Pareto, Mill and the cognitive explanation of collective beliefs. Unnoticed “middle-range theories” in the Trattato

International audienceEven if it is probably not the real core of Pareto’s sociology (Turner et al.,1981, and Powers, 1987), the most original part of Pareto’s Trattato deals with the very contemporary issues of the rationality of beliefs and the rationality of actions (Valade, 1990). These two latter issues are connected, given that many actions can be considered as rational or irrational depending on the rationality of the beliefs upon which they are based. I will concentrate in this paper on the sole issue of the rationality of beliefs, more precisely, collective beliefs. I do not contend that Pareto made a thorough examination of these general issues. Indeed, in the final part of this paper I will bring our attention to certain important limitations of Pareto’s general account of collective beliefs. I also do not maintain that his analyses of singular examples were always profound, because clearly often they were not, but that he gave probably the largest account that has ever been given of the various roles that justification of beliefs and actions (“argumentation”) can take within a sociological framework. Chaïm Perelman and Lucie Olbrecht-Tyteca (1969), along with contemporary argumentation theoreticians such as Toulmin (1958), Hamblin (1970), may have gone further in the investigation of certain micro-mechanisms of argumentation. But neither scholar spoke in general of the issues of argumentation procedures within the framework of social science as broadly as Pareto did. I will proceed step by step, discussing several of the most widely accepted views of Pareto’s theory of collective beliefs, arguing that each one is partial and too narrow. In each case, I will introduce new distinctions, which will result in a much broader view of Pareto’s conception of beliefs. I will try to avoid Pareto’s very idiosyncratic vocabulary as much as possible because, according to Pareto himself, this may be deeply misleading (Pareto, 1935, §119, 868) . Furthermore, the meaning of Pareto’s specific vocabulary varies largely in the Trattato depending on the specific context. On several occasions, I will prefer John Stuart Mill’s wordings, generally clearer than Pareto’s. At the end of this paper, I will even use one of Mill’s specific psychological micro-models, because it turns out to be more refined and empirically more relevant than Pareto’s models. By referring to Mill, who gave both a psychological and logical account of logical errors, I will remain very close to Pareto’s programme. As Pareto himself wrote, after speaking of Mill, “It is the province of logic to tell why a reasoning is false. It is the business of sociology to explain its wide acceptance” (§1411). Pareto thought of his sociology as the complement of Mill’s logic, (§ 1410-1412) but he seems to have underestimated the conceptual accuracy and the empirical relevance of psychological models involved in the System of Logic. The careful investigation of Pareto’s micro-models should lead us to re-read Pareto’s Trattato di Sociologia Generale in the continuity of Mill’s System of Logic and to emphasize the cognitive aspects of Pareto’s sociology (Bouvier, 1992) counter to the usual ultra-emotivist interpretations (Aron, 1965). But it should also lead us also to re-evaluate the Machiavellian aspect of his thought, upon which many scholars have insisted for some time (Burnham, 1943, Fiorot, 1969, Femia 2006, Marshall, 2007), in favor of a more complex, less dissembling, and more communicative view of political, moral and religious discourses. The first two parts of this chapter will be devoted to the latter concerns and the remaining sections to the cognitive dimension. The issues, though, are related, in particular the role of emotions in beliefs and in actions