top of page
  • Admin

“Tremendous Uncertainty”: David Logan on China’s Nuclear Capabilities and China-US Nuclear Dynamics

Dr. David Logan speaks to CACR about China’s improving nuclear forces, no-first-use declaratory policy, and concerns about US capabilities. This interview has been edited for length and clarity. All comments are David Logan’s personal opinions and do not necessarily represent the views of the Department of the Navy or the Department of Defense. This interview is edited for clarity and length. (March 7, 2023)

David C. Logan is an Assistant Professor in National Security Affairs at the U.S. Naval War College. His research focuses on nuclear weapons, arms control, deterrence, and the U.S.-China security relationship. He received his Ph.D. from Princeton University and has served as a Fellow with the Princeton Center for International Security Studies and a Stanton Nuclear Security Fellow with the MIT Security Studies Program. The views expressed are his personal views and not those of the U.S. Naval War College.


I want to start with something you said in 2017 in a piece for War on the Rocks. You were arguing, I think very cogently, against overhyping or overstating China’s development of its nuclear forces, and you said, “China’s nuclear evolution appears to be driven by a desire to maintain a secure second-strike capability in the face of advancing U.S. capabilities, which Beijing believes might threaten its nuclear deterrent.” However, a few years later, in a 2022 piece for the Pacific Forum, you made a very different point. You said, “Recent changes to [China’s] nuclear forces might be an attempt to maintain a survivable second strike capability in the face of expanding US capabilities or might signal a significant departure from China’s historically restrained approach to nuclear weapons and the adoption of a substantially more assertive nuclear strategy.” Can you tell us what the changes have been in China’s development of its nuclear forces over the past few years that have made it a little bit less clear that this is purely about maintaining a second-strike capability?

That’s a great question. I’ll start off by saying that there have been significant changes. Part of the divergence in those two publications is just a reflection of new information that came out in the past five years. But part of it is also a different emphasis in the arguments.

To a large extent, I think the arguments that I made in a more recent War on the Rocks piece in 2020 still hold. I was essentially trying to push back against three myths: that China had this massive hidden nuclear stockpile, that the no first use policy was a sham– and absolutely there is nuance there and there are risks even if it is not a “sham”– and that China had this extensive stockpile of what we could consider tactical nuclear weapons. The third point is something that there are no hard and fast rules about, because clearly China has deployed systems with theater ranges that probably are envisioned for theater operations, but they still don’t constitute the kind of short-range low-yield weapon systems that the US traditionally views as tactical nuclear weapons.

That being said, what characterizes changes in China’s nuclear forces not just in the past five years but over the past decade or more falls into four categories. One is the number. In the early 1980s, which is when China first started to deploy missiles that could hit the continental United States, China only had maybe a few dozen missiles that could target the US, and that was still largely true up until well until the 2000s. By contrast, today, the latest DOD report estimates that China has 300 ICBMS deployed and that the operational warhead stockpile is around 400– the previous year’s report was only 100 ICBMS. I think there’s probably a little bit of uncertainty there– but it goes without saying that the number of deployed nuclear-capable missile systems, especially those with intercontinental range, and the number of warheads has grown in recent years, and I think the most prominent example of that is the silo fields that have been discovered, though we still don’t know how those will be operated.

The second category is that we’ve also seen greater diversity in the force. Originally, China’s nuclear forces were a few liquid-fueled intermediate-range intercontinental range ballistic missiles and a SSBN that was of terrible quality and never performed an operational patrol. Today, China has arguably all three legs of the triad, to varying degrees of quality and quantity, and even within each leg, we’re seeing diversity in the kinds of delivery systems. China has road-mobile ICBMs, silo ICBMS, and the theater nuclear force; there’s work taking place on a next-generation long-range strategic bomber with nuclear capabilities; and they have the sea-based force.

The third is that we’ve seen a significant increase in the quality of those forces. The move from liquid to solid fuel has improved the responsiveness of those forces, because when you have liquid-fueled missiles, you can’t really store them– the fuel is very dangerous and the fumes can sometimes even be corrosive to other components in the missile system. But solid-fueled missiles are mobile, they have quicker reaction times, and some of the most recent systems like the DF-31 AG and the DF-41 are believed to have off-road capability. Off-road capability is important because it broadens the operational area that China’s missiles can exist in, and it also makes it so that those missiles don’t have to rely on pre-surveyed launch sites. In the past, some of China’s previous generation of road-mobile systems could be identified even through commercial satellite imagery of the pre-surveyed launch sites, because they had to make sure that the ground was level, that there was enough room to operate, and that the missile had reliable infrastructure to get there. Today these offroad systems give China more options to move and to hide the systems, which is important.

The last category of the four is hints of changes in operational practices. There have been some suggestions that China may be moving to a launch on warning posture, to improving the reaction times of Rocket Force brigades, and to potentially maintaining a higher state of day-to-day readiness.

Let’s talk about China’s no first use policy. Looking at the Soviet example, my understanding is that the Soviet Union had a no-first-use policy that was belied by documents that came to light after the dissolution of the Soviet Union planning for contingencies in Europe. Is that true, and does that suggest that even though China’s no first use policy is also expressed clearly and without loopholes in official policy documents, it is reasonable to suspect that that might not be the whole story?

I’m not an expert on the Soviet Union, but my understanding of the Soviet no first use declaratory policy first is that it was hinted at first in 1977 in a public speech by Leonid Brezhnev and then more formally introduced in 1982 at the UN, and so we’re really talking about the last decade or so of the Soviet Union and the Cold War. My understanding of that policy is that yes, Soviet war plans did absolutely envision first use of nuclear weapons, contrary to the declaratory policy. But in my reading, every example that I can find of Soviet war plans that envisioned first use of nuclear weapons did so under the assumption that the Soviets had highly credible information that NATO was prepping for its own first use of nuclear weapons, so it always was envisioned under the condition that they believed that nuclear use was imminent and inevitable and that they simply wanted to pre-empt it. I think it’s important that even in the Soviet example, we don’t have evidence that they envisioned using nuclear weapons first when there wasn’t evidence of potential NATO first use, as far as I’m aware.

I actually think that in some ways the Soviet case could be a good comparison to the Chinese case, especially today. Though I think it’s really important to distinguish between what I would call the “sincerity” of the policy and the “credibility” of the policy. I think those are two very different things. I think that up until recently, Chinese officials have been sincere about the no first use policy, but that doesn't mean that it would hold in all circumstances, particularly in a conflict.

So how can we say that it’s sincere? We don’t just have to look at the public-facing documents, but we can actually look at other, more credible indicators. One is documents from the PLA that are not created for public consumption, such as curricular materials, training manuals that are produced and used at places like the National Defense University and the Academy of Military Sciences in Beijing, and other materials that within the Chinese system are considered classified, outlining how the missile forces would operate. In all of those documents we consistently see that no first use principles are still being applied. The second is reports of PLA training and exercises in which they are operating in a simulated post-nuclear environment after China has suffered a nuclear strike. A third indicator is that the operational features of China’s nuclear forces have not really been configured to support first use, so warheads and missiles are stored separately, launchers are stored separately, and we have evidence of extremely tight political control. Historically, this has not been a force that’s well postured for first use. The last indicator is that the force is so small and unsophisticated that there’s limits to what you can accomplish with a first use threat. So I do think we have many indicators– it’s not simply saying, oh, Chinese officials say they have a no first use policy, therefore they have a no first use policy. We have other indicators that have more credibility because they’re tightly coupled to the operational practices, so propaganda or misinformation signaling logic just doesn’t apply to the same degree.

Now, some of those constraints are changing, so all of that does not mean that China’s no first use policy is inviolable. We do have hints that it may be relaxed under certain conditions. There have been PLA writings as old as 2004 that have talked about lowering the nuclear threshold under certain instances, such as large-scale strategic attacks on important population centers or bombing nuclear reactors or really important infrastructure projects such as the Three Gorges Dam. Now, even in those instances, it appears that what “lowering the nuclear threshold” means is not using nuclear weapons but threatening to use nuclear weapons and creating an environment in which they’re trying to deter an adversary with those threats, but that still suggests at least some bounds in which moving towards use is envisioned. There’s also some more recent evidence about Chinese missile brigades, including nuclear ones, conducting what is called combat readiness duty in which they may move to higher levels of alert, and this could tie into a launch on warning posture, which could be an attempt to pre-empt an incoming attack against China’s nuclear forces. We haven’t seen any detailed public information about this. The information about this comes I think largely from a DOD report, and I have seen at least one report in PLA media of an exercise in which a rocket force brigade equipped with nuclear-capable missiles clearly fired under conditions of an expected incoming attack but before the attack actually struck. So there is some evidence of a potential loosening of the no first use policy.

Even aside from that, we shouldn’t believe that the policy is completely inviolable. No declaratory policy by any state, let alone China, is completely inviolable. These policies can change, especially in the midst of a crisis or conflict, so I do think it would be very dangerous for American military planners to proceed under the assumption that China would never use nuclear weapons first. Especially because that opens the door for potential US military operations that could be escalatory and would be dangerous precisely because we don’t know how China might respond in those situations. And I don’t even think we have to rely on this narrative about deceit– I don't even think Chinese officials know what they would do in those circumstances.

Do you think fear of nuclear use could be a constraining factor in a Taiwan conflict? More generally, what role do you think nuclear signaling or the threat of nuclear use could play in a future conflict between China and the United States?

We could envision nuclear threats being used by China to try and deter US allies and partners in the region from supporting a US military operation, and really we’re mostly talking about Japan in that instance. So maybe some of these systems in theory could be used to more credibly threaten Japan with a nuclear strike. Now, that would violate China’s declaratory no first use policy, not just in the ways that we talked about before, but also because China explicitly says that it will neither use nor threaten to use nuclear weapons against non-nuclear states. But then that raises the question of whether those promises apply to a non-nuclear state that is hosting military forces from a nuclear state that are clearly engaged in combat against your own military. I also haven’t seen any credible evidence that the PLA is thinking about using nuclear weapons in that way, but we’re sort of envisioning possibilities.

So I think the ways that nuclear use or nuclear risks or threats are more likely to play a role are the following. One is this idea of the stability-instability paradox, which is this concept that as the strategic level becomes more and more stable, if China has more confidence in the survivability and capability of its nuclear forces, then it may be more emboldened at lower levels of violence, in other words at the conventional level. I think whether or not China is engaging in modernization and expansion for those purposes, China having more room to operate in the conventional domain is still a possible byproduct of a more secure nuclear force. The second is this concern about entanglement. Folks have written about the extent to which China’s conventional and nuclear forces are entangled. The Rocket Force operates both conventional and nuclear systems and systems that are supposedly “hot swappable,” like the DF-26 intermediate-range ballistic missile, meaning you could switch out a conventional or a nuclear warhead. There are concerns that in the fog of war in a crisis or a conflict, this kind of entanglement could generate escalation risks in a couple of different ways. So maybe you have strikes that are intending to target only Chinese conventional missiles inadvertently hit a Chinese nuclear brigade, or maybe strikes degrade overlapping command and control infrastructure that is relevant for both nuclear and conventional brigades. From the Chinese perspective, if they become more insecure about the survivability of their nuclear forces, this may generate “use or lose” pressures and encourage them to at least threaten nuclear use. From the US side, you have this potential problem of what’s been called “warhead ambiguity.” If you have entangled missiles and dual-capable missiles in the PLA Rocket Force, it may not be possible for the US to identify whether an incoming strike is conventional or nuclear, and in theory this could cause some American officials to assume that it’s nuclear.

From research looking at the 1969 border crisis between the Soviet Union and China, we saw similar dynamics where we had vague and perhaps even unintentional nuclear threats from the Soviets that were perceived as absolutely credible by a very insecure, anxious, and alarmist Chinese government. It’s not hard to see that maybe the PLA could learn, either from Soviet signaling in 1969 or Russian signaling in 2022 in Ukraine, that nuclear threats, even if they’re vague and ambiguous, work, and make it more likely for those to be used in the future.

Regarding the security-insecurity paradox, I was alarmed to read in your 2022 Pacific Forum article that there were studies that indicate that China’s second-strike capability may not be secure. If that’s true, I think that that poses a lot of serious questions about the stability of a China-US conflict.

I think whether China has a survivable second strike depends on whom you ask, what you mean by survivable, and your assumptions. If you ask me, China has a survivable second strike. But the most sophisticated open-source nuclear exchange model, published in 2020 by Wu Riqiang, determined that at day-to-day alert, which is a very low level for China, there was only a 10% chance that three warheads from China would survive and penetrate US ballistic missile defense capabilities, while even at maximum alert status, that percentage for three warheads would only rise to about 50%. Now, there’s a lot of uncertainty– not only inherent uncertainty about the physical capabilities of nuclear forces and ballistic missile defenses and their operational practices, but also an information asymmetry. The US has tremendous uncertainty in the effectiveness of its ballistic missile defense capabilities, but it certainly knows a lot more than China does, so I think you often see Chinese analysts adopt their worst-case scenario assumption, which is that US BMD capabilities are targeted towards China and they’re highly effective. And then the reverse is probably true for US analysts, who are aware of the severe limitations of US BMD capabilities.

You saw an increase in Chinese concerns about survivability really jump in the early-to-mid 2000s with the US withdrawal from the ABM treaty. There have been other developments since then, including concerns about advanced ISR capabilities and US conventional strike systems, which could, in theory, make it easier for the US to target Chinese nuclear forces without having to revert to nuclear first use by the United States, but I think China has a survivable second strike capability. This is something that is reflected in US policy documents. The last couple of nuclear posture reviews have said that it is a goal of US policy to essentially maintain strategic stability with China, which to me indicates a desire not to undermine China’s nuclear capability. Although there are sometimes debates about whether the US should try to pursue a damage limitation strategy against China. That’s a whole other issue.

How important are China’s advances in hypersonic missile technology for the nuclear dynamics between the United States and China?

I think hypersonics could matter at the conventional level, but I guess I’m skeptical that they represent a revolution at the strategic nuclear level. I think the technical characteristics of hypersonic weapons have been a bit overhyped. I’m not a technical expert, but there’s been work on that recently showing that traditional ICBMs already possess many of the supposed advantages of new hypersonic systems. Also, from a strategic interaction standpoint, it’s not clear to me that hypersonic weapons systems provide a meaningful capability to China that it doesn’t already have.

I think China has a survivable second strike, so the additional potential ability to launch a weapon over the South Pole or to have it come in and evade or have a lower radar cross section doesn’t really matter. If they can already strike the continental US with a nuclear weapon, I don’t really care how the missile arrives.

If the United States ever did negate China’s second-strike capability, would that be a bad thing? Would it create some new dangers?

I think that when it comes to nuclear dynamics between the United States and China, the US needs to prioritize diminishing arms race incentives and maintaining crisis stability. I think that has been reflected in US policy in the desire to maintain strategic stability with China. I don’t think that from a technical perspective it’s possible [to negate its second-strike capability], especially given China’s resources today and defense industrial base. It would be even more challenging to build towards primacy or damage limitation today than it would have been in the past.

There are risks from what’s occurring now. If China were to officially move to a launch on warning strategy, to adopt a higher alert status among its nuclear forces, to continue to build out the force, all of these things increase the risk of nuclear use, increase uncertainty, and increase competitive pressures between the US and China. They also erode non-proliferation norms. China is in violation of its disarmament commitments under the NPT, especially by increasing the quantity of its nuclear forces. But I don’t think it would be feasible or beneficial for the US to attempt to achieve nuclear primacy.

Do you have an impression of what China-US nuclear competition is going to look like in the future? Do you see it evolving in parallel with the Cold War, or is it too soon to tell?

It’s very hard to tell, in part because I think there’s still tremendous uncertainty about China’s goals in its nuclear modernization and expansion, and goals not only from an endpoint standpoint but also from a drivers standpoint.

I do think that competition in the nuclear domain is going to be driven in large part by overall strategic competition and the larger bilateral relationship, and Chinese strategists often conceive of strategic stability as not just narrow technical considerations but within the broader view of relations between Washington and Beijing. So I think if the US and China can better manage their relations in other domains, that will tamp down on nuclear competition.

Originally, I think China’s nuclear modernization and expansion was focused on maintaining a technically survivable deterrent. I think that is still a major driver, but I think in addition there’s growing evidence that China is investing in its nuclear forces for reasons of status and prestige. We see that from senior leaders in the Rocket Force, we see that from senior party leaders, we see it in how the PLA media and broader Chinese media talk about developments in the nuclear domain connecting technological breakthroughs to national status. There’s also been this interesting development that Leveringhaus over at King’s College London has done some work on about how Chinese media and officials increasingly valorize China’s strategic missile and nuclear program histories. That’s another way that overall competition and the relationship feed into the nuclear domain.

You also see evidence of [a mentality of] nuclear power as an indicator of comprehensive national power and that China may need a larger nuclear force, almost so that the US in particular but also the world more generally takes China seriously. PLA and some civilian writings say that the gap in nuclear capabilities between the US and China is so big that it tempts US strategists to try to pursue damage limitation or nuclear primacy, and so China needs to invest in its nuclear forces not only to maintain a survivable second strike, which it may already have, even in the views of these strategists, but simply so that they close the gap enough that the US isn’t tempted by these fantasies of negating China’s nuclear forces. So it’s eliminating even the perceived opportunity from the US standpoint. In short, I think it’s in part going to be driven by the quality and status of the relationship.

Is there anything else that you think China hands who are not nuclear experts should know about China’s nuclear forces or the nuclear competition between the United States and China?

I would emphasize the tremendous uncertainty on this issue. The last DOD report said that if China’s modernization continues on its current trajectory, by 2035, it will have an operational nuclear warhead stockpile of 1,500. Without a doubt, the warhead stockpile is going to increase. The delivery vehicles will increase and become more diverse. But we don’t know how many there will be, how they will be operated, or exactly what the drivers are. There’s been Chinese resistance to track-one discussions on nuclear issues, but I do think that conditions in some ways may be better for that– not from a political standpoint, but just because if China is more confident in their survivability of its nuclear forces, that potentially makes it easier for them to come to the table. Previously, when you had such a small, unsophisticated, and vulnerable nuclear force, opacity was a deliberate strategy to maintain survivability.

China also has other “investment needs,” whether it’s military spending or domestic investment, so I would caution against automatically assuming worst-case futures. They might come true, and we certainly need to envision them and we certainly need to prepare for them. But worst case doesn’t mean most likely, and the policy choices that the US makes can have a significant impact on which of those scenarios develops.


bottom of page