Joseph Torigian
2026-01-16 08:30:00
As this article goes to press, security officials in Iran are carefully weighing their options on how to dole out violence to preserve an authoritarian system. It is an existential moment for any authoritarian regime, and one that Chinese leader Xi Jinping must be watching closely. He strongly believes the People’s Liberation Army must be prepared to defeat any enemies within the elite, among the Chinese population, or overseas. Given the military’s fundamental role in the designs of one of the world’s most powerful men, outside observers have watched carefully as rolling purges continue to hit high-ranking officers.
Suddenly and unexpectedly, a few weeks ago a six-hour recording of the closed trial of Chinese general Xu Qinxian was published online. Xu was commander of the elite 38th Group Army in the early days of the Tiananmen Square protests in 1989 before he was removed for wavering when ordered to enforce martial law.
The video is a stunning revelation for anyone interested in civil-military relations in China. It provides crucial new hints about one of the biggest mysteries surrounding the June 4 crackdown —a watershed moment in modern Chinese history. Xu’s defense raises universal questions about what can be done in a disciplined organization when problematic orders challenge moral and ethical codes.
This is not the first time a major leak has led to a big jump in what we know about decision-making during the June 4 crisis. Various internal speeches, memoirs, and even diaries have become available over the years. Each new revelation gets us closer to what actually happened.
And yet no breakthrough ever provides a complete answer. In the video, Xu denies some of the factual claims made against him in witness testimony. Moreover, it is not easy to characterize Xu’s behavior in legal terms. In some ways, his reaction to martial law was explosive. Yet during the trial he rejected the idea that he “disobeyed” the order, and other actions demonstrated that he was still adhering to military discipline in important ways.
Perhaps these continuing questions are a point of frustration. And yet the ambiguities that emerge from the video tell us a great deal about what Chinese military officers face in the People’s Republic of China.
They show that the Chinese Communist Party was unable to figure out how to balance the “democracy” and “centralism” in “democratic centralism” even shortly after the Cultural Revolution had shown the danger of both extremes. In the People’s Liberation Army, smart, experienced military officers are often unsure about what is permissible and what is not. Xu’s case reveals the extraordinarily complicated political calculus that goes into how and when military officers are punished and why it is so hard for outsiders to figure out what’s happening.
And finally, Xu tells us something about Xi’s greatest fear: hesitation in the military, the party’s final guarantor of power, during an emergency. The ghost of an idea — the “nationalization” of the People’s Liberation Army — has haunted the party for decades. In the 2010s, after Xi came to power, the pro-reform history journal Yanhuang chunqiu guaranteed to the leadership it would not touch eight subjects of particular sensitivity, and this topic was the first one. And now we have a six-hour recording of the trial of the most famous case of a Chinese officer allegedly claiming this principle to stop a massacre. What does the video tell us?
What Did Xu Actually Do?
On May 17, at a meeting at paramount leader Deng Xiaoping’s home, Deng decided to introduce martial law. General Secretary Zhao Ziyang accepted the decision but refused to play the role of enforcer. The next day, Yang Shangkun, a vice chairman of the Central Military Commission and its secretary-general, communicated the martial law order to a group of military leaders. A few hours later, Xu Qinxian, the commander of the 38th Group Army, was summoned to the Beijing Military Region headquarters. The elite 38th Group Army was stationed in nearby Hebei province, but Xu was already in Beijing for treatment of kidney stones.
He left directly from the hospital for the fateful meeting. Upon receiving an order to bring the 38th Group Army to Beijing to execute martial law on May 20, Xu said that he disagreed. In his mind, the protests were a political matter that involved “the masses.” Therefore, political means should be used to resolve the situation. Public security and the Beijing Garrison forces would be enough and the People’s Liberation Army should not be involved. He suggested that the decision should be reviewed at higher levels. Since good and bad people were mixed together among the protesters, Xu warned, if the order was executed poorly they would become historic villains. Most dangerously, he said that he was unwilling to participate.
Nevertheless, he communicated the order to the 38th Group Army by telephone. The following day, his fellow officers convinced him to lead them after all. Xu even issued directives on how to prepare. However, unbeknownst to them at the time, Yang Shangkun had already decided to strip Xu from command the previous night.
What Do We Still Not Know?
Two big puzzles remain. First, according to prosecutors and witness testimony, Xu demanded that the National People’s Congress — the Chinese state legislature that typically serves as a rubber stamp for party decisions — review the order because the People’s Liberation Army obeyed the National People’s Congress, not the party. Xu also faced the accusation of raising questions about the format of the order. At the trial, Xu denied those accusations. What are we to make of these discrepancies?
Xu himself admitted that he mentioned the National People’s Congress because the latest constitution had incorporated the armed forces into the state system, which, in his mind, meant the Congress legally could play a role in deliberations about the armed forces. During the trial, it is possible that Xu downplayed what he actually said about the Congress because he wanted to avoid a more serious punishment.
My own hypothesis is that Xu did mention the National People’s Congress but likely, in my judgment, did not go so far as to say that the party did not command the People’s Liberation Army.
In his defense, Xu claimed to have said that the Politburo, State Council, and Central Military Commission should review the decision, and that the National People’s Congress could also play a role given the nature of the Chinese constitution. Although we cannot say for sure, I suspect that to be true. It is hard to imagine a product of the People’s Liberation Army, especially at that level, disagreeing with the idea that the party did not “command the gun.” Xu stated at the trial that even a low-ranking officer with little experience would understand that basic fact. Hopefully, new evidence in the future will provide more context on this crucial question.
The prosecution might have chosen to emphasize the National People’s Congress for two reasons. First, in whatever context, it was dangerous for Xu to mention the Congress at that moment. And second, during the Tiananmen Square protests, the biggest danger was indeed the possibility of the Congress slipping out of party control, as I write about in my book The Party’s Interests Come First: The Life of Xi Zhongxun, Father of Xi Jinping. By the time of the trial, the prosecution would have known about the party’s political objective of accentuating that the Congress did not play an independent role in Chinese politics. In other words, even limited or hedged references to the Congress could have been interpreted as existentially threatening, especially in retrospect.
Did Xu complain about the legality of the order itself? This is perhaps an even bigger puzzle. In subsequent years, Xu would privately suggest that the order was problematic because it was in oral, not written, form. Yet in those conversations Xu also admitted that it was common practice for an oral order to come first and a written one later. So, in a way, both sides were right.
Curiously, in an internal speech, Liu Yazhou, a prominent commissar known for princeling connections and popular writings, said that Xu refused the order because Zhao Ziyang, the first vice chairman of the Central Military Commission, did not sign it along with Deng and Xu. Even more curiously, Yang Shangkun stated on May 24, 1989 that some people in the military indeed believed that all Deng, Zhao, and Yang needed to sign an order for it to be legitimate.
At the trial, Xu accepted that he had said a written order would be better. Yet he provided important context. The problem was not the legitimacy of the order, but because he had a poor memory, so, “if any errors occur in the future, it will be difficult to investigate.” Xu stressed: “I didn’t question or ask why a written order hadn’t been issued for such a significant matter; I didn’t bring that up. I remember the commander at the time even said that in past operations, we could issue verbal orders first, and written orders were often issued later. I said that’s also true, and issuing it later would be fine with me.”
Once again, it is hard to guess what exactly happened. Perhaps at the trial Xu was downplaying what he said. Or, the fact that Xu had raised the issue of a written or oral order in any context was dangerous enough to use as a weapon against him and make a point to other officers.
Did Xu Transgress?
In both later conversations about the incident and at the trial itself, Xu referred to the party document “On Several Principles for Intra-Party Political Life,” adopted in February 1980 in the wake of the Cultural Revolution. According to that document, cadres who disagree with party policy are allowed to maintain their position and express their views within the party. They are obliged to execute policy “except in extremely urgent situations where implementing the decision would immediately cause serious consequences.”
In certain ways, Xu’s behavior accorded with the spirit of this document. He was right that executing the order would “cause serious consequences.” On March 20, when 38th Group Army tried to enter Beijing, it was a disaster. The citizens of the capital rallied to stop them from entering the center of the city. The fact that the 38th Group Army entered Beijing without Xu substantiates another one of his defense tactics: that the soldiers would enforce martial law whether he commanded them or not, as the military served the party and not Xu the individual. Moreover, Xu did agree to execute the order the following day. In other words, his refusal happened during the preparatory phase. The fact that it was conceivably possible he might execute the order after all, despite the tensions of the May 18 meeting, suggests that it was not a decisive showdown. Xu flatly denied that he “refused” the order, saying it did not reach that level at the May 18 meeting.
Yet in other ways it is clear why Xu’s behavior was so dangerous. What if every commander refused to execute the order? What, indeed, is the difference between “not personally wanting to execute an order” and violating an order? The news that something happened with Xu immediately reached the students — a damaging revelation for the party. Military commanders were distracted by his behavior when they were trying to focus on the protests.
What exactly was Xu’s calculus? Perhaps he felt that martial law was such a disastrous decision that the chance his act would cause the leadership to pump the breaks was worth it. He knew how beneficial it was for the party to leave him in command of the 38th Group Army to create a sense of unity among the ranks. Maybe he thought they would try to win him over. He might have feared that he would be blamed for poorly executing an inherently flawed decision anyways. Or he was making a moral choice not purely shaped by consequentialist thinking.
Naturally, we can ask whether his calculus was carefully considered. He was well educated and known for his calmness as a commander. He cut a thoughtful figure at the trial. Yet he had just left the hospital and was in great pain. The martial law order stunned him. The fact that he came around the next day suggests he might have calmed down.
In any case, in the Chinese Communist Party, punishment is only indirectly related to what the accused actually did. It’s also about political messaging. For understandable reasons, Yang Shangkun felt a need to make an example of Xu and preclude the possibility that Xu would waver again at an even more critical moment.
Why Didn’t Xu Do More?
The enduring puzzle of the June 4 crackdown is how Deng engineered such an outcome when so many people within the military and party leadership did not want to use violence. Seven retired generals wrote a letter calling for a peaceful solution. Two retired marshals did the same. Many in the party hoped that an emergency session of the National People’s Congress would find a peaceful resolution. The general secretary himself, Zhao Ziyang, refused to enact martial law.
Yet even the most “brazen” faced powerful ideational and institutional constraints. Zhao accepted Deng’s decision to announce martial law as legitimate even though a majority of the Politburo Standing Committee thought it was a terrible idea. Zhao even said at Deng’s home that it was “better to have a decision.” Zhao tried to resign but was convinced to change his mind.
In the wake of the Cultural Revolution, senior figures did reflect on the dangers of strongman rule. Yet they allowed Deng to dominate. Why?
The answer is that Leninist systems are designed to be organizational weapons. They require power to flow downwards, not upwards. The party needed a core leader, and Deng was that core. Rejecting Deng would mean rejecting the party, which was the source of life’s meaning to this generation. As a revolutionary elder known as a military man, the men with guns saw Deng as a natural leader. Open resistance would only lead to a worse punishment, and it might implicate more people too. The party’s “enemies” might use that resistance to hurt the party. When the 38th Group Army convinced Xu to return, they emphasized the need to “keep the big picture in mind.”
What Does This Tell Us About Xi Jinping?
Xi speaks openly and frankly about risks. He worries about small problems becoming big problems; economic and social problems becoming political problems; butterfly effects; international and domestic threats uniting; black swans and gray rhinos. To ensure that the People’s Republic of China does not fail like so many previous dynasties and regimes, he speaks of self-revolution — a continuous struggle to achieve the right value system. And self-revolution is nowhere more important than in the military, the party’s most important final guarantee.
Shortly after coming to power, Xi famously declared that the Soviet Union collapsed because the Soviets lost control of both their ideology and their military. He spoke about how the Soviet Army was “nationalized,” meaning that it was no longer in the hands of the Communist Party of the Soviet Union. Later, in an internal speech to the Beijing Military Region, Xi was explicit that, “During the political turmoil of the spring and summer of 1989, our Party was able to persevere because the military remained loyal to the Party’s command and not a single soldier was swayed by the enemy … Therefore, we must never relax our grip on the Party’s absolute leadership over the military. … In fact, various hostile forces at home and abroad have realized that as long as our army remains under the command of the Communist Party of China, their schemes to destabilize and destroy China will never succeed.”
To achieve “absolute leadership,” Xi calls for a continuous battle to inculcate idealism and dedication among the ranks. For Xi, all mistakes are at heart a problem of political consciousness, or at least it makes the most sense to talk about continuing challenges with that kind of language. That explains why Xi is so preoccupied with corruption: Materialism facilitates an individualistic mindset that is vulnerable to Western values, like a national, not party, military. Therefore, any “mistake,” even in the course of operating normal bureaucratic procedures, is very dangerous because it is an ideological problem.
Xi never wants something like mass protests to happen ever again in the first place, which is why he cares so much about expanding party control at the grassroots. Unlike during the Tiananmen crisis, when Deng outsourced much of the early decision-making, Xi would likely act more decisively.
But as Xi himself admits, it is hard to predict when and if mass protests might occur. Xu’s case helps us formulate questions about what might happen if they do. Would the military swiftly agree to kill for Xi — someone without the revolutionary prestige and authority of Mao Zedong or Deng? What might the military elite do if they were ordered to execute a risky gamble in the Taiwan Strait? When faced with such a situation, what are Chinese military commanders capable of doing to prevent a disaster? And as they consider a complicated set of calculations about what is and is not appropriate, would this newly released video of Xu’s closed trial play a role? Would they see inspiration not only in his bravery at the time but in the pride his behavior has triggered in the people who learned about what he did? Or will they see a cautionary tale for anyone who wants to take a risk to do the right thing in a system where any misstep could cause terrible repercussions?
Joseph Torigian is an associate professor at American University’s School of International Service and a senior fellow at the Council on Foreign Relations. He is the author of The Party’s Interests Come First: The Life of Xi Zhongxun, Father of Xi Jinping and Prestige, Manipulation, and Coercion: Elite Power Struggles in the Soviet Union and China after Stalin and Mao.
Image: Pete Campolongo via Wikimedia Commons
