„Violence has its own time-zone“

This interview picks up and extends the discussion of the book Explosive Conflict. Time-Dynamics of Violence (2022) between its author, Randall Collins, and the sociologist Eddie Hartmann, who reviewed the book on Soziopolis.

Randall, in your path-breaking book Violence: A Micro-sociological Theory from 2008 you announced that a second volume will focus on issues of institutionalized violence on a macro-sociological scale. This prospect stirred speculations about how your theoretical approach to violence could be extended when dealing with phenomena such as revolutions, widespread protest violence or wars. Can you tell me why it took so long in the end for the book to be published?

The problem that took me the longest and that represented the greatest difficulty to research was the high-tech revolution in the military. Most of the literature is speculative, consisting of best-case scenarios (i.e. advertisements) for all the things high-tech can do. It promises to remove the human (i.e. sociological) element in combat, by replacing human senses and emotions with technical instruments and data-driven calculations.

But best-case scenarios (just like worst-case scenarios) do not describe the actual experience of war. I spent several years conducting interviews – thanks to Anthony King especially, I was able to meet with officers who had fought in Iraq or Afghanistan. Talking to these specialists, I found out that high-tech weapons are powerful when available, but much of the time they are being repaired, considered too expensive to waste, or have run out of fuel or ammunition and need to go back to resupply.

What did you draw from that observation?

The longest section of my book consists of four chapters on the effects of the high-tech revolution. My conclusion is that computerized high-tech is still very far from making such militaries invincible. Sociology remains crucial in understanding violence, even on such a scale, because human beings still interact with a technology which sometimes works extremely well and sometimes causes a lot of delays and frustration. In the end, I was able to develop a theory of terrorism as itself a result of high-tech war; and to show how the more serious dangers of cyberwar come from increasing reliance on computerization. It took me a long time to put it all together.

And the second reason for the delay of the second volume is that I was trying to come up with a new idea. I didn’t want the book to be an encyclopedia of existing research on macro-violence, a kind of textbook saying, for instance, Michael Mann says this or that about large-scale war. My ambition was to say something new about areas in which very good work has been done in the past.

Did that ambition lead to any new findings or insights in particular?

It led me to the concept of time-dynamics as the central theme of my new book. Instead of writing about the causes of wars, revolutions, or violent movements (topics that have been extensively studied), I focus on the course they take. My observation is that once a conflict comes to the brink of violence, it enters a new time-zone, so to speak. What caused the conflict in the first place, becomes a separate question from whether the conflict will be long or short, who will win or lose or if it will reach stalemate, and how destructive it will be. It becomes a matter of turning points within the time-zone of violence, sometimes redrawing the map of the world, sometimes leaving things much the same as they were before. Within a time-zone of 3 months, conflicts tend to be popular and generate enthusiasm; but if they go on longer without victory, they lead to divisions between one faction favoring peace and another pressing on to continue no matter how high the casualties and the cost.

Many conflicts abort at the outset, like the Trump supporters occupying the US Congress for six hours on January 6, 2021; others turn into civil wars that last for years. What makes the difference? Comparing different revolutions shows that the most successful ones are rapid because the regime becomes emotionally demoralized. The classic case here is 1848, where the French king resigned but nobody else in the royal family wanted to take over, and suddenly the revolution succeeded. Revolutions of this sort have relatively low casualties. The opposite is the case when both sides muster their forces, and especially when there are two mass movements opposing each other, each with a separate rallying-place (e.g. the failure of the Egyptian revolution in the 2011). These multi-sided uprisings become quite bloody and can go on for a very long time (the Syrian “Arab Spring” shows the prolonged time-dynamic at its worst).

Does the war in Ukraine fit the pattern too?

It is a more recent example of it, indeed. Putin apparently thought that it would be a situation like the US withdrawal from Afghanistan in 2020, where the government fled the country in panic, allowing a rapid take-over. But Zelenskyy did not panic; the US-equipped troops remained loyal and put up strong resistance against the advance on Kyiv in the early days of the invasion. As the Russians anticipated a quick victory, they did not provide sufficient logistic resources for their troops, and the emotional swing happened within weeks – it was the Russians who became demoralized, whilst the Ukrainians remained resolute. Both sides were well-equipped with weaponry, but what made the difference was the dynamics leading to the emotional turning point.

Then comes a second phase of time-dynamics. Failing in a quick victory – on either side – the Ukraine war turned into a long attrition war. The conflict moves into a stretched-out time zone where the extent of material forces each side can mobilize and their willingness to accept a great deal of casualties become the critical issues. During such times, macro-politics shifts too, changing world alignments of allies and opponents. It also becomes a lesson or experiment in high-tech war.

What does that mean for the further course of the war?

High-tech wars are a particularly expensive kind of attrition war – unless the side with high-tech superiority gets a decisive victory right at the beginning. The war in Ukraine turned into a high-tech war with mostly US-supplied weaponry on one side vs. Russian high-tech on the other (supported by new suppliers like Turkey and Iran). And although Western weapons are more advanced, we see the patterns which I predicted based on the Iraq and Afghanistan wars: slow delivery of the most advanced Western weapons because of their extremely high cost, high need of repairs and training, tremendous demands for ammunition and other supplies. It then becomes a war of world industries and economies, and meanwhile emotional enthusiasm wears down. The Counter-escalation/De-escalation model, as spelled out in my book, is applicable to the Ukraine war, as it was to previous wars.

The composition of Explosive Conflict differs from the micro-sociology book from 2008. To me, it seems a bit more like a puzzle of different journal papers, whereas the micro-sociology book reads more like a single piece.

Yes. Also, I agree that this volume is not as good as the first one, as you argued in your review essay. The failure of my second volume on violence to live up to the first fits a general pattern among multi-volume series. When authors write books in long series, usually it’s the first volume that’s the famous one.

So publishing books also has its own time-dynamics?

For instance, in the case of Wallerstein’s The Modern World-System, it’s the first volume that made him famous. I think there is better material in some of Wallerstein’s later volumes, but it’s the groundbreaking first volume that gets the fame. We also find this pattern with Marx’s Das Kapital, or Gibbon’s Decline and Fall of the Roman Empire, and even in Proust’s À la recherche du temps perdu, where the most famous part is its first page. But there are examples where the long-term focus on a topic keeps providing new discoveries; Michael Mann’s multiple volumes had some later success, such as his book The Dark Side of Democracy, which is an offshoot of his work on the later volumes.

There is a significant shift in focus and scope between your first and second book on violence. How would you describe the progression in between the two?

There is some important continuity between the theoretical strategy of both books; focusing on time-dynamics is in line with the finding that the time-patterns of emotions drive micro-conflict, rather than sheer physical force. And both books make the same gestalt-switch, away from “root causes” and into the processes that, once started, have their own patterns irrespective of who the conflict is between and what happened before it started.

The first book tries to show that what happens in the micro situation is crucial. It answers a criticism raised against the ethnomethodologists, which was to say, “What does it matter if you record conversations and look closely at face-to-face interaction? Aren’t you just showing that everyday life is routine, and that nothing new happens there?” But micro-sociological research on violence shows that there are turning points that cause or prevent violence, and there are time-dynamics for how long a confrontation can last before it becomes boring.

To what extent did you succeed in analyzing those in your new book?

As far as the newer book is concerned, I have illustrated time-dynamics in a number of cases. But we haven’t yet studied time-dynamics well enough to show not just how long it takes to mobilize and de-mobilize various kinds of conflicts; we need more comparisons to find the theoretical mechanisms that explain the variability – the conditions that determine how long violence will go on and with what intensity and damage. As you also recognize in your essay, this area of research hasn’t yet been explored thoroughly enough as to announce definite new discoveries.

As I mentioned at the beginning, the biggest problem on the conceptual frontier is cyber warfare, taking place not just on the battlefield but penetrating into everyday life; not just hacking individual victims but possibly destroying an entire economy, its banking system and public trust, since everyone is relying on electronic communications. Here again the self-promoting capitalist high-tech industries are an obstacle to identifying the crucial patterns. But we will learn more on this issue from Anthony King’s new research on computerized warfare.

From a theoretical perspective, the problem of time-dynamics seems to be the main research focus of Explosive Conflict, whereas the micro-sociology book focuses on the question of how social actors can overcome the quasi-natural barrier of confrontational tension and fear so that violence can occur in interactions in the first place.

Yes, that’s right. But Explosive Conflict also provides more time-sensitive case studies. For example, let’s take the chapter about nationalist movements at the time of the Arab Spring. For years, I have been clipping out newspaper articles about protest demonstrations and revolutionary uprisings, documenting the day-to-day timeline as long as the demos continue. The Arab Spring uprising happened in 2011, when the Internet was just beginning to become accessible to a mass audience in the Middle East. Many people were saying: “The internet changes everything; now you can’t stop the revolution because people learn about it on the internet and alert other people to join in.” That is partly true; it is a typical time-pattern that during the early period of mobilization people get enthusiastic about the sheer size of their movement. But to overthrow a regime, there must be a point of physical confrontation. Huge crowds protested at Tahrir Square, chanting for the soldiers to join them, which they did on day 7, Mubarak was forced to resign on the 18th day. The pattern is similar to the February 1917 revolution in Petrograd, when Russian soldiers refused to shoot protestors; and the Czar abdicated on day 7.

What does the time-dynamics dimension add to your earlier work?

These patterns echoed my previous research on how many people put out flags after the 9/11 bombings: there was strong and widespread solidarity for about three months, but then social dissensions reappeared. Similarly, in Egypt, Christians, Muslims, and secularists were initially united in the protest movement and women were welcome at the demos. But after seven weeks they started being attacked by men, Muslims and Christians started demonstrating separately in different parts of the city. This illustrates an important theoretical point: if there are two different centers of protest, it becomes impossible to reach a tipping point. One huge demonstration creates the feeling that the People, as a totality, are exercising their will. But if there are counter-demonstrations, who counts as “the People” is a matter of contention; the idealism and enthusiasm of the early period of unity turns into mutual hatred. Within six months, Egypt reverted to authoritarian rule; rival demonstrations signal that the conflict is endless, people become tired of participating, and willing to accept a dictator to stop the fighting.

Don’t you think that such a strong focus on time-dynamics comes at the risk of missing crucial aspects of the whole story? For instance, if we look at the disturbing events of January 6, 2021 on Capitol Hill, shouldn’t we also focus on the cultural cleavages of the Trump years? I mean, why did all these people go there in the first place? Do you think your theory could also address questions about the sociopolitical situation in which these events occur in the first place?

Yes, I believe that the main driver of the severe polarization in the USA is social media. When social media first became omnipresent, many people proclaimed that all barriers would break down as everybody would communicate with everyone else. This early expression of naive enthusiasm is itself a phase of time-dynamics. From a social psychology point of view, it is impossible to communicate with everybody at the same time; it would be an extreme form of cognitive overload. Today it’s widely recognized that the world of social media is very balkanized. There is no single audience, but many different groups that communicate only among themselves. Social media helped foster the many extremist movements in the United States (also true elsewhere), but I’m not so sure that the underlying issues have changed. What’s different is that it is much easier to mobilize people to protest and attack via the internet. This happened for several issues that peaked during the Trump years.

Could you spell those out?

I don’t think Trump is so unusual as a politician; he’s very opportunistic and demagogic, in the sense that he’s skilled at stirring up crowds of his own followers. We’ve seen people like that in the past, such as John F. Kennedy (whose sexual affairs were kept quiet in an era when the new media were much more centralized). The reason that Trump crystallized so much hostility was the MeToo movement on the internet. Accusations of sexual aggression against men, especially against well-known men, would go viral, i.e. generate a tremendous burst of attention, and trigger both support for the accused and #MeToo responses adding further accusations. Complaints ranged from rape to microaggressions. Allegations on the internet prompted rapid responses and often resulted in having the accused fired from his job, rather than going through the lengthy procedures of a court of law. The MeToo movement grew and became further influential, because people realized that voicing accusations of sexual misconduct against public figures may well have immediate consequences – in the case of a television comedian facing such allegations, for instance, the board of directors would feel pressured to take swift action and fire him from his show. The mechanism at play here is what scandal theorists (such as Ari Adut, referred to below) call a secondary scandal: if you don’t punish the scandal, then you, too, become part of it.

Can this mechanism fail to work in some cases?

A few weeks before the 2016 election, the Democrats publicized a recording of Trump talking privately to another media personality about how easy it was to sexually abuse women who work in their TV business. The Democrats expected this scandal to win them the election; and as the scandal went viral, many Republicans were ready to bail out on Trump. But Trump refused to be emotionally intimidated into withdrawing; and in a series of face-to-face meetings convinced his Republican supporters that they could ride out the scandal if they did not join it. And in fact, they did – they broke the emotional dynamics of the spreading scandal. The Democrats and the MeToo movement were outraged that they couldn’t get the secondary scandal going: this target wouldn’t budge. The same thing happened again when Trump nominated Brett Kavanaugh as Supreme Court Justice. There was a serious MeToo accusation against him from many years earlier; the US Senate held public hearings about the accusations, but Republicans also mounted a defense by defining it as a politically motivated tactic. The failure left Democrats and feminists feeling morally outraged that the scandal ritual itself had been violated.

Do you see similar dynamics at play when looking at other contemporary social movements?

A similar dynamic of polarization happened with the Black Lives Matter movement. It started with protests against police killing a Black man; it was not a new issue or a new cause, even in recent years, but it became an especially large, nation-wide protest since it happened just as the Covid restrictions on public gatherings were being challenged in the summer of 2020. It then took up the MeToo pattern of accusing well-known figures on the Internet who allegedly used racist expressions in the past. The mechanism of secondary scandal became operative, and numerous well-known entertainers, athletes and business executives were fired by their employers who would otherwise be accused of doing nothing about racism. In schools and universities, the movement contributed to emphasize matters of identity such as the privileges of being white over questions of individuals’ beliefs and actions, culminating in the doctrine that everyone who is white is racist.

Since the US are deemed to have always been racist and to celebrate racist historical figures to this day, local political pressure was mobilized in cities to tear down statues and change names of streets and schools named after figures such as Christopher Columbus – because he colonized America – and even Abraham Lincoln, because he fought in the Indian wars. These attacks gave rise to counter-escalation on the other side; attacking national monuments and symbolic heroes of the past generates its own moral outrage (like Sunnis burning down Shi’ite mosques in Iraq). In sum, the USA has gone through a series of ritualistic outbursts, mobilized by the Internet; each spreading moral fervor, applying emotional pressure through the mechanism of secondary scandals. It is in such a context that many people do not trust the official news and believe an election was stolen; and others believe that everyone but themselves are racists.

Having all this in mind, what is your outlook on the time-dynamics of Trumpism in view of the US presidential race in 2024? Do you think he will be able to re-establish the same kind of emotional energy among the audiences and to mobilize a similar kind of movement once again?

One way to estimate this sociologically is to ask how long charisma lasts. In my book Charisma: Micro-sociology of Power and Influence, I tally how many years various people were charismatic. Jeanne d’Arc: two years; Jesus Christ: two years. Gorbachev was a charismatic reform leader for about two years and then lost prestige completely. At the longest end of the distribution, Julius Caesar, Napoleon and Hitler were charismatic leaders of their followers for about 15 years. Almost all charismatic figures come to a bad end, among other reasons because they make many enemies. The ones who lasted the longest were at the head of military or authoritarian regimes. The only one who lost his charisma and then got it back was Napoleon: defeated in 1814, he was forced to abdicate and exiled to the little island of Elba; a year later he came back to rally his followers and was defeated at Waterloo after three-and-a-half months. Exiled permanently to St. Helena, he finally lost his charismatic style and died 6 years later.

I would say Trump had a confident, charismatic style from the election campaign of 2016 until the outburst of Covid in 2020; since then, he seems like a nostalgic figure, harking back to past events instead of pushing forward as charismatic leaders do. According to Max Weber charisma is unstable, but what he meant by that is that charisma can’t be institutionalized and transferred to the succeeding generation. I would add that charisma is intrinsically unstable in itself; it requires being at the center of attention of a growing movement, feeding on emotionally engrossing interaction rituals. That is a difficult formula to maintain for very long.

I would also like to ask you about the fatal beating of Tyre Nichols by a group of Memphis police officers on January 7, 2023, which you analyzed in your article “Three Police Tactics Led to Memphis Killing of Tyre Nichols”. I guess a lot of people struggle to make sense of what happened here because the police officers who beat this young Black man to death were also Black. Hence, there’s a wide-ranging discussion on whether the underlying problem can still be racism in that case. For instance, some of the more liberal newspapers in the US talked about institutional racism. What is your point of view here?

Focusing exclusively on racism is a blinder, like a gestalt that renders one unable to see any other causal factors. Black people get stopped by the police in the United States about three times more often than white people. Altogether, there are about 20 million police traffic stops in a year. Of these, about 5 million involve Black people. The number of persons killed in a police stop is about 100 per year. That tells us that out of 5 million stops of Black people, 4,999,900 (or more, since those who get shot are not all Black) do not get killed by the police. Neither being Black nor White is a good predictor of how likely you are to get shot by the police, since it is an extremely small fraction in any case. We must look for other conditions – the surrounding events or the micro-behavior in the situation – for an explanation of what happens in a particular encounter.

These situational and micro-interactional conditions operate whether the police are Black, White, or any other race. In my blog I pointed out three of these: (1) Police anonymity: unmarked cars, no uniforms, wearing hoods, an ominous and threatening self-presentation. (2) Large numbers of officers on the scene – the crowd-multiplier of violence. (3) Rumor transmission among police, radio traffic, phone operators and dispatchers, amplifying false beliefs about the dangerousness of the suspect. Racism can play a role in creating suspicion or tension, but it is neither a necessary nor a sufficient condition for police violence.

What other factors need to be considered?

There is very good research on gangs and violence in the Black community. In my opinion, the best analysis comes from Elijah Anderson, my former colleague at the University of Pennsylvania, who uses Goffman’s concept of the “code of the street”. It is a cultural code of acting like a tough guy, ready to use violence, because of the belief that you can’t rely on the police to protect you. It is a Goffmanian presentation of self, since most Black people, even in poor communities, are not criminals; but they can imitate that style when they feel threatened. Given the existence of the street code, it doesn’t surprise me that Black cops beat up Black suspects. It’s much the same code as the one gang members operate by. In addition, there is the evidence of statistics as well as ethnographies that most violence on Black people is committed by other Black people. The hotspots for frequent violence are typically in Black neighborhoods. Black police are aware of all this; and they act by the street code when they think they are being disrespected. Solving the problem is not just a matter of changing the color of the police; it requires better training, focusing on awareness of the micro-sociological conditions that unleash violence.

As far as I know, the spot where Tyre Nichols was killed was located in an area of Memphis with a particularly high crime rate. The police officers were part of a SWAT team, a special force within Memphis police.

Violence is a serious problem, in all its forms. But simplistic ideologies – that the basic problem is racism or that violence is fundamentally the culture of white men – do nothing to help solve the problem. Historically, there is violence in almost all societies; and tactics do not differ much by skin color when people acquire the weapons. Nor is it simply gender. When females get the opportunity to engage in conflict, wield power, or have weapons, they act the same as men; non-white people act the same in physical confrontations, social movements, and when exercising police authority or political power, as white people. The dynamics of conflict and violence are universal, they can be found in tribal peoples as far back as we can observe, as well as in video evidence from across the world today. As political leaders, women act no differently than men. Here my conclusions may be pessimistic; there will be no utopian future if only we get rid of chauvinistic white men. But some points allow us to be optimistic; both micro-patterns and macro-patterns show that conflicts have processes that bring them to an end. My finding that rape can be successfully resisted (Chapter 13 in Explosive Conflict) shows the victory of micro-situational emotions over sheer physical strength. The “white men” ideology doesn’t point to a solution, but the study of micro-interaction does.

Randall, thank you very much for your time and the interesting conversation.

I have to thank you as well! Thanks for advancing the project of how to do research on violence. It’s nice to be in an academic field where things are moving forward. For example, I can mention two important books which are coming out in the future. One is Michael Mann’s book On Wars, to be published soon by Yale University Press. In his typical style, Michael Mann assembles the best information about wars going back to the Aztecs and the Babylonians and up to the Ukrainian war. He concludes that wars almost never achieve the ends that the people who started them expected, whether seeking material aggrandizement or imposing an altruistic ideal on the world. Nevertheless, people keep on starting wars. Mann answers the puzzle why over and over again we do something that doesn’t work, and he makes a convincing case against the belief that wars solve things.

The other book is by Ari Adut, who wrote the most important book on the dynamics of scandals (On Scandal, 2008). His current book is about the dynamics of the Reign of Terror in the French Revolution. We tend to believe the Reign of Terror started because the Jacobins were so radical, but Adut shows that almost every politician during the early years of the Revolution was a constitutional monarchist. The spiral of radicalization and paranoia resulted from creating a structure where all the power was concentrated in the legislature. There were no longer any independent courts – all courts were set up specifically by the legislature, for trials of its enemies. There no longer was an independent executive – only committees appointed by the legislature. Thus, when the assembly got into disagreements, they purged the weaker party; this happened repeatedly, creating an atmosphere of distrust and fear that was almost impossible to break. If you didn’t chop off somebody’s head first, they were going to chop off yours. The problem wasn’t ideology or fanaticism, but overconcentration of power in one decision-making body. Adut’s analysis makes me worry about movements in the United States who want to change our electoral system, concentrate all power in Congress, and manipulate the number of judges in the Supreme Court so that it always aligns with the party in power. That pathway could turn the US into France in 1792.

To conclude my point: the sociology of violence and conflict has made important findings, not only theoretical advances but also practical advice on our greatest sources of trouble. It is a field of sociology that has many constructive insights to offer.

So we will have two more books to discuss on Soziopolis soon.