In conversation with the American Sociological Association President: Randall Collins on emotions, violence, and interactionist sociology.

Walby, Kevin ; Spencer, Dale

Randall Collins is Professor of Sociology at the University of Pennsylvania and is the President of the American Sociological Association. In the following interview, Collins discusses the sociology of emotions, the interactionist tradition, and violence. The discussion situates Collins' contributions as part of an intellectual trajectory that incorporates elements of critical sociology and the micro-sociology of interaction.

Randall Collins est professeur de sociologie a l'University of Pennsylvania et President de l'American Sociological Association. Dans cette entrevue, R. Collins parle de la sociologie des emotions, de la tradition interactionniste ainsi que de la violence. L'entrevue permet de situer les contributions de Collins dans le developpement contemporain de la sociologie critique et la microsociologie interactionniste.

RANDALL COLLINS IS ONE OF THE most renowned American sociologists alive today. His works span the gamut of sociological inquiry, from macro- and systems-level considerations to detailed analysis of interactional processes. Some of Professor Collins' key texts include Conflict Theory (1974), The Sociology of Philosophies (1998), and Macrohistory: Essays in the Sociology of the Long Run (1999). He is author or editor of 17 books. He has written 150 articles and chapters, many of which are translated into several languages. Professor Collins' two most recent books--Interaction Ritual Chains (2004) and Violence: A Micro-Sociological Theory (2008), both with Princeton University Press--focus principally on issues related to interaction, emotions, and corporeality.

The following interview was conducted as part of the Emotions Matter workshop May 8 to 9, 2009 at Carleton University, which was generously funded by the Social Sciences and Humanities Research Council of Canada. In the first half of this interview, Professor Collins comments on issues concerning writing, the use of metaphor, and his own intellectual biography. He also opens up the lives and contributions of Goffman and Blumer, offering historical insight concerning the "backstage" of symbolic interactionism. Toward the end of the interview, Professor Collins discusses emotions and violence more specifically, expanding upon two of his key concepts: emotional entrainment and confrontational tension. Randall Collins is President of the American Sociological Association.

When did you become interested in the topic of emotions?

RC: Back in my undergraduate days at Harvard in the early 1960s, when I was immersed in Freud and Piaget. In fact, I went to graduate school in psychology at Stanford. But Piaget had not caught on in the United States at that time, so I decided to switch to sociology. I remember trying to check out what kinds of psychology of emotion existed at that time. There was very little.

And how did the interest in violence enter your focus?

RC: I was part of the generation of young sociologists who broke with functionalist theory and moved toward conflict theory. In my case I was working with a mixture of Marx and Weber. Conflict theory at the time implied interpreting Max Weber in a left wing or conflict direction. I published a book in 1974 called Conflict Theory, but a few years later I realized there was not any conflict in the book. It was all about structures of domination. So I started looking at the conflict literature and teaching a class on conflict. Military violence was the first place where there was some good empirical material. I gradually spread out from there.

Unlike most other sociologists, you had a career in creative writing beforehand. How has that experience with creative writing influenced your sociological writing?

RC: Quite early I was interested in being a novelist. I wrote a couple of aborted novels, then in about the middle of the 1970s I became very disillusioned with my department at the University of California San Diego and decided to quit the academic world. I had just written a critique of educational credentialing as the major contemporary form of stratification. I thought to be honest I should quit this racket. So instead I made a living as a professional writer. One thing I learned from that time was not to screw around while writing. I realized that if you were going to support yourself from writing you could not afford the luxury of writer's block or writing incoherently. It really sharpened my techniques of writing.

On this topic of writing and vocabulary, can you say more about the use of metaphor in sociological writing?

RC: This has fairly deep theoretical significance. Metaphors are needed because of the micro-macro relation. In our experience, everything comes to us in the micro, in the here-and-now of some particular situation. This is true even when we are seeking information about larger patterns. When we ask somebody about their social class, their occupation or their education, we get a brief cryptic answer, but it is a mistake to take that word--typically a noun, or a number--as if it were an entity. To depict one's occupation in a word is to compressa huge number of experiences in situations, and to gloss over the way in which that person's experiences are related to the chains of experiences of the persons they have interacted with. Class, occupation, education--even age or race--are actually metaphorical transformations of social processes which play out over large numbers of microsituations. My old teacher Herbert Blumer used to challenge us by saying: "What are you actually talking about when you say structures, or social class? Show me social class, where do you see it?" Of course we need ways to summarize patterns that operate in chains of situations. In the case of class, we have metaphors like strata--things are said to be higher or lower, although in fact that is not usually what class interactions look like if we actually see them in a situation. Unlike the early ethnomethodologists, who dismissed all macrostructure as rhetorical gloss and concentrated instead on the patterns of situated cognition, I am quite willing to admit that macropatterns exist--as chains of microinteraction. But we need better ways of getting at their real character and especially their dynamics. The question I am puzzling over is whether metaphors help or hurt. If you get rid of the metaphor of higher or lower, what do you replace it with? Do you get something analytically useful out of changing vocabulary? In the case of violence, it was helpful to get rid of the conventional metaphors, which badly got in the way of seeing what the phenomenon actually is.

Both Goffman and Blumer interpret Mead. Mead does not show up a lot in Goffman, here and there with references. Blumer wrote a whole book about Mead, and many papers. Where do you think the key turning points are in how Goffman and Blumer interpret Mead? How do Goffman and Blumer differ in their use of Mead?

RC: Blumer and Goffman were two of the big stars of the Berkeley department of the 1960s, which had a lot of important people working in it, including macrosociologists. But Goffman and Blumer were the ones who inspired people to look deeply at the micro. Although Blumer was pretty respectful of Erving Goffman, it was not true the other way around. Often Goffman was a wise guy, ironic, sardonic, puncturing people, whereas Blumer was an old-fashioned mid-western gentleman. Also Goffman did not consider himself a symbolic interactionist, even though he had come from Chicago and had been there under the Blumer regime. Goffman had studied with a British social anthropologist when he was at the University of Toronto and although I do not think he had crystallized his micro position at that time, later he liked to consider himself as doing a version of Durkheimian ritual analysis but on new materials. I think the turning point did happen at Chicago. Everett Hughes, who was a leader of ethnographic research in the symbolic interactionist camp at Chicago, told me that Goffman arrived from Toronto as an arrogant young man and in his early phase was very Freudian. This probably did not last long. You can see that references to Freud are absent from Goffman's writings even though he was quite interested in mental illness. Goffman's move was to investigate mental illness as a form of social activity rather than something one dug up from the past in psychoanalytic sessions. I think under the influence of the ethnographic emphasis at the Chicago School, Goffman started trying to see what he could find in everyday life interactions, but what he was looking for, unlike Blumer, was not a version of Mead's theory. Instead Goffman was looking for a version of Durkheimian ritualism. The other piece of the puzzle is W. Lloyd Warner, trained in British social anthropology, who worked on Aboriginal society in Australia but had come to the United States and did one of the first famous community studies of stratification, in Newburyport, MA. Warner interpreted social classes as if they were tribes with distinctive rituals. Goffman was his research assistant. Goffman's very first paper is called "Symbols of Class Status," and there you can see the combination of Durkeimian ritualism and the empirical topics of American sociology.

In memoriam and tribute to Goffman, you once wrote that you would have liked to see Goffman write a book about sex. In your own book Interaction Ritual Chains (2004) you have written about sex. How does what you have written about sex and interaction rituals differ from a more conventional Goffmanian account?

RC: I once made this same comment to a friend, another sociologist, who had been at Berkeley with Goffman that I wished Goffman had written about sex because it obviously can be analyzed from his perspective. For instance, sex has front stages and back stages. My friend said that he did not think Goffman had enough sexual experience to write such a book. You never know. Goffman was a very private person. Goffman says very little about sex but there is a footnote I think in The Presentation of Self in Everyday Life (1959) where he raises the question of whether there is an ultimate self behind the backstage, and he says some people think your sexual self is the ultimate backstage. Then he discourages this reading by saying that sex is a performance and that in Italy there is a saying that sex is the poor man's opera. My early thinking was, how do you translate Freud into microinteractions? Ina way, Freud's "conscious" and "repressed" map onto the front stage and backstage. Later I developed a more explicit analysis of what makes rituals work, which led to seeing the various kinds of sexual activities themselves as rituals that produce different degrees of bodily and emotional entrainment. There are a number of puzzles. Most naturalistic or biological theories of sex have a great deal of trouble explaining almost anything except vaginal heterosexual intercourse. Simple things like kissing, or more elaborate acts like oral sex, are very difficult to explain in a biological way. So you have to start thinking about various sexual practices as social techniques to get people mutually entrained. Sexual practices are very intense forms of interaction rituals.

On the issue of sex and emotions, can you explain the importance of research that starts from analyzing interactional process and sequences?

RC: There is a lot of very good empirical work in the sociology of sexuality in recent years, but it is still not close enough to its own topic. It talks around it. I am not very happy with generalized concepts of masculinity or hegemony because that is like where I was in conflict theory when I was still into Marx and Weber without closely looking at anybody in conflict with anybody else. A lot of the sociology of sex is about what it is like to have a particular kind of erotic career, without getting into the microprocesses of what people actually do. This is a certain vestige of the Victorian veil of secrecy, perhaps, although we are pretty intrepid these days so I think it is just that people are not thinking micro enough. Everything happens second by second in little sequences and that must be where the basic reality is. If you apply the interaction ritual model to erotic interactions, erotic interaction rituals that are successful produce feelings of solidarity, a symbolic halo and particular kinds of sacred objects--bodily objects--that people carry around in memory and which they orient themselves to subsequently. Successful rituals produce emotional energy, which is an individual's confidence and initiative toward particular kinds of activities. Sexual desire or sexual drive is not primordial but is shaped by experience in erotic interaction rituals, especially via which rituals succeed or fall, which are energy gainers or energy drainers. What in abstract terms is called sexual proclivity is to develop a specific emotional energy about one's sexual interactions with others. Microerotic experiences must be the crucible of peoples' sexual lives.

What many people find interesting about your approach toward emotions is that you begin with interactional sequences but then there is a link up to macroprocesses. There is Jack Barbalet whose work is very similar. But some have indicated that he comes from the opposite direction, starting with something greater than interactional sequences. What do you think of this approach?

RC: I am in favor of what Barbalet is doing. My own approach is to push much more on the micro and the empirical side. There is a question here of starting from top and bottom and seeing where they come together.

But if one does not start with the sequences of interaction and instead starts from a different level of analysis, is there not a potential problem with this conceptualization?

RC: I think we can cobble it together as long as we can make some movement among the levels. My take on this terminology is somewhat different. Entities that are middle-sized, like social movements, have an emotional atmosphere that shifts over time. The key to social movements is whether they can chain enough high-intensity group rituals together so they have enough rallies or sometimes battles with opponents to keep the movement going. There are interesting ways this can be achieved with different emotions. My former student Erika Summers-Effler has worked on movements, like the antideath penalty movement, which do not have much success--so how do they keep themselves going? At one point she quotes a movement leader who says something like "our energy is getting down, so I think it is time to get arrested." A lot of the techniques of these middle-sized organizations are about establishing and trying to keep up an emotional atmosphere.

In your new book Violence (2008) you argue violence is not easy. It must be achieved through interactional processes. Can you explain a bit more what the emotional dynamics of achieving violence are?

RC: This is an aspect that surprised me as I went along. The initial thrust of my conflict theory was to try and make things more vivid and material and bodily, so I was looking for more violent conflict. I had become very critical of the theory of the state as this institution that had legitimacy, as if this were a matter of course. Instead I tried to push Weber in a more dynamic direction, conceptualizing the state as the institution that attempts to monopolize violence, its legitimacy coming from the extent of success in its monopolizing activities. But as I got more and more into it, I realized that if you look at specific violent situations there is much more posturing and bluster than actual violence. So that is a first consideration, that people are not very competent at violence. Furthermore, there is stratification in violence, between those who are good at it and those who are not.

But when I pushed it further I realized that actual physical violence itself largely hinges on the prior establishment of emotional dominance. Microtechniques of emotional interaction are crucial in whether violence is successful or not, and indeed in whether it will happen or abort. Sometimes successful violence has a very rapid onset and a strong temporal rhythm-for instance, a successful armed robber has a technique of finding when is exactly the right moment to pull out the gun and threaten others. I am certain now that microtechniques are crucial in almost every form of violence. If you observe or examine videos of people who argue and sometimes have fist fights, you always see the process of attempting to establish emotional dominance. The turn of phrase from sports--"having the momentum"--is actually quite accurate. But it is a reciprocal and interactional process. When one side has the momentum the other side does not. Sports commentators do not think this through. They say the defense is getting tired because they have been on the field too long. But the offense has been on the field exactly the same time as the defense so there must be a process where one side is getting more energized and the other side is unenergized. It is more a matter of who is establishing the initiative, who is setting the rhythm in this situation.

Can you say a bit more about your concept of confrontational tension?

RC: This is the central concept of my theory of violence. It is an empirical discovery, but I do not want to claim originality because a U.S. Army psychologist named Dave Grossman has a somewhat similar concept. He argues that there is an innate fear of harming other people, which can be documented both behaviorally and physiologically, such as what happens to the heartbeat when people threaten each other. Instead of fear of harming others, I put the emphasis on the tension that appears when people go into action at cross-purposes. This tension permeates one's body; it does indeed drive up the heartbeat while you are pumping adrenalin and cortisol. Psychologists have called these the fight or flight hormones. But in an antagonistic social interaction, more can happen than fight or flight, and the third alternative is crucially important. If both sides are at the same level of tension the third alternative takes over and that is stalemate. Most confrontations, in fact, become stuck at the stalemate level. From a practical point of view, this is a good thing, because if we want to reduce violence the best way to do so is to get people to stick at the stalemate level. A related point is that someone who is good at fighting is someone who has developed techniques for using the other person's confrontational tension against them. That is the subtle technique of winning. It is not that good fighters have no confrontational tension but that they make the other side suffer from it more than they suffer from it themselves.

Perhaps I should add here that some philosophically oriented sociologists may object to the notion that there is such a thing as an empirical discovery. Of course one always approaches one's observations from a background of ongoing theoretical discourse, bur we are not locked into it and it is possible to discover things that we did not see before. Also, I think it is not worthwhile to rigidly operationalize everything beforehand. Good sociologists need to be pragmatists and let concepts emerge in interaction with all kinds of considerations, empirical and theoretical. Confrontational tension is a concept of that sort. It ties together many empirical findings and it does a lot of work in generating explanations of how many different types of violence operate, as I have tried to show in the topics taken up in Violence: A Micro-Sociological Theory. For instance, in the book I discuss why Bourdieu's concept of "symbolic violence" is really obfuscating, the kind of metaphor that leads us in the wrong direction by rhetorically muddling very different processes.

Can you explain how confrontational tension across microsituations leads up to an organizational level of analysis, using the military example?

RC: Every larger social entity is put together out of microencounters. From a conceptual standpoint it is generally more convenient for us to lump it all together and say this is the army, this is the battle, this is the business corporation, but indeed everything referred to in this way is happening as chains of face to face interaction or technologically mediated interaction. Pragmatically it may be useful to lump it together, and theory regarding organizations has made a number of important discoveries on that level of analysis. Is there anything more that we get as sociotogists from considering military organizations as chains of microinteractions? In the case of the army, the larger network exists primarily to get soldiers to some physical situation where they can attempt to threaten violence against some other group of soldiers. The larger structure does most of its work preliminary to the time when the fighting actually takes place. The moment of fighting is the time when confrontational tension emerges, because fighting has to go through the microlevel. In this respect, let me compare traditional or historical forms of combat and newer forms. Older armies, as described for instance in the accounts of Julius Caesar, operated by marching several thousand soldiers up to a point where they can come into direct contact with the front line of a similar force of opponents. The side which broke down least from confrontational tension would win the fight; the art of warfare was to try to keep from losing group coherence until the other side broke apart, at which point they could be killed because they were no longer resisting.

We still have fights like that. To some extent brawls and riots look like Caesar's legions when they entered the melee. But over time armies have developed long-distance weapons. Up until the 1860s weapons were inaccurate enough so that soldiers still had to get within a small number of meters or else the chance of hitting anyone was not very high. Fighting remained within the zone of face-to-face confrontational tension, and managing that emotional tension was the key to winning or losing.

But now our weapons have become fairly precise at a distance where you can barely see the enemy, or only on an electronic screen, and battles can be carried out from miles away. Does this mean the confrontational tension disappears? The research question is open. But as far as I can tell, it does not disappear. Even in a long-distance combat zone, soldiers using weapons still experience some confrontational tension. The most advanced militaries, notably the United States and the United Kingdom, are now attempting to control combat entirely by computers, using informational inputs from long-distance sensors. The aim is to take humans out of the loop because they are the fallible element. Since the time of Clausewitz military thinkers have had the idea of friction, or what in popular parlance is called the fog of war. On the microlevel, this friction or fog--notice the use of metaphors--translates into the emotional processes of confrontational tension and the way it propagates through the links of an army's human network. Today's military planners are optimistic that for the first time in history, armies can get rid of the fog or friction. If that is so, it would completely change the process of winning a war, since in traditional battle the more coherent organization--in traditional terms, the one with the higher morale in the confrontational situation itself--would beat the organization which is less able to cope with confrontational tension. But if warfare can be carried out entirely at a distance, or by long-distance controls, the confrontational tension disappears, and winning is entirely a matter of having superior technology.

I am attempting to analyze various aspects of this now. One of the things that complicate matters is that most wars right now are so-called asymmetrical wars, between a high-tech army on one side and guerrilla fighters on the other, who play mainly for political effects concerning civilians on both sides. But I am also looking at the situation where one high-tech military fights another high-tech military--would this just be a war of technological behemoths, or would social processes of emotion come back into play? I suspect the latter.

High-tech warfare may seem a somewhat marginal topic for sociologists, although if the United States fights China, for instance, in the world-system transition about 20 years from now, it could--frighteningly--come about. But my concern here is also in wider theory. The United States military is now attempting to create a truly Orwellian kind of organization, in which everyone is linked by remote surveillance technology and coordinated by computers. This type of organization might become the prototype for civilian organizations in societies of the future. Sociologists need to be aware of the issue and start thinking about what we can do about it.

Thanks to Nicolas Carrier and Justin Piche for help with the translation.

INTERVIEWERS: KEVIN WALBY AND DALE SPENCER

Carleton University

Kevin Walby, Department of Sociology and Anthropology, B750 Loeb Building, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario, Canada K1S 5B6. E-mail: walbymswresearch@gmail.com