I give a pretty straightforward definition based on how the term is used in the social movements literature today and in some of the more contemporary studies of revolutions: a revolution is a mass siege of an incumbent regime by its own population aimed at bringing about regime change and substantive political or social change in its wake. As Leon Trotsky put it, a revolution is about citizens seizing back control over a regime through mass mobilization from below.

Revolution has been an evolving phenomenon since the seventeenth and eighteenth centuries, when modern revolutions first came into being. There’s a debate over whether there were revolutions in the ancient world. If there were, they were of a different sort and were on a very limited scale.

For instance, some believe that there were revolutions in ancient Athens. But the total population of ancient Athens, at least in terms of free citizens, was small (at most, about one hundred thousand people). Athens was a city-state, on an entirely different scale than the contemporary nation-state. But when we talk about revolution in the modern era, we’re talking about the engagement of very large numbers of people and the modern nation-state, with its territorial ambitions, its economic goals, and mass political organization.

Revolutions across history have been put to many different purposes. They began largely as a way for society to contain the usurpations of monarchs. They evolved in the nineteenth century and gained a social component, aiming at the transformation of the class structure of society. That social element dominated for a long time. But there were always political revolutions that occurred throughout this period, and the political dimensions of revolution predated its social dimension.

In addition to anti-monarchical revolutions, there have also been revolutions aimed at attaining democracy, revolutions for independence from a colonial power or from an existing state, revolutions to invert a racial hierarchy or ethnic hierarchy, and Islamist revolutions aimed at transforming a secular state into a religious one. That doesn’t exhaust the various purposes to which revolution has been put across history.

My book is essentially about how revolutions have evolved over time. I don’t look at all different types of revolutions in detail. I was particularly interested in the decline and marginalization of social revolution and the rise of what I call urban civic revolutions. Social revolution had its heyday in the mid-twentieth century. But in recent decades, social revolutions have faded, and political revolutions aimed at containing corrupt and repressive dictatorships have multiplied in their place. This has been accompanied by a shift in the location of revolutions; they’ve moved from the countryside to cities. In the nineteenth century, social revolutions were predominantly an urban phenomenon. But they migrated to the countryside in the mid-twentieth century and became predominantly rural.

Urban civic revolutions have become the predominant form of revolution around the world. An urban civic revolution seeks to mobilize as many people as possible in central urban spaces in order to bring about regime change through the power of numbers rather than the power of arms. Because they try to mobilize as many people as possible in order to disrupt government, they’re quite diverse in the coalitions that underpin them. Part of why they have become the predominant form of revolution around the world has to do with the movement of millions of people into cities over the past century, the massive urbanization that has transformed the nature of our world.

In fact, you don’t see this form of revolution until the late twentieth century, when very large numbers of people concentrated in cities. In 1900, something like thirteen cities around the world had a million inhabitants or more. Today, we have 548 such cities. So it’s much easier today to generate very large crowds and use them as a basis for bringing about regime change than it was a century ago.

Traditionally, revolution was an armed phenomenon. But armed revolution in cities was generally a losing proposition, as the state possesses overwhelming force in cities. The state has greater numbers of armed combatants, better weapons, and better training, and these forces tend to be concentrated in cities, where the nerve centers of government are located. This was something that revolutionaries recognized already in the nineteenth century.

Friedrich Engels, for instance, wrote that armed revolutionaries in cities were at a big disadvantage vis-à-vis the state because the state had overwhelming firepower. This was the main reason that armed revolutionaries moved from the city to the countryside in the mid-twentieth century. In the process, social revolutionaries discovered the revolutionary potential of peasants, who previously had been thought to be reactionary and focused largely on the issue of access to land rather than transformation of class structures.

There were other reasons besides urbanization that made urban civic revolutions possible in the late twentieth century. In the late nineteenth and early twentieth centuries, governments dealt with unarmed crowds in very violent ways. Unarmed revolutionary crowds in the early twentieth century were six times more likely to involve deaths than they are today, as governments were more likely to shoot at them or ride over them. So revolutionaries had to arm themselves simply for protection.

But today, with the invention of “nonlethal” methods of crowd control, unarmed revolution has become much less dangerous. It still involves significant risk. Obviously, people are still shot at and killed in revolutionary crowds today. But it is much less likely to happen than in the past.

There are also technological reasons why the urban civic revolution didn’t occur until the late twentieth century. In the early twentieth century, for instance, you couldn’t hear people in a large crowd at a distance, since there were no systems for sound amplification. It wasn’t until the 1930s, when the Nazis first applied sound amplification to large demonstrations and street processions, that you could hear people in large crowds at more than ten meters. In earlier periods, revolutionaries also relied largely on highly localized neighborhood and factory networks for mobilizing people. This also limited the size of crowds.

But by the late twentieth century, all these things changed. With the rise of television and the internet, there was a major shift in the technological environment, allowing for the mobilization of larger numbers. Digital technologies have transformed the process of revolution, accentuating visuality and simultaneity and traversing political boundaries with great speed.

The political and international environment also shifted in ways that were conducive to the rise of urban civic revolutions. And the demographic concentration of people in cities made possible revolutionary challenges based on the power of numbers rather than the power of arms. So revolutions today are markedly different than in the past — not only in their relocation to cities, but also in the entire manner by which they are carried out.

There have been multiple transformations in the physiognomy, social structure, and character of cities over the last two centuries that have affected the nature of urban revolutionary challenges. When you think back to the nineteenth century and to industrialization, the classic situation was, say, Paris, which contained very densely settled working-class neighborhoods near centers of power. These dense working-class warrens provided a conducive physical environment for barricade warfare.

But in order to combat rebellion, the government broke up these neighborhoods. The neighborhoods were cleared out, and large, open spaces were created in their place. In Paris, this was associated with Hausmann and then was later imitated elsewhere around the world. States also sought to create large ceremonial open spaces in cities as they grew in power. And as cities grew, larger streets and boulevards became necessary simply for the movement of people. These large boulevards and squares were not particularly good for armed rebellion. But today they are exactly where large numbers of people concentrate in revolutionary crowds.

In short, the opening of spaces in cities provided the physical environment conducive to urban civic revolts based on the power of numbers.

Neoliberalism has also had effects on cities. For one thing, gentrification has often pushed poor and working-class people to the physical periphery in many cities around the world. The urban poor generally don’t participate in urban civic revolutions in numbers relative to their share of the population. Rather, the educated upper-middle class participates disproportionately. I’m not talking about the bourgeoisie, the capitalist class. Capitalists generally don’t participate in revolutions, as there’s no good reason for them to do that. They are usually fairly well plugged into the ruling regime. But the educated service class does participate disproportionately.

In my book, I use representative national surveys from four different urban civic revolutions — two from Ukraine, one from the Egyptian revolution, and one from Tunisia — to investigate who participates in urban civic revolutions. The surveys show that across these cases, the educated and better-off participated disproportionately relative to their numbers in society. But they also show that the crowds that participated in these revolutions were quite diverse, not only in terms of their class profile but also in terms of their political views. Urban civic revolutions are quite diverse coalitions. You couldn’t generate the huge crowds necessary to dislodge a regime with just upper-middle class people alone.

That diversity is reflected in their goals and demands. They revolve around negative coalitions, focusing on the eviction of a corrupt and repressive regime rather than for a particular set of changes. In this respect, urban civic revolutions are often framed as democratic revolutions. But as the surveys show, most participants have a weak commitment to democratic values. Moreover, the issues that motivate most participants tend to revolve around corruption and economic issues, while political and civil liberties are a motivating factor for only a minority of participants, usually about a third or a quarter.

In terms of the effects of neoliberalism, there have been a significant number of revolutions sparked by crises resulting from neoliberal reforms on urban populations, changes like the contraction of public services, or price increases that have been imposed on society. Not all urban civic revolutions come into being in response to neoliberalism. But there is a subset that has been sparked by neoliberal reforms. Neoliberal globalization has also connected cities in a new way and made the global diffusion of revolution a quicker and more widespread phenomenon. The diffusion of revolutions across countries has become more pronounced as a result of globalization.

Contemporary scholars have generally moved beyond the Leninist conception. Revolution is not merely a matter of the depth of grievance or the level of repression. These have never accurately predicted revolution. Studies show that most people suffer in silence in the face of grievances. Certainly, grievances matter. But they are grossly insufficient to explain revolution. Leadership matters, as Lenin himself would have argued. Resources matter. And opportunities matter. But even these do not fully explain the outbreak of revolutions, which have been notoriously unpredictable.

In terms of what a revolutionary situation is, scholars today instead take their cue from Leon Trotsky, who argued that a revolutionary situation is a situation of dual sovereignty, when competing claims for sovereignty over the same state emerge. Trotsky was obviously writing about the Russian Revolution, in which two competing centers of power, the Provisional Government and the Soviets, vied for predominance. It’s not always the case that you get to two formal centers of power claiming sovereignty over the same government. Sometimes, in contemporary revolutions, where the leadership of revolution is often diffuse, it’s just people rejecting the sovereignty of the incumbent regime, with the alternative to the incumbent regime being more implicit.

The Leninist conception is highly structural. But revolutionary situations are not merely a matter of structural conditions. They emerge out of interactions between governments and oppositions that can turn periods of reform or repression into open revolt. There are multiple possibilities that occur within these conjunctures, and a great deal of unpredictability is built into revolution due to the choices that governments and oppositions make in reaction to one another.

This doesn’t mean that there’s not a structural dimension underlying revolution. As I show in the book, there are a series of structural conditions that accurately predict the emergence of about 80 percent of the urban civic revolutions that occurred: high levels of corruption, intermediate levels of repression, leaders who have been in power for a long time, a lack of oil resources, and lower-middle levels of development.

But these conditions also seriously overpredict the likelihood of revolution. Based on these conditions you would expect to see many more revolutions than actually occur. That’s even taking into consideration certain triggers that are known to set off revolutions, like price hikes, a financial crisis, or international wars.

The structural conditions that underpin revolutions overpredict revolution because of the role that government-opposition interactions play in the emergence of revolution. For example, in response to a challenge, governments can respond with co-optation. Alternatively, they can respond with reform or repression. These sometimes dissipate revolutionary challenges before they gain momentum.

Problems also emerge within oppositions that prevent them from cooperating across groups. In the book, I trace out several examples where there was a higher-than-normal risk of revolution in a country based on the structural conditions that mattered elsewhere, but revolution didn’t happen, and I look at why it didn’t occur despite the heightened risk. What I find is that agency and choice make a difference in these situations.

Revolution is a structured phenomenon. It generally occurs where we would expect it to occur and under a particular set of structural conditions. Nevertheless, revolutions also depend a lot on the choices people make, and these interactions often occur with quite unpredictable results. Errors in revolutionary contention are also quite common. For all these reasons, revolutions typically come upon us as surprises.

It’s hard to say. We don’t really have a handle on all the routine modes of contention and reformist movements in the same way as we do revolutions, which are a far less common situation. I can’t say that reformist movements, for instance, are more or less successful than revolutions in terms of achieving their substantive goals. Change is hard.

However, I think we can say that for certain types of goals, there has been a shift away from revolution as a way of achieving change. For instance, there haven’t been successful social revolutions since the late 1970s or early 1980s in terms of gaining power, and since the mid-1990s relatively few new revolutionary episodes have emerged that have sought class transformation of society. However, we do know of nonrevolutionary movements that have had goals of transforming the class structure of society that have come to power through the ballot box or that have influenced government in other ways. Revolution may have become a less useful way of changing the social structure of society. Part of the reason for this is because changing the social structure of society tends to divide society and elicit a great deal of violent resistance, making social revolution more violent and more difficult to achieve.

To me, the Orange Revolution is the archetypal urban civic revolution. It occurred because a corrupt and repressive government tried to engage in electoral fraud to stay in power, and this elicited significant opposition across broad sectors of society. It tried to mobilize as many people as possible in central urban spaces, in large open areas in order to bring about regime change. And it was spectacularly successful in mobilizing large numbers. They had crowds of up to a million people in central Kyiv.

But very quickly afterward, the negative coalition that dominated that revolution split after coming to power. It was so diverse that it couldn’t be held together. It also didn’t push aside or transform the state it inherited in any way, and corruption continued to flourish. This led to a situation where Viktor Yanukovych, the candidate who ran in the election that sparked the revolutionary crisis in the first place, was elected to the presidency six years later in 2010. So the very people whom the revolution evicted from power came back into power at the ballot box not too long afterward due to the failings of the governing coalition that emerged from the revolution.

This lack of persistence is typical of urban civic revolutions. They produce governments that are highly fractious. Part of this is because they have very minimalist, negative goals that seek to unite as many people as possible. They are largely about evicting a repressive and corrupt government and are more about what people are mobilizing against than what they are mobilizing for. If you’re going to rely on the power of numbers, you must have demands that are going to appeal across large numbers. If you live by the power of numbers, you die by the power of numbers.

Of course, there was a second revolution in Ukraine, largely because the first one didn’t end the corruption or repression that had sparked the first revolution. That second revolution, the Euromaidan revolution, was somewhat more successful in transforming the state. But it has also struggled quite a bit in doing so, largely because it inherited the corrupt state that Yanukovych built. Certainly, part of the challenges facing contemporary Ukraine have been due to war. But the Russian annexation of Crimea and the invasion united Ukrainian society in a way that otherwise probably wouldn’t have been the case in the wake of revolution. It allowed a new coalition to emerge around Zelensky that has seriously attacked corruption.

Exactly. The coalition in Egypt also couldn’t hold itself together in the face of what came afterward. It split, and eventually things deteriorated to the point that the liberals who had allied with the Muslim Brotherhood to evict the Mubarak government allied themselves with the remnants of Mubarak’s military in order to evict the Muslim Brotherhood from power. That story is the tragedy of the Egyptian revolution, and we know the consequences in terms of the enormous repression that is taking place there today.

For urban civic revolutions to succeed, it often takes an external threat to keep revolutionary coalitions united, because the natural tendency is for them to break apart after coming to power.

Overall, violence has declined in revolution over the long term. Revolutionary civil wars have become less common than in the first half of the twentieth century or the Cold War period. Even within those wars, there are fewer deaths. And even in unarmed revolutions, there has been a decline in lethal violence.

There’s a unique element to the Ukrainian situation given Ukraine’s relationship with Russia and the way in which this sparked war. But after a long decline in violence in revolution, revolutions today are once again becoming somewhat more violent, though in a different way. We’re not seeing civil wars so much as more riotous violence, especially in cities. You saw that in the Euromaidan revolution. What began as a typical urban civic revolution mobilizing very large numbers of people largely in response to government repression evolved into violent street riots. However, the urban civic element was not capable of evicting the Yanukovych government.

While the urban civic model has generally been quite successful on its own terms, the targets it’s being applied to in recent years have grown more difficult to topple. Protesters are confronting more repressive governments that aren’t as easily moved by the power of numbers. These governments have learned how to deal with urban civic protests, how to manage them spatially and how to wait them out until they exhaust themselves. You see this in Belarus as well. This failure of the power of numbers has pushed several cases into more riotous forms. So yes, they are becoming more violent, but more in terms of street violence than civil war.

The collapse of the Soviet bloc was certainly a contributing factor. But it’s bigger than that. Take Theda Skocpol’s classic work on social revolution, which is generally held to be the model of scholarship on the topic. Skocpol sees social revolutions as rooted in a particular type of social formation, what she calls an agrarian-bureaucratic society. Agrarian-bureaucratic societies are societies in which the surplus produced by peasants is essentially shared between a government and an aristocratic or landed elite that’s in alliance with the government. This is the classic type of society that was vulnerable to social revolutions throughout the nineteenth and twentieth centuries.

Accordingly, social revolutions historically were closely connected with land inequality and lack of access to land. They typically occurred in peasant societies. Even when mobilization in social revolutions occurred in cities and was led predominantly by urban classes, the peasant element was still prominent. But by the late twentieth century, that agrarian-bureaucratic society had begun to fade.

What happened? Well, a third of the world experienced communist revolutions that wiped out that aristocratic landed class altogether. Then you had land reform in other places that occurred as a result of the threat of social revolution and that was aimed at undermining its potential. Land inequality is still quite extensive around the world. But a lot of land has been redistributed to populations in many places, and that has mitigated elements of the social revolutionary impulse.

Then you have the massive migration of people into cities. Who are the people who migrate into cities? It is typically young men, the very same people who are most likely to participate in armed rebellion. When they migrate to cities, they leave behind a disproportionately older and female population in the countryside that often depends upon wages earned in the cities and remittances sent back to the village. So access to land is no longer as important as a source of subsistence, and often it’s access to wages in cities that has taken its place. You also have developments like the green revolution in the countryside, which has increased productivity in the countryside in some countries, so that people can produce more.

Finally, one of the main solvents of that aristocratic class has been democratization. Studies show that the power of the aristocratic, landed elite tends to be undermined through democratic reforms.

All these factors taken together have undermined agrarian-bureaucratic society. So social revolutions as we traditionally knew them no longer occur, not just because of the end of the Cold War and the Soviet Union. The Soviet Union didn’t create social revolutions. The role of the Soviet Union was essentially to provide arms to revolutionaries. Rather, the conditions that underpinned these revolutionary conflicts have deteriorated, and instead large numbers of people have moved into cities. This movement of large numbers into cities instead created a conducive environment for the urban civic model of revolution.

It would have to be reimagined. The agrarian-bureaucratic model of social revolution is not going to work anymore. But social inequality is still pronounced and has even gotten worse over time. So it’s not as if the issues underpinning social revolution have disappeared in urbanized societies. The eclipse of agrarian-bureaucratic society has deteriorated and marginalized that model of social revolution. But it doesn’t mean that some other model of social revolution might not be invented.

Such a model, given the urban character of societies, is probably going to have to rely on the power of numbers, not on the power of arms. So, it wouldn’t be as heavily ideologized and total as the earlier model, which generally sought to completely transform society. Again, if you are going to rely on the power of numbers, you must figure out how to appeal to very large numbers of people, and that is going to water down your demands. Also, because the coercive forces of the state are concentrated in cities, armed revolution in cities is a losing proposition. It’s just not going to work, at least not across a significant number of cases. A new model of social revolution, one that would attack class inequalities, would have to be different.

Right now, most of these issues are being handled through the ballot box to the extent that they are being dealt with. Generally speaking, democracy is the great solvent of revolution, because there’s really no reason to risk your life on the streets if in a few years, you can change your government through the ballot box.

All revolutionary episodes, for that matter. Generally, there are no revolutions after a certain degree of democracy has been attained. Revolutions have occasionally occurred in democracies, but they are very, very rare. The sweet spot is at a middle range of repression — not in the most open regimes and not in the most repressive regimes, though the most repressive regimes do experience revolutions more often than democracies.

This is where democratic backsliding could play a role. We don’t know what the future of democracy is today. It’s in great doubt and under threat. We have seen backsliding in democracies around the world and a movement toward more authoritarian regimes. If democracies do move back toward authoritarian forms of government, that could make revolution a more attractive proposition.

Yes, that’s what I have in mind — but even more so, the growth of dictatorial power that overrides any restraints on executives. There have been cases in which revolutions have occurred under those kinds of circumstances. The real question is, how much influence do people have at the ballot box to restrain these tendencies? Can they change their government through the ballot box when there is the will to do so?

Historically we have had occasional right-wing revolutions. It’s not a totally unprecedented phenomenon. The rise of Mussolini might be interpreted as a revolution, and in fact in my book I do count it as a revolutionary episode. So, it’s certainly possible.

I wouldn’t count January 6 as a revolutionary episode, in large part because it was carried out by someone who was already in power in order to maintain himself in power. The other thing is that it lacked the conviction of a revolutionary episode. It was a one-off event. I would say it was closer to a riot than a revolution. A siege implies some kind of commitment to going out and staying out there no matter what, to achieve regime change.

In the case of January 6, they smashed up Congress and then they went home. Others have called January 6, I think correctly, a “self-coup,” an autogolpe as it’s known in Latin America. This is when a leader tries to perpetuate themselves in power by carrying out a coup or uprising against their own government, essentially in order to seize permanent control over that government.

Luckily, the self-coup on January 6 was foiled. The same with Bolsonaro’s attempt in Brazil. It lacked the commitment and conviction of a mass siege of power, resembled a riot more than a revolution, and was a self-coup aimed at perpetuating Bolsonaro in power rather than a mass revolt against the regime.

It could happen. Revolution is a phenomenon that can be used by all different social forces — it’s not the sole property of the Left or of liberals. It’s quite possible.