Professor Minxin Pei on his book “The Sentinel State: Surveillance and the Survival of Dictatorship in China”

Minxin Pei is the Tom and Margot Pritzker ‘72 Professor of Government at Claremont McKenna College. He is also a non-resident senior fellow of the German Marshall Fund of the United States. In 2019 he was the inaugural Library of Congress Chair on U.S.- China Relations. Prior to joining Claremont McKenna College in 2009, he was a senior associate at the Carnegie Endowment for International Peace and served as its director of the China Program from 2003 to 2008. He is an opinion columnist for Bloomberg and the author of From Reform to Revolution: The Demise of Communism in China and the Soviet Union; China’s Trapped Transition: The Limits of Developmental Autocracy; China’s Crony Capitalism: The Dynamics of Regime Decay, and The Sentinel State: Surveillance and the Survival of Dictatorship in China.
Anna Behuniak '26 interviewed Dr. Minxin Pei on March 5, 2024.
Photograph and biography courtesy of Dr. Minxin Pei.

You mention that China’s surveillance system is very young despite its sophistication. Can you explain how repression tactics in China have evolved since the Mao-era and what events have accelerated their development?

The surveillance state is in fact quite old; it is the technological surveillance state that is relatively young. There are several important characteristics of the Maoist era surveillance. First, it used very crude, very violent, and very restrictive tactics. For example, roughly 20 million people, mostly former landlords, rich peasants and capitalists, were put under surveillance. The way they were kept under control was through the deprivation of their civil liberties. These people couldn't move anywhere which was really cruel. Another thing about the Maoist era is there was a lot of political turmoil which meant that stable or steady surveillance was very hard. The Great Leap Forward, for example, caused such an economic depression that they had to let go a lot of police and they did not have enough resources to keep files on people. The Cultural Revolution was also a pivotal moment as the party had to dismantle the spy network. The party in the Maoist era used cruel tactics, but it was also its own worst enemy. The real change came only after the Tiananmen crackdown when the party realized that they had to strengthen their surveillance system with preventive repression and invest in the system like there was no tomorrow. They recruited a lot of spies and police and invented newer tactics. For example, they developed much more sophisticated ways of watching the people in order to identify key targets more efficiently. And then later, they adopted modern technology. 

One of the central tasks of a surveillance state is to handle domestic threats to political stability. In your book, you discuss how China has accomplished this through its Key Populations and Key Individuals programs. How do these programs work?

A key component of any surveillance system is that you have to know who you want to watch. China has two mass surveillance programs. One is called Key Population which is maintained by the police. It is a very administratively labor-intensive program. There are procedures the police must follow like regularly checking on people in the program, which is very labor-intensive. People placed on this particular blacklist are mostly ex-convicts that are viewed as potential threats to public security, as well as drug addicts and Falun Gong members. Once somebody is released from prison, that person stays on the list for five years. The program is mostly set up for public safety as only about 3% of the individuals in the program are categorized as political threats. In terms of the number of people, my estimate is about 35 people per 10,000 which for China is roughly 5 million people. According to some public sources, I found that the police don't like the program very much because it is so labor intensive. Police need to check on a lot of people and fill in a lot of paperwork. Also, because this program is based on hukou, the fact that people move around so much makes it administratively difficult to keep track of people. 

The other program, called Key Individuals, is not maintained by the police but by local authorities. While you can find official documents on the Key Populations system from the Ministry of Public Security that detail how that system works, I have never seen a government document explaining the Key Individuals program. However, we have enough evidence to show the following. First, designation is most likely done by local officials who are quite familiar with an individual. Second, data for these individuals are collected by the local political-legal committee and there is a national database. If you are designated as a Key Individual then all your information, especially your national ID card, your age and gender, gets entered in the national database. Police in any part of China can access this data and the high-tech system will capture your movement and alert local police that you are in their area. Key Individuals include a lot of political targets, unlike Key Populations. Although we do not have enough information to break down this population, local yearbooks tell us what these people are. For example, if you're a frequent petitioner then you are more likely to be a Key Individual. Others include ex-PLA soldiers and Falun Gong people as well as other underground religious practitioners. Because it is a much more political category, I suspect the number of people considered as political targets is much higher in this category. Based on local data, this category is bigger than the key populations by about 15% to 50% more. If you put the two categories together, we're talking about roughly 8 to 12 million people in China on both lists.

How does China utilize informants as a tool for political repression? How productive are these networks in terms of quality and quantity of intelligence?

Because China is a huge country of 1.4 billion people you cannot give the task of surveillance to only the police because you would have to hire a gigantic number of them, which is not economically viable. Like other dictatorships, China has informants, however, because the Chinese Communist Party is everywhere in China it has the capability of recruiting a very large number of informants. These people are given a variety of tasks such as being assigned to watch Key Individuals. Over the course of my research, I have found that many KI’s will be assigned informants who tell local authorities what that particular person is up to. The system also covers several subtypes of informants. The police recruit what they call special intelligence personnel or teqing whom they use to infiltrate suspected criminal or dissident groups. They will also station these informants in important public venues, such as hotels, conferences, conference centers, railway stations and shopping centers just to watch people coming and going. They particularly like taxi drivers, security guards and sanitation workers because they are always watching other people and they can use them as an extra pair of eyes at venues they are monitoring. The other more generic type of people are those that regularly report to local authorities and police what they've been hearing. 

On paper, China has as many as 15 million informants, which is about 1% of the population. However, it is hard to gauge how productive they are. Based on some local reports, it seems that 60% of them do not actually report any information which means that probably half of China's informants exist on paper only. But this does not show a failure of the program because the very awareness that there are so many informants lurking in society will deter people. People don’t trust their colleagues or their neighbors which means it's much more difficult to organize. Additionally, we do know one measurement of quality – out of all the information informants produce, only about a quarter gets reported, which means probably 75% is not deemed as very useful.

In your book, you emphasize characteristics of China’s surveillance system that differ from those of past dictatorships, such as its “multilayered structure.” Can you explain these differences and why they have been so important to the CCP’s sustained power?

In practically all other dictatorships, surveillance is given just to the police and mostly to the secret police – the Stasi in East Germany, the KGB in the former Soviet Union and the SAVAK under the Shah in Iran. China, however, does not follow that model. Instead, it has a multi-layered system in which China distributes tasks among a variety of both police agencies and non-police agencies. For example, while the Ministry of State Security performs domestic spying in China, it does not watch people who have no foreign connections. This division is the agency that comes closest to the KGB, but it is still different since it watches only certain groups of people. Then, there is China's domestic secret police which is part of the public security apparatus and has about 60,000 to 100,000 people, according to my estimate. This agency watches a lot of dissidents or underground church members who do not have foreign connections. Then, there is also community police that watch less high priority people. You can see there is a division of labor even among police as there are three agencies performing the same task of surveillance. 

The other interesting thing about the Chinese multi-layered system is that universities, state-owned enterprises and neighborhood committees all perform surveillance functions. Because of this, the Chinese multi-layered system is much bigger and covers a lot of blind spots that the KGB or Stasi couldn’t cover. China also saves money by employing a lot fewer expensive law enforcement official as secret police personnel are very expensive. Politically, this system also prevents the emergence of a powerful secret police agency. If you distribute the tasks among a very large number of actors, then you don't need to form a very large and powerful secret police agency; that is the Chinese secret. 

Your book offers new insight into the hidden system of China’s oppressive surveillance state – a topic that is very difficult to research. What were the biggest challenges in writing this book?

The biggest challenge is to get hold of the sources. Most materials related to Chinese law enforcement are classified or internal so they are very difficult to find in the public domain. However, while China may be a very secretive society, its government sometimes leaks so you can find these materials in local yearbooks, or local police gazettes which are basically police histories. In the 1990s, and the early 2000s, the Chinese government was a lot more open than it is today, so it released a lot of sensitive materials. I was able to go to Hong Kong, where there is a library that contains a wealth of local yearbooks and local police gazettes. The amount of information I was able to get is enough to paint a basic picture but there are still a lot of gaps. I just hope this book offers a framework or some kind of roadmap for future scholars to follow in order to find out more about how this system works.

In the future, what will be the biggest threats to the performance and survival of China’s surveillance state?

There are at least two. One is that it deals with a particular context. The system only works well when the overall situation is relatively stable. Out of this context, the system does not work well. The people on the blacklist make up only 1% of the population so if China becomes less politically stable then it might have to watch 2%, which would actually be very hard. The system itself could be challenged by the general changes or developments in Chinese society and economy. The other thing about the surveillance system is that it is quite expensive. People don't realize how expensive it is. The technology part is very costly as there is constant maintenance and updating when things break down. People also don't realize that 85% of all law enforcement budget is covered by local governments and local governments do not have a lot of money. Going forward, if the economy does not do well, then we can expect some kind of erosion of surveillance capabilities.

Anna Behuniak'26Student Journalist

Share this:

Leave a Reply

Your email address will not be published. Required fields are marked *