---

Sheena Chestnut Greitens is an associate professor at the Lyndon B. Johnson School of Public Affairs at the University of Texas at Austin, where she also directs the Asia Policy Program. Her research focuses on national security, East Asia, and authoritarian politics. She is the author of Dictators & Their Secret Police: Coercive Institutions & State Violence (2016) and Politics of the North Korean Diaspora. She is currently working on a book entitled, Preventive Repression: Internal Security and Chinese Grand Strategy under Xi Jinping.

Q: How does China view national security?

A: There’s no question that national security has been a defining feature of Xi Jinping’s leadership and of the way that he has governed China. But what’s interesting about the use of this term “national security” — which you can also fairly translate as “state security” — is that when Xi has applied that concept, he means something quite different than the way that the United States approaches and organizes for national security. 

For Xi, the centerpiece of his approach is the comprehensive national security concept, which he announced in April 2014, and then developed into China’s first national security strategy, which the Politburo approved in early 2015. What’s interesting is that there’s a very strong emphasis on internal security. The use of the term “national” security primed American analysts to think that the Central National Security Commission that Xi was establishing was going to be a corollary or counterpart to the U.S. National Security Council. And that just hasn’t been the case. The center of gravity of national security work for Xi is much more explicitly internal. 

Chinese sources talk about how the foundation of national security is political security. And political security means defense of China’s socialist system, of the leadership of the Communist Party, and of Xi Jinping himself at the core of that leadership. And so this focus on internal, non-traditional, emergent security threats is fundamental to the way that the party-state thinks about security. It is fundamentally a regime security concept, which makes it distinctive. It’s important to understand that, because it has a comprehensive effect on Chinese internal security policies, but also increasingly is affecting China’s foreign and external security policies as well.

Is this new?

There’s a long history of thinking about external and internal security threats as interconnected. The old saying “nei luan wai huan” can be translated as “internal disorder, external disaster,” with the implication that the former invites the latter. And so that concept has a very long history. 

What’s different about what Xi did is that he took this concept — which was pretty vague when he first announced it — and used it as an intellectual architecture and blueprint to guide organizational reforms and a legal overhaul of China’s security laws, which has had both domestic and international implications. That concept has also prompted internal security policy changes in, for example, Xinjiang, and also has produced a marked increase in the international activity of China’s police forces. What he did was take this concept and make it into a systematic framework — one that has been a pretty good guide, in my view, to understanding Chinese domestic and foreign security policy since it was announced. 

You can look at the comprehensive national security concept and view it as a grand strategy for China. That’s the argument of the book I’m writing now. When Western writers and historians talk about grand strategy in the Anglo-American tradition, they tend to think about external security and primarily (if not exclusively) about military power. Whereas for China, grand strategy is fundamentally a theory about how to make the party-state and its leadership secure — and the answer to that question has both internal and external dimensions that are very closely connected.

How much of this is Xi’s own focus on national security as opposed to the Chinese political system’s broader focus on it?

It’s a bit of both. Xi seems to have a particular worldview about what threatens China and about what threatens the Communist Party — in particular, one that seems to be linked to his diagnosis of what doomed the Communist Party of the Soviet Union. There’s a decades-long precedent of thinking of the Soviet Union as a potential example or a cautionary tale for China; there was a Cold War-era phrase, ‘the Soviet Union’s today is our tomorrow,’ which initially was very positive; and then as the Soviet Union changed, first with the Sino-Soviet split, and then with the tumult of 1989 and the collapse of communism in Central and Eastern Europe, that phrase took on a much more worrisome cast for the Chinese leadership. 

One of the ways this matters is in anti-corruption. There is a book about the fall of the Soviet Union called Stealing the State, where the author [Steven L. Solnick] writes that Soviet communism wasn’t abandoned from the top or overthrown from below, but hollowed out from within. If you look at some of Xi’s rhetoric about anti-corruption and national security, he appears to believe that corruption risks hollowing out the party’s grassroots foundation and the foundation of social control. And so there is a securitization element to the anti-corruption campaign that he’s pursued. 

A second area is ideology. Things like Document Number Nine [a leaked internal document from 2013 enumerating risks to the CCP] highlight the fear that the threats to the Chinese party-state, and to the political security of China, are not just threats of physical harm. It’s actually a threat of ideological contamination and ideological sickness that will weaken the Chinese body politic. This is not just about education or propaganda; there’s a very concrete set of organizational changes that have operationalized this concept. And again, the threat has both internal and external dimensions. 

The third piece is about whether the Soviet Union maintained adequate control over its coercive apparatus, both the internal security forces and the military. So under Xi, there was a reorganization of the military, including reorganization and moving of the People’s Armed Police, and then the creation of national security commissions below the [Central] National Security Commission; these now exist down to the local level, to ensure that national security is a priority, not just at the top, but all the way down through the local leadership. 

This approach to national security also has foreign policy implications. For example, what do you think of the recent reporting about the overseas police stations?

I do think that the increased activity of the Chinese Ministry of Public Security as a global actor is important. A narrow focus on the PLA [People’s Liberation Army] is not sufficient for understanding all the different facets of China’s global security presence. One reason it matters is that Beijing’s offers of police and domestic security assistance seem designed to make China into an internal security partner of choice. That’s particularly appealing to countries that might not want internal security assistance to come with human rights conditions or democratic accountability requirements that Western aid programs often demand. So China’s role in this space could have a negative effect on democratic consolidation and democratic stability worldwide. 

But the other thing is that China’s political-legal apparatus doesn’t make a clear distinction between the work of everyday law enforcement for criminal policing purposes and the political policing that assures the CCP’s hold on power. So the projection of the Ministry of Public Security activity abroad, when the system doesn’t maintain that distinction internally, does raise the risk that this recent phenomenon of transnational repression is here to stay. That’s something that governments around the world are increasingly going to have to grapple with. The United States and the international community are overdue in just getting a comprehensive picture of what the actual activities that are going on are. 

I don’t think the reporting on the police stations, specifically, has always been sufficiently clear in terms of what’s happening. That’s not unusual in new developments in global politics: we don’t have the clearest picture at first. But we’re overdue in getting a clear picture of what is going on where, so that we can identify precisely what the risks of that activity are, why we care about them, what tools are available to deal with them, and to ensure that citizens and countries are appropriately protected where they need to be from that kind of activity. 

… I don’t want to overstate either the omniscience or the omnipotence of the security state, and it’s easy to do that. This is still a system that has a lot of fragmentation.

So given all of these changes and all of these concepts that fit within comprehensive national security, is there anything that can’t be viewed through the lens of national security in China today?

Well, almost anything can, because the concept is very broad. But I should be careful here, because I don’t want to overstate either the omniscience or the omnipotence of the security state, and it’s easy to do that. This is still a system that has a lot of fragmentation. It has a lot of principal-agent problems in terms of managing its own personnel. And it doesn’t work perfectly or omnisciently. 

I’ve done a lot of work on China’s use of surveillance, both domestically and overseas. Starting around 2012, local and provincial budgets for domestic security spending increased substantially. Around that time we saw reports that Chinese spending on domestic security was exceeding spending on the military and national defense. Much of that increase in local spending was for the procurement of surveillance technology and cameras, and there’s been some really excellent reporting about the creation and buildup of China’s surveillance state. But we need to put that in the right political context: it goes back to this idea that the comprehensive national security concept is hyper-preventive. Even the “stability maintenance” framework of the Hu Jintao-Wen Jiabao era have been implicitly (or occasionally explicitly) deemed to be too reactive, or not sufficiently preventive. 

Xi often uses this phrase fangkong, “prevention and control.” In 2015, there was a directive issued by the State Council and the CCP Central Committee that talked about constructing a “multi-dimensional, information-based prevention and control system for public and social security,” which is basically a jargon-y way of saying, let’s create a surveillance state that can do this preventive work. 

It’s easy to focus on the aim, and on what the technology can do, and we need to talk about that. But in looking at how this worked on the local level, I found that the ability to collect information initially outstripped the back-end capability to combine different sources of data, to coordinate it, and then actually use it effectively in governance. That was both a technical problem, in terms of how you take different streams of information and put them together, and it was a bureaucratic politics problem, because the local state remains pretty fragmented, with lots of different actors who are assigned roles for social stability and security who don’t necessarily all play well together, or share information well. And lastly, there was a human capital issue; you have to know what to tell the system to look for to use the information effectively for governance. The ability to collect data at least temporarily outstripped authorities’ ability to combine and actually harvest it for governance. So around 2017, you see local and senior officials talking about this problem of ‘information islands’, which was more or less saying, we’ve got all this information, but it’s not being put together and used effectively. The party-state was aware of this, and was putting effort into solving it, but that also meant that it was still an issue that hadn’t really been resolved. 

Inside the Huaxian Traffic Monitoring and Command Center in Huaxian, Henan. August 15, 2018.Credit: Huaxian Traffic Police

It’s easy to view the system as very powerful; it is. But in some ways, that is exactly the purpose. Believing that the surveillance state is all-powerful actually helps the party-state accomplish its goal, because if citizens believe that the surveillance state is omniscient, and that things are being monitored and responded to very quickly, that has a real deterrent effect on mobilization and collective action. There’s an interesting question about whether these analyses that emphasize how powerful the surveillance state is, actually help the party-state accomplish the goal of the surveillance state itself, because they’re sort of augmenting that deterrent effect that the party wants the system to have. 

What did you learn about the state’s ability to maintain control from protests in November?

If I was a member of the Communist Party leadership — which thankfully I am not! — I would be concerned about the emergence of these protests because we’re 10 years into development of the surveillance state I just described, and 10 years into this hyper-preventive security concept being a huge emphasis and priority for officials. So to see a set of protests emerging across different locations in China that share common information, and common grievances and complaints — to see that escalate beyond a local policy implementation question, which is where most protests often stay — and to have multiple, simultaneous points of failure in “prevention and control,” would probably be concerning. 

It does look like [during the protests] the system got temporarily overwhelmed, and the authorities temporarily had trouble figuring out how to respond. But the surveillance state — and the security apparatus that employs the surveillance state — pretty quickly caught up. Some of the videos that were shared on social media can have very powerful mobilization effects on the citizen side, and the side of people who are opposing government policies — but then that same information later became useful to local authorities in tracking people down and saying, essentially, “Alright, zero-COVID is gone now, you don’t have anything else to complain about, so stop,” and in some cases punishing and putting pressure on people, particularly those who might have been involved in organizing. I think it shows both the limit of what a system like that can do, but also then the capacity to respond pretty quickly. And it shows the sharing of information to be a double-edged sword: who it benefits can flip pretty quickly, in a short amount of time.

Do you see the subsequent reopening as a response to the protests?

Not solely a response to the protests. I think they were probably a factor. 

There are three longer burning issues. I hesitate to comment on the course of the virus in China, because I am not a public health expert or an MD. But from what I’ve read, the consensus seems to be that COVID was already spreading rapidly, and so there was a sense that this might be already leaving the space where the CCP could really control it. And there are two other factors. Historically, both in Chinese politics and in non-democracies writ large, elite politics are much more threatening to the stability of non-democracies, than ground-level protests or activism. Two-thirds of dictatorships fall to elite insider threats, not to popular revolution. So the fact that Xi Jinping got through the 20th Party Congress and had been able to make the personnel changes that he wanted, lowered his risk at the elite level. And the third factor is the economic costs, and the fact that poor economic performance can also be very destabilizing to non-democracies. 

…the pandemic, combined with the tightening of political control in China, has made the system even more opaque and difficult to scrutinize.

Those three factors, operating on a longer time horizon, had a cumulative impact — and then you had protests and unrest, and the party-state could look responsive by changing the policy. 

It was clearly an abrupt policy change. But in some ways the puzzle is actually why China didn’t loosen the policy earlier and accompany it by a vaccination strategy that could have reduced the cost a bit more. That continues to puzzle me. I don’t have a good answer, but it’s also outside my area of expertise. 

This is one of the really interesting and difficult things about studying Chinese politics and foreign policy at this particular moment: the pandemic, combined with the tightening of political control in China, has made the system even more opaque and difficult to scrutinize. Not impossible by any means, but there’s no question that understanding Chinese decision-making is becoming more challenging, at least for particular questions and research areas. To the extent that power has become more concentrated, and decision-making has become more concentrated in a single person, in Xi Jinping, you leave the realm of political science where I was trained and move more into the realm of psychology. And that’s just a very different way of understanding and predicting decision making, particularly when it depends on an individual that we don’t have a lot of information about, including who he talks to, and how he makes decisions. 

What are the downsides to this? What does the state lose by focusing so much on national security?

First, there’s been a change in the way the party thinks about economic development. Zero-COVID may be a case of that, but this is a broader phenomenon, where the party-state has been willing to accept limits on economic performance in order to keep political control and control over Chinese society. There’s a line that you’ll sometimes see now, that “security is a prerequisite for development.” It’s had some pretty profound impacts, and the costs are especially high in a place like Xinjiang. 

But you also see some concern among Chinese scholars and analysts that securitizing everything isn’t the right way to go, that you risk viewing everything through a single lens, and that leaves the party with real blind spots, and very high costs. A few Chinese sources have raised questions about whether all these problems should be viewed through a security lens. 

To go back to COVID, I would say the CCP treated the management of the coronavirus as a public security problem, from the initial policing, and crackdown on the original whistleblowers or reporters in Wuhan, Dr. Li and others, to the broader infrastructure of ‘prevention and control’ (that phrase is used in public security, but it’s also used a lot in public health). Treating that as a social control problem, and focusing on quarantine, or lockdown, that set of policies, had a real cost — instead of focusing on things like mass vaccination campaigns that were particularly concentrated among high-risk populations, like the elderly, or building up health care and augmenting the basic capacity of the Chinese medical system. To me, that’s probably the easiest and clearest recent example of where there was an alternative way to approach this problem, but the use of the security framework predisposed the party towards some choices and some ways of managing the pandemic and not others. And those really did have costs, economically and socially. 

Can you describe your research on Xinjiang?

The basic phenomenon that I wanted to understand was that it seemed like there was a sharp escalation in this campaign of collective repression against the Uyghur Muslim population in spring of 2017. That escalation followed a meeting of the Central National Security Commission — so it is linked, process-wise, to the comprehensive national security concept. And you also see this use of the language of ‘prevention and control,’ which in Xinjiang takes on a sort of extended medical metaphor. Political threats are described as viruses and tumors, and there’s rhetoric about the need to immunize people in Xinjiang against politically problematic thinking and behavior. 

This is also a time at which Xi Jinping has been telling officials, look, we have to be really attuned to the interconnectedness between external and internal security threats, and the risk that external developments will destabilize China at home. So one way to understand the escalation that we see in Xinjiang under the comprehensive national security concept, is the system’s extreme sensitivity to the potential for changes abroad to have profound destabilizing effects at home. Chinese leaders appear to hyper-focus on a fear that small numbers of Uyghurs, who had fled China and gone abroad, through Southeast Asia and into the Middle East and Syria, could be radicalized, could get training and support from abroad, and that that could then diffuse back into China as a more serious security threat. That’s the Xinjiang-specific variant of this internal-external security nexus. And if you combine that with this preventive logic of immunization, which is operationalized as forced re-education and internment, you get the grossly disproportionate violation of human rights that we have all witnessed. 

The comprehensive national security concept gives us a framework to see why there was a change. Xinjiang had been a security problem in the view of the party-state for a long time. The comprehensive national security concept primes officials to think in terms of this, and gives us a useful backdrop for understanding why the response in Xinjiang was so escalatory, so repressive, and so destructive to Uyghur families and the fabric of society there.

Why was there such a focus on the impact of Uyghur contact with overseas groups, why did that end up striking such a nerve?

There’s a long history in China of this fear that external actors will destabilize Communist Party rule at home. That’s driven Beijing’s fear of color revolutions, and of ideological infiltration, which is a big emphasis in Document Number Nine, and it appears in some of the analyses of what happened at the end of the Cold War: foreign actors contributed to the ideological contamination and thus the disintegration of communism. And so, this perceived risk that foreign thinking could cross borders and create an infection in Xinjiang that would metastasize into actual political instability: that’s not a Xinjiang-specific phenomenon or fear. There’s a long history of being concerned about that risk, to a degree and a level that might be hard for outside observers to understand. 

We tend to look at the Chinese state and see it as very powerful. Chinese thinking about that power seems to me to be much more dialectical. Phrases like “great changes not seen in a century,” or “China approaching the center of the world stage” — we have a tendency to read those as almost triumphalist. But if you look at those passages in context, in the original Chinese sources, they’re often paired, in that with increased opportunity comes increased risk and threat and danger. So there’s a darker side to China becoming a more powerful actor that, when we talk about China’s national security framework, we don’t always appreciate. But it’s very much baked into the discourse, this pairing of risks and opportunities. 

Saying that the political system is primed to react to those fears and perceived risks, is not the same as saying that’s a legitimate response. I want to be very clear when I say that explaining that the Chinese political system is primed to emphasize and perceive a certain threat is not the same thing as saying that threat is being accurately perceived. Non-democracies inflate and mis-estimate their perception of threats all the time. 

…there’s a darker side to China becoming a more powerful actor that, when we talk about China’s national security framework, we don’t always appreciate.

So when I ask the question, “why did this happen?” I’m looking at analytically, what the party is seeing. They had a security concept coming down from the leadership, telling them to be vigilant and to worry, and that produces a massively disproportionate reaction that’s had horrific human consequences. 

That is an explanation for why the escalation occurred. But ultimately, it’s going to be unsatisfying to many people, because the reaction does seem so disproportionate and so unjustified. That gap between explaining what led to a certain set of security policies in Xinjiang, between that and having any sense in which those policies are justified — which they’re not — is one of the reasons why this is a really hard question. Because ultimately the answer is not very satisfying, and the stakes for other human beings are so high. 

---