It is impossible to say what type of scholar Bernard E. Harcourt is. He is Professor of Law and Faculty Director of Academic Affairs at the University of Chicago Law School. He is best known for his detailed econometric critique of the "broken windows theory"—the notion that an atmosphere of disorder created by graffiti, loitering, and prostitution encourages more violent crime. Harcourt has argued that this is in fact an illusion created by our desire for orderliness—an illusion that masks the deeper effects of surveillance and that goes unrecognized by the policy-makers who have championed it, such as former New York City mayor Rudolph Giuliani and his first police chief, William Bratton. At the same time, Harcourt is part ethnographer. He has explored the socio-cultural aspects of such legal issues as firearm registration, urban redevelopment, and gay and lesbian rights. Finally, a deep appreciation and enthusiasm for law and philosophy infuses his work. His students will tell you, some with admiration, others with frustration, that you are just as likely to read multiple-regression analyses as you are Michel Foucault in Harcourt's classes.

Harcourt's latest book, Language of the Gun: Youth, Crime, and Public Policy (Chicago 2006), neatly reflects these various facets. He interviewed thirty inmates of the Catalina Mountain School, a juvenile detention facility outside Tucson, Arizona housing minors who had repeatedly run afoul of the law. The interviews began simply: Harcourt placed three full-color photographs of handguns in front of the youths and asked them what they thought. As it turns out, at-risk youths know their guns and have a lot to say about them.

AS: Language of the Gun has striking ethnographic elements. It permits the Catalina Mountain youths to speak in their own voices. In this sense, it has a kinship with Philippe Bourgois' work in the Puerto Rican barrio, In Search of Respect, which you both praise and critique in your book. You describe, for instance, how gun carrying is not just about self-defense, but about aggressive, preemptive protection—about avoiding "getting dogged," "punked," or "jumped." These subtleties can be lost in the statistical analyses that seem to dominate public discourse. Where does ethnographic work fit in the public policy debate?

BH: It's true that ethnographic methods have taken a back seat to econometric studies in legal and public policy debates today. The reason is that ethnographic interpretations tend to be perceived as overly subjective—as too easily influenced by the preconceived ideas of the researcher. It's precisely to challenge this tendency that I wrote the book. The method I develop, which combines in-depth interviews and a free-associational method with quantitative analyses, seeks to render the interpretation of qualitative data more measured and objective. But beyond that, I'm convinced there's a false dichotomy between qualitative and quantitative research methods. There is today an illusion about the objectivity of econometric modeling. In the book, I spend a lot of time discussing how to interpret the Catalina interviews. I try to demonstrate that all empirical approaches and their corresponding methods—whether multiple-regression analysis, survey questions, or qualitative interviews—are based on specific assumptions about human behavior. They each adopt a discrete theory of human action. They each rest on a subjective belief that we act rationally and deliberately, or that we follow scripts unconsciously, or that we are determined by larger structural forces. My central point in the book is that we need to lay out our subjective choices about human action—I call these "ethical choices"—and defend them when we offer interpretations of our data, regardless of whether the data are quantitative or qualitative. In this sense, presenting ethnographic detail is no more subjective than interpreting a multiple-regression analysis.

AS: Language of the Gun is also a statistical work. You hand coded your interviews with the youths to find the distinctive meanings that the youths associate with guns, such as protection, danger, attraction, power, commodity, recreation, and jail. You then use correspondence analysis, a statistical technique that measures the connections between these different meanings and how they match up with the youths' backgrounds. Why did you choose this somewhat unusual method? What advantage does it have over more familiar approaches?

BH: I decided to use correspondence analysis precisely because it is the most rigorous and effective method to visually represent the statistical relationships between "social meaning" variables—here, between the meanings of guns for the Catalina youths. This book is a study of the symbolic dimensions of the gun as object. It's a semiotic of the gun—an attempt to capture, in a more measured way, the meanings of the gun in order to draw legal and public policy implications. It represents an effort to get serious about social meanings—a concept that has received a lot of attention, especially in the law and social norms movement born here at the University of Chicago in the early work of Larry Lessig, Dan Kahan, Tracey Meares, and Eric Posner. Correspondence analysis is a method used in Europe and Japan, and it's perfectly suited to the task because it allows the researcher to visually graph how meanings relate to each other in different social contexts. Pierre Bourdieu pioneered the method in his fascinating research on the social dimensions of taste and the academy. Using it here allowed me to map on two dimensions the symbolic realm of guns for the Catalina youths. That's the best way to begin to decipher the complex world of social meaning. 

AS: Through correspondence analysis, you find that the meanings associated with guns group into three "clusters" that you call "registers of gun talk." In the first, youths talk about guns as "dangerous yet attractive, necessary for aggressive, preemptive protection." In the second, they talk about selling guns or trading them for drugs and favors. In the third, guns are for hunting, target practice, and recreation. These clusters, you show, are closely linked with actual gun carrying and gang status. For example, youths in the action/protection cluster are more likely to carry a gun, be a gang member, and deal drugs. What in these findings surprised you? How should public policy reflect these distinctive approaches to guns? 

BH: You're getting at the most important contribution of the book: the effort to develop legal and public policy interventions that are specifically tailored to the different registers that the youths deploy. In other words, the attempt to think about these Catalina youths through the lens of language, rather then race, ethnicity, education, family background, or prior criminal history. The central insight here is that language reflects something important about how we think, reason, and act. Youths who talk about guns as a commodity, for instance, are more likely to be pursuing their goals in an instrumental way and, as a result, may be more amenable to rational choice interventions. In contrast, youths who talk about guns only in terms of aggressive, preemptive protection are less likely to be amenable to deterrence approaches. To these youths, guns are a life or death proposition filled with desire and attraction for the guns. Increasing the cost of carrying a gun will have little effect on them. With regard to these youths, a more promising avenue may be to develop practice-based approaches that will provide them with different scripts for how to resolve conflict, refocus their desires, and spend their time. In this sense, the larger theoretical contribution is that the language youths use to talk about guns can tell us something important about the way they think, desire, and act—which in turn can inform the kind of legal and policy interventions that might be most effective.

AS: You have often been critical in your scholarship of approaches to crime policy that rely too heavily on deterrence. In Language of the Gun, you criticize your colleague Steven Levitt for just such an error. However, your work with the Catalina youths is permeated with the deterrent effect of punitive gun laws. One youth states that, in a gunfight, "I just froze up, and I was just 'Oh man, please, God, please don't make me use this. I don't want to go to prison, really,'" and then explained that he no longer carried a gun because he feared the adult justice system. How did your experience with the Catalina youths affect your opinion of deterrent gun policy?

BH: You have to realize that for every youth who said "guns carry too much time" there were others who explicitly resisted the idea. I spoke with one youth, in fact, who told me that "I never think the police are gonna catch me. . . I know they'll catch me sooner or later, but I don't think that 'tonight, I'm gonna get caught.' Or I shouldn't have this gun because I think I might get caught. Because I just think that's kind of like jinxing myself." He didn't want to think about the cost of crime because the very thought might jinx him. As a result, it's crucial to avoid taking an all or nothing approach to deterrence. Instead, it's important to figure out which of the youths might be amenable to deterrence-based strategies and which will be immune. Again, it is in their language that we can begin to decipher this question. The key, though, is to take a more nuanced approach that acknowledges the possible role of deterrence in some, but not all cases. The larger project is to tailor the legal and policy interventions to each individual youth using the medium of the language of the gun.