Canadian Journal of Sociology Online November-December 2007
It is no secret that there is no consensus in the social sciences on fundamental issues such as the agency-structure debate. In brief, we can identify three types of social theory: determinism (structure —> action) (Parsons, Althusser, etc.), co-determinism (structure <—> agency), (Archer, Mills, Giddens, etc.), and relationism (actor <—> actor = structure) (Elias, Mann, Emirbayer, etc.). This debate and other related issues have shaped the evolution of the social sciences since their emergence in the 19th century. For this reason, many social scientists might be attracted to Agency and Causation in the Human Sciences, especially because most of the texts are written by philosophers, some of whom have published many texts on this issue (e.g., (E.J. Lowe, T. O'Connor, and A. Mele). Eleven philosophers joined by three human scientists (including R. Boudon) should be able to make a significant contribution to this never-ending issue in relation to the human sciences. There is no doubt that philosophers can help us — human scientists — to move forward with this key problem. "Outsiders" can provide refreshing perspectives, questions, solutions, etc. that "insiders", limited by their own disciplinary habitus, can hardly imagine.
Upon first glance, the book is promising. Chapters titles refer to important debates and problems in human sciences, such as: "Making Something Happen — Where Causation and Agency Meet" (G. Keil), "Three Tasks for (Hard Interactionist) Dualists" (U. Meixner), and "Teleological Explanation, Emergence and Downward Causation" (A. Corradini). The editors state in the introduction that one of the key questions discussed in this book is:
"Is the causal model of the natural sciences sufficient to explain human actions or can we expect to gain explanatory advantage from the classical concept of agent causality? … As a matter of fact, the social sciences increasingly use the causal model of the natural science in order to explain human actions. Therefore, the contributors of the first chapters discuss the limits and boundaries of deterministic causation, and the problems connected with the application of 'causation' to human actions" (p. 10).
Unfortunately, my colleagues in human sciences will be disappointed by this book. If we do not take into account R. Boudon's chapter, there is not a single reference to any human scientist! Durkheim on the notion of "social fact"? Nothing. Comte on positivism? Marx on economic/historical laws? Weber on causation and "ideal-types"? Nothing. Any reference to relevant contemporary works such as the "structuration theory" of Giddens? None. The critical realist approaches of Bhaskar and Archer? None. Mouzelis? Sayer? At least Bourdieu? No, no, and no. Elias? Mann? Tilly? Latour? Forget it.
The surprised reader will get the impression that the authors either do not know anything about theories, approaches, etc. on agency and determinism in the human sciences, or they do know they exist but they think they do not deserve to be discussed or even mentioned. At best, the knowledge of the authors on "agency and causation in the human sciences" is desperately weak; at worst, the authors refuse to look outside of philosophy. Agency and Causation in Philosophy would have been a more accurate title.
Once again, a trans-disciplinary approach could be more than relevant, even necessary. There is no doubt that philosophers may make a significant contribution to "our" debates. But it would be imperative to talk about and criticize theories and concepts used in the human sciences from a philosophical perspective. In this respect, some philosophers could discover that the notion of freedom — which is important in many texts in this book — is not really relevant in social theory. In the social sciences, it is seen more as an abstract notion without any empirical relevance. Therefore, agency (and not "free agent") is usually related to — or embedded into — social structures or social processes. In spite of their limitations and their need for help from "outsiders", the human sciences have moved beyond the classical opposition between freedom and determination. These considerations should be part of discussions on agency and causality in human sciences — even among philosophers.
One could hope that the text of R. Boudon ("Which Theory of Behaviour for the Social Sciences?") would be more useful for human scientists. At least, the reader will find some references to works in social sciences (Durkheim, Geertz, Lévy-Bruhl, Nisbet, Runciman, Sahlins, and Weber). However, there is no mention of any other recent and significant contributions (Elias, Giddens, Archer, etc.), and only readers who do not know Boudon's previous work will be interested by this text. Rational choice theory, or the idea that actors "do what they do because they think that their action will bring them maximum satisfaction", is rejected by Boudon because it cannot explain many behaviours. He also rejects the idea that "behaviour should in most cases be explained as the effect of psychological, cultural or biological forces". Boudon favors a "third model", "used explicitly by Max Weber", but also by other classical authors such as Durkheim and Tocqueville. In this last "model", these classical thinkers "consider the cause of any individual action to be its meaning in the minds of people." Individual actions are "the outcomes of understandable motivations and reasons" (p. 198). Once again, Boudon's interpretations of classical texts are presented as data supporting his general thesis. Actions are motivated by various forms of rationality. Thus, even if many actions seem to be irrational, there are always "rational" in one way or another since their "causes are the reasons and motivations that people have for believing what they believe. … Hence, explaining an action, even apparently irrational ones like rain rituals, means reconstructing the motivations and reasons of the actors." (p. 207) The solution offered by Boudon could be labelled as a "neo" or "soft" rational choice theory where the notion of "rationality" becomes more flexible or integrative. Rational behaviours are not always instrumental. Behaviours can also be caused by "cognitive rationality" (p. 208).
Overall, this book cannot be used for teaching at any level in the human sciences, since it does not really talk about the human sciences. It is composed of philosophical argumentations on causality and agency that students in philosophy might find useful. Human scientists might find a couple of interesting general ideas in some chapters. It also helps to identify some contemporary publications on causality and agency in philosophy. Unfortunately, it does not offer much more than that for human scientists. One hopes that next time, disciplinary walls will finally begin to be replaced by footbridges.
François Dépelteau
Laurentian University
http://www.cjsonline.ca/reviews/agencycausation.html
November 2007
© Canadian Journal of Sociology Online
