China’s ongoing modernization and expansion have eased the technical constraints on its nuclear forces, providing it with many more options for first use than in the past. This inherently enhances the credibility of any nuclear threats Beijing might make and might make them more attractive.

 

Second, the survivability of China’s strategic nuclear forces could decrease the perceived costs of engaging in nuclear signaling compared to in the past.

 

Third, Chinese decisionmakers might view nuclear signaling as more valuable if they believe that it has aided Russia’s military invasion in Ukraine. Many observers believe that Russia’s nuclear signaling over its invasion of Ukraine has limited the extent of U.S. and NATO direct participation in the conflict. Some Chinese scholars have argued that Chinese strategists might learn the value of these kinds of threats to limit U.S. intervention in a crisis or conflict (though, as I say earlier, the limited writings I’ve seen on this suggest that Chinese strategists are actually quite concerned about the escalatory implications of those kinds of signals over Ukraine).

 

Fourth, the domestic political incentives in China for nuclear signaling might be stronger today than they were previously. China’s domestic media appears to have given more attention to the country’s nuclear forces in recent years which could strengthen the view among the public that these weapons are valuable for defending Chinese sovereignty and thereby create public expectations that the weapons would be used in a crisis or conflict, even if “only” for signaling purposes.

 

I’ve recently researched the nuclear signaling dynamics around the 1969 border clashes between China and the Soviet Union. Those clashes erupted over a territorial dispute and are believed to be the only time in history that China placed its nuclear weapons on alert. One of the features of that crisis is that the actors involved, particularly on the Chinese side, seriously misperceived the signaling that was taking place. In particular, Chinese observers eventually came to believe that the Soviets were preparing for a surprise nuclear strike precisely at the time that Moscow was preparing for negotiations to deescalate the crisis.