Regime stability is a funny thing. One day it’s there; the next day, poof—it’s gone.

For the past several months, I have been talking to experts about a possible coup in Russia. I approached the question gingerly. It seemed too much to hope for; it seemed naïve. Vladimir Putin had been in power for more than two decades. Many had predicted his demise—always prematurely. There was a small cottage industry on Twitter of people insisting that Putin was ill. They liked to post photos of him sitting at meetings, clutching his desk as if he were about to fall. I didn’t want to be like that. “Is this ridiculous to even think about?” I would ask the experts. The experts laughed. They felt the same way. A coup was unlikely, they agreed. A popular uprising—a “Ceaușescu scenario,” in which the people stormed the Party’s headquarters, convened a hasty trial, and murdered their dictator—probably even less so. To a scenario like the one that actually played out last weekend—one of Putin’s warlords raising a mutiny, taking over one of the country's military headquarters, and marching on Moscow, all while Putin was still in power—we gave very little consideration. It just seemed too outlandish to talk about.

And yet, since the war began, all of the experts had been thinking about ways in which the Putin regime might collapse, and watching what Putin was doing to protect himself. Peter Clement, a former director of Russia analysis at the C.I.A., noted a televised meeting, days before the war, in which Putin browbeat members of his security council into pledging their support for his Ukraine policy. It was a brilliant move by Putin, Clement thought, to bring his senior administration officials in line. “They’re all complicit now,” Clement said. “It’s not like one of them can say, ‘I thought this was a stupid idea.’ They all signed on.”

For that reason, Clement thought it more likely that a move against Putin would come from the second circle, from someone less in the public eye, someone we’d not heard of. Clement was willing to speculate with me, but he considered the chances low. You’d have to have the security services on board, he said, because you’d need to physically arrest the President, and it was unlikely you could appeal to security hawks with an antiwar agenda. And you’d have to be prepared to run the country. It’s a big country and in the thick of a long war. “It can’t just be, ‘We got rid of the Wicked Witch of the West! Let’s all stand up and cheer!’,” Clement said. You’d have to have a plan, and Clement was having trouble thinking of people who might have one.

Another former C.I.A. analyst, Andrea Kendall-Taylor, who was a deputy national intelligence officer for Russia and Eurasia between 2015 and 2018 and now runs the Transatlantic Security Program at the Center for a New American Security think tank, walked me through the political-science literature on how authoritarian regimes tend to fall. Of the four hundred and seventy-three authoritarian regimes that had fallen between 1950 and 2012, a hundred and fifty-three had done so via coup. But the coup was on the wane; after the end of the Cold War, the U.S. had stopped propping up quite so many military dictatorships, which are what tend to get militarily couped. It was unlikely, Kendall-Taylor explained, that the security services, or anyone from Putin’s inner circle, would move against the Russian President, because the regime had entered the stage that the political scientist Milan W. Svolik called “established autocracy.” In an established autocracy, the leader has monopolized power to such an extent that he can no longer be threatened by what Svolik calls an “allies’ rebellion.” The truth is, Kendall-Taylor said, most personalist dictatorships, such as Putin’s, ended with the dictator dying in power, especially when the dictator was older than sixty-five (Putin is seventy). “That is by far the most likely scenario,” she told me. She put the chance of regime change in Russia in the next two years at ten per cent, and that “ten per cent includes Putin having a heart attack.”

The historian Vladislav Zubok, who is the author of a recent book on the collapse of the Soviet Union, described the various ways in which other Russian and Soviet leaders—Nicholas II, Nikita Khrushchev, Mikhail Gorbachev—had been ousted, and explained why none of those scenarios mapped onto this one. Nicholas II had abdicated, in 1917, after large protests in Petrograd (current-day St. Petersburg, then the Russian capital) shattered confidence in his regime, and the military joined the mutiny; Putin, Zubok pointed out, had made sure that his capital, Moscow, was well-provisioned and maximally isolated from the war in Ukraine; there is a loyal paramilitary force to control protests. Khrushchev was overthrown, in 1964, by a plot from within his own inner circle, led by his deputy, Leonid Brezhnev, who worked within the structures of the Communist Party to urge others to turn against their leader. The K.G.B. played a key role in the coup. Putin’s regime, by contrast, is highly informal, much more like Stalin’s, with all paths leading, in the end, to Putin. It is hard, under such circumstances, to plan a coup. And there are several branches of secret police, each competing with the others, making any plotting very complicated. As for Gorbachev, the comparison seemed the least apt of all. He had not only allowed his rival, Boris Yeltsin, to run for President of Russia—he permitted the government to finance his campaign. Putin was unlikely to do something like that. If there was a leader to whom Putin could be compared, Zubok said, it was Ivan the Terrible, who ruled Russia in the second half of the sixteenth century. Ivan fought a long war of attrition with his Western neighbors; he demoralized his ruling élite, and murdered his own son and heir. After his reign was over, the country eventually fell into civil war, the period known in Russian history as the Smuta, the Time of Troubles.

Two experts on Russian public opinion described their understanding of Russian attitudes toward the war in Ukraine, and what might cause those attitudes to change. Oleg Zhuravlev, a founding member of the Public Sociology Laboratory, an independent Russian research collective, summarized a series of in-depth interviews that his team had done with young Russians in the past year. They had found that support for the war was both thinner and narrower than it looked. There was a small group, about ten to fifteen per cent, of genuine supporters; there was a similarly small group of genuine opponents. In between was a large group of people, most of whom had come around to supporting the war not because they thought it was a good idea but because they didn’t know how to oppose it, and because they felt totally alienated from the people in charge of it. “Over and over we heard the same thing,” Zhuravlev said. “ ‘If there’s one thing I know about politics, it’s that I don’t know anything about politics. The people in the Kremlin are foreign to me; they are not like me. But they must have their reasons.’ ”

It was depoliticization in its purest form. Zhuravlev’s occasional collaborator, the longtime polling expert Elena Koneva, had spent the year and a half since the war began running a project called ExtremeScan, through which she designed polls to figure out the basis for Russian public support of the war and what could cause it to contract. She had seen signs, mostly in the border regions of Russia, that, when the war began to truly affect people’s lives, their opinions started to change. First they experienced fear of retribution—“We have done so many horrible things to Ukraine,” one respondent said, “that the Ukrainian Army will inevitably come here”—but the actual experience of war, of shortages, of shelling, of people being forced to evacuate, began to erode support for the war. And Koneva predicted that, if things got worse, support would erode further. “If people are constantly having to sit in bomb shelters, and women are giving birth without medicine,” she said, “then an end to the war will become their most passionate wish.”

Yevgeny Prigozhin figured in our conversations as a grotesque and somewhat comic character. When looking at the Putin regime, one Moscow-based historian said, “We’re all wondering who the Beria figure is going to be,” referring to one of Stalin’s most efficient henchmen, tried and executed by his former comrades after Stalin’s death. “Who are they going to take out and shoot right away? And then you look at the criminal types who are working for the Kremlin—and you see Prigozhin. There’s your Beria.”

For Kendall-Taylor, speaking in May, Prigozhin’s antics—his profane insults and increasingly aggressive rants, which included accusations of treason against the Russian Army’s leadership—were a sign of élite discord. In a post-Putin world, she said, the presence of warlords like Prigozhin and Ramzan Kadyrov, the head of the Chechen Republic, could lead to a “Sudan scenario,” in which these forces would start a civil war. In the near term, though, with Putin still in power, she did not think Prigozhin would undertake an actual rebellion. At the time, his criticisms of the military seemed only symbolically significant, a sign that the élite was in disarray and that protest actions, whether secessionist or antiwar, might not be met with as much force as people had once thought.

Regime stability is a funny thing. One day it’s there; the next day, poof—it’s gone. The Moscow-based historian, who asked that his name not be used since he was still in Russia, recalled what it was like to observe the Politburo in the early nineteen-eighties. “They looked like a totally homogeneous mass,” he said. “There was no indication, in their public statements or in anything else, that any of these people thought differently from one another.” But Gorbachev, it turned out, did think differently. In the years to come, he undertook a series of reforms that ended with the Soviet Union ceasing to exist. Authoritarian regimes could seem very stable, until suddenly they weren’t.

On the first anniversary of the Russian invasion of Ukraine, Kendall-Taylor convened a group of experts to compile a “stability tracker” for the Putin regime. The tracker identifies ten “pillars,” ranging from “Absence of an alternative to Putin” to the idea, among Russian citizens, of “Russia as a besieged fortress,” and tries to indicate whether these are growing stronger or weaker. As of this spring, several factors were going in the wrong direction for Putin: his élite was becoming fragmented; his economy was suffering the effects of the war and of sanctions; and his military, historically apolitical, was being pulled into the political arena by concerns over Prigozhin’s rising influence and its access to military resources. But the factors going in the other direction were more numerous: according to Kendall-Taylor’s experts, Putin had strengthened his control over the information environment; the people most discontented with his rule were leaving the country; and the idea of Russia as a besieged fortress was gaining rather than losing adherents. Most important, there remained no viable alternative to Putin: his warlords were politically unpopular, and his heroic opponent, Alexey Navalny, was being denied food, sleep, and medical care in a Russian prison. In the absence of an alternative, the status quo would continue.

Among experts thinking about the Russian regime, there are, very roughly speaking, two kinds: those who look at Russian and Soviet history and culture to determine what might happen next, and those who look at Russian authoritarianism in comparative perspective—that is, alongside authoritarian regimes in Egypt and China and Turkey. This is also known as the political-science-versus-area-studies debate. The different approaches yield slightly different hypotheses. Before joining the C.I.A., Clement wrote a Ph.D. dissertation on the so-called Congress of Victors, in 1934, at which Stalin consolidated his rule and also began to see that not everyone was satisfied with it. In the eighties, Clement analyzed the succession struggles during the post-Brezhnev period. His interest in regime insiders partly stems from this experience. Kendall-Taylor, a comparativist with a Russian focus, who studied in graduate school with Barbara Geddes, one of the founders of modern quantitative authoritarian studies, prefers looking at the numbers: this many regimes of this type fell in this manner; this many regimes of this different type fell differently. But everyone agrees that the life of a regime is full of contingencies; leaders can make mistakes. They can, for example, start a brutal and senseless war against a neighboring country, and refuse to relent even when the war is going badly.

War is a known stressor for personalist dictatorships, which (paradoxically, one might argue) also tend to start more wars. War puts pressure on the economy and on security services, and also has a way of being unpredictable. Between 1919 and 2003, according to the political scientists Giacomo Chiozza and H. E. Goemans, nearly half of all rulers who lost wars then lost power within a year. (Of these, half were sent into exile and nearly a third were jailed.) This is true of Russian history as well. War losses, such as Russia’s 1905 defeat by Japan, have sometimes led to tectonic shifts in the country’s political life; in 1905, it led to an uprising that forced the tsar to grant his people a constitution. In 1917, the Russian Army’s struggles in the First World War were a major factor in pushing the tsar out of power.

But wars have an upside for authoritarians, too. They offer an excuse to increase repressions and take over the information space. And repression, unfortunately, works. It makes people more fearful of voicing their dissent and of coming out into the street. In the current situation, Kendall-Taylor gave credit to the work of the Russian security services, who have tended to respond to protest with some restraint. “They don’t overreact in ways that could spiral and trigger a public reaction,” she said. “They’re using facial recognition and other things. It’s a knock on the door out of the public eye rather than beating people in the square.” This tends to mitigate some of the dangers of public disapproval that are inherent to a repressive system.

On the other hand, repression’s ability to decrease the flow of information can be dangerous for a regime. People don’t know how dissatisfied other people are—and neither does the government. Kendall-Taylor recalled Mohamed Bouazizi, the Tunisian fruit vendor who set himself on fire in December 2010 and started the Arab Spring. This was an example of the downside of repression. Repression works, until it doesn’t. In Tunisia, one desperate act led to a popular uprising and the fall, in weeks, of a regime that had been in power for decades.

Prigozhin’s march to Moscow came as a shock to just about everyone. Clement had been on high alert ever since the Russian authorities declared, on June 10th, that they would require all Wagner soldiers to sign a contract with the Ministry of Defense—that is, technically, to dissolve Wagner as an independent entity. Zubok had also noted the increasing fractiousness of the Russian élite, and wrote to me before the uprising that the emperor might not be clothed. Kendall-Taylor, in an article she co-wrote for Foreign Affairs, mentioned Prigozhin as a possible pretender to the throne after Putin’s departure. But no one expected the sequence of events that unfolded last weekend.

In their aftermath, there were more questions than answers. Zubok wrote a piece for the New Statesman in which he compared Prigozhin’s act to Julius Caesar crossing the Rubicon and marching on Rome in 49 B.C.; he recalled that the oligarch Mikhail Khodorkovsky had once proclaimed, when he was a young man on the make in the new Russia, that the man with the rifle—an iconic symbol of the Bolshevik Revolution—had been replaced by the man with the ruble. But now the man with the rifle was back; a new political era had dawned in Russia.

I asked Zubok why he had decided to go outside of Russian history to try to explain the Prigozhin phenomenon. He said that Russian history can sometimes feel like a straitjacket. “ ‘Hey, it’s the Smuta! Hey, it’s 1917! Hey, it’s 1991.’ I’m the last person to argue that those patterns don’t matter.” But sometimes you want a little something different. Sometimes you want to explore other historical connections.

But, he added, “there is a huge caveat.” Despite its many wars and campaigns, Russia has never actually produced a Caesar—that is, a warlord who marches on the capital with his men and takes political power. There is a reason for that. The traditional political system, arguably still in place to this day, is a triangle comprising the tsar, the boyars, and the people. In times of trouble, the tsar can play the people against the boyars, and vice versa. If things go bad, the boyars can take the blame. This is, in effect, what Prigozhin was asking for—that Putin sack his boyars in the Army, who had made such a hash of the war. Zubok acknowledged the danger inherent in such a strategy for Putin: “You may think that someone is a Red general, but next thing you know they’ve turned around and are executing the Bolshevik leadership”—as happened with Ivan Sorokin, a revolutionary commander in the Russian Civil War who went rogue in the North Caucasus and attacked the Soviet leadership in his own district before finally being killed himself. But this is the sort of thing that happens in the absence of a tsar, when the Smuta is in full swing, whereas Putin, however weakened, remains the tsar. “You have to acknowledge the sources of resilience in this crazy system,” Zubok said. His prediction was that Putin would remain in power, chastened but basically unchanged.

This was also Clement’s analysis. There were many things about the events of the weekend that he found notable, including the fact that Prigozhin’s column of trucks and armor, travelling, exposed, along Russia’s highways, had managed to go as far as they did. “When was the last time you saw a column on a highway?” Clement asked, recalling the Russian trucks and tanks on the road to Kyiv last year that had been methodically picked apart by Ukrainian forces. His conclusion was that local commanders did not feel that they could take the initiative to destroy Prigozhin’s column; it was above their pay grade. He was also fascinated by Putin’s five-minute video address, recorded during the uprising, in which Putin spoke of treason and betrayal and seemed to compare himself to Nicholas II, unable to prosecute a war because of intrigues behind his back. “Was it absolutely necessary to make this speech?” Clement asked. It made Putin look panicky and weak, he said: “If you really think it’s a full-blown rebellion, why don’t you take him out?”

Nonetheless, he could see no pathway to a Russia without Putin. An analysis in the Times had suggested that there could be talk in his inner circle of asking Putin not to stand for reëlection in 2024. Clement was skeptical. “The trouble with that is, who’s the person who’s going to go in there and say that to him? Who is going to say, ‘Vladimir Vladimirovich, look: you are a very rich man—we think you should go and retire and just live happily ever after’?” Clement recalled an incident during the Iran-Iraq War, in which one of Saddam Hussein’s advisers suggested that a way to forge peace would be for Saddam to temporarily step down as Prime Minister. The man was executed and his body parts delivered to his family the next day. “Dictators don’t like to be told that they should retire,” Clement said. He added that, in this, they weren’t necessarily wrong: Could Putin actually retire? Who could guarantee his safety? Wouldn’t whoever replaced him as the ruler of Russia find it very uncomfortable to have Putin still hanging around? “This isn’t like Khrushchev, where he can just go live quietly on his farm,” Clement said. This was a person with a lot of enemies.

For the moment, at least, it was also a person with a lot of power. “He still controls the Army and the F.S.B.,” Clement said. “And people are still afraid of him.” Clement suspected that Prigozhin would meet an unhappy end. Putin is notably vengeful, widely believed to have approved the murder of people he deemed to have betrayed Russia, long after they had done it, though the Kremlin has repeatedly denied the country’s involvement. But Clement also believed that the experience with Prigozhin could make Putin more cautious: he might make some changes among his advisers, as a way of explaining the failures of the war; he might even conclude, Clement speculated, that the war was taking away too many resources, that he needed to focus on domestic concerns, and that he would therefore consider engaging in ceasefire talks, so that he could regroup and possibly resume war later. That, Clement went on, was still low probability. But the probability had increased.

Zhuravlev, the sociologist, observed Prigozhin’s uprising from a hospital in Almaty, Kazakhstan, where he had gone to do research and had come down with appendicitis. He found the spectacle fascinating—both terrifying and encouraging—but the videos of ordinary Russians in Rostov-on-Don greeting Prigozhin’s fighters and cheering them on did not entirely surprise him. Zhuravlev interpreted it, optimistically, as a sign of engagement. In his interviews with young Russians, he had been struck by how many of them wanted to talk about the war but had no opportunities to do so. Now there was an opportunity—not owing to a democratic movement, to be sure, but an opportunity nonetheless. The uprising was encouraging in another way, too: it showed that it might not be so hard to organize a revolution in Russia. “For now, the people who want to do it don’t have the means,” Zhuravlev said, “and the people who have the means don’t want to do it.” But perhaps this could change.

Kendall-Taylor was quick to clarify that the events of the past weekend were not an attempted coup but an insurgency, a sign of frustration rather than a planned attempt at regime change. Nonetheless, the message that Prigozhin’s actions had been sending for months—that the regime was not as strong as it seemed; that you could defy it and survive—had been significantly strengthened. “So much of the glue of these regimes is that no one knows how widely held the discontent is,” she said. “When something like this happens, it sends such an informative signal that others are as dissatisfied as you are. It starts to change people’s calculus about what is possible.” The next time something happens in Russia that people do not like—it could be a major military defeat, or, slightly more likely, according to Kendall-Taylor, something to do with the 2024 Presidential election—they may not be so worried about going out into the street to say so. Prigozhin may or may not survive his stay in Belarus, but here was a person who marched with a small army several hundred miles through Russia, without encountering any real resistance. That didn’t mean the regime was in danger of imminent collapse—but it suggested that the chances had increased slightly. “That’s the way these regimes unravel,” Kendall-Taylor said. “At the end of the day, whether it comes from a coup or an insurgency or a protest, Putin will at some point give an order to crack down and fire, and people won’t do it. And that’s the end of the regime.”

No one could predict the future. But it was worth trying to analyze the situation and think it through. Earlier, Kendall-Taylor had said that the chances of Putin no longer being in power in two years were ten per cent. Now she was willing to go up to twenty.