Can discipline predict what one says about race and crime?

[julienlarregue]

Here is the answer: Painting the Discipline-line black. Psychology, Sociology, Criminology Departments and the Discussion of Race.

Abstract: While sociology and psychology are long-standing disciplines, criminology departments are a recent creation. Their development in the US can be traced back to the 1940s’ with strong ties with law enforcement. Since its inception, criminology has received numerous critiques on its legitimacy, the most radical critique labelling it a “State science”. This paper quantifies how criminologists, sociologists and psychologists study the question of race through 229 papers published between 1945 and 2014. I find that sociologists and psychologists have a scientific production that tend to support welfare policies slightly more than criminologists do. Criminologists tend to explain African American crime with factors that leave little room for anything else other than penal policies. Overall, my results support the critical theory and Bourdieusan claim that criminology is closer to state interests, but the gender and background analysis shows that researchers’ agency is still a major player for criminologists.


Please feel free to contact me if you have any question / comment / critic: [e-mail]

Best,

Julien Larregue
PhD candidate, Aix-Marseille Université, France

[julienlarregue]

If some of you are more familiar with SSRN: ssrn.com/abstract=2649876

[response]

Please don't take this personally, but I see some major issues with this paper.

- It starts by reviewing a lot of superficial statistics on crime, but misses the main literature it should be referencing. What this paper claims to be about is a sociology of science, but there is little to no discussion of that literature.
- It is incredibly thin on describing its coding schema. You need to detail how you coded the papers much more clearly. Additionally, some pretty outlandish claims are made. Claiming that people who find that crime is explained by "socioeconomic situation" do not necessarily imply that social policies could improve the situation. Likewise, claiming that psychology has an impact does not mean that the way is clear for the "custodial state" to dominate.
- Likewise, it does not follow at all from Abbott's statement that departmental structure is unchanged that departmental affiliation is then an accurate predictor of what discipline a scholar belongs to.
- Which brings me to the final, and major, problem with the paper. Just counting how many articles use this or that variable is meaningless. Articles have different visibilities, different audiences, and so on. The number of articles published on some topic is a poor measure of anything, especially if their visibility is different. To do the sort of study that you claim to make, you'd need far more sophisticated methods. It would have to include measures of visibility (e.g., citations), patterns of co-citations and so on. It may be that sociologists have engaged psychologists in debate and are trying to contradict psychological explanations with socio-economic ones (in which case, we'd see a significant pattern of one literature citing the other), or it may be that they have different audiences, and therefore focus on different things while being silent on each other (in which case the literature cited would be different).

This paper has a worthwhile goal. It is just very far from achieving it. I'd suggest taking a look at the significant literature on citation patterns and policy results that is out there, a lot of it published in places like Research Policy, to get a better sense of how to do this.

[julienlarregue]

Thank you very much for your comments, and don't worry, I don't take this personally.

I agree with you when you say that I don't discuss the sociology of science literature. The fact is I tried to adapt my paper to the journal I submitted it to (that does not mean it improves my work, I know...). Same answer for the coding schema, I could be much more precise, but it appears that method is not the main interest of this same journal.

Your last comment is very interesting. Actually, psychologists were in the debate before sociologists (in the beginning of the century). But still, the idea of different audiences is a very good one. I had began to look at citations patterns, but I didn't link it yet to the historical part of my analysis (development of criminology departments). I could definitely investigate this more deeply, thank you for your suggestion.