Jean-Léon Beauvois (1943–2020): A Key Figure in Social Psychology

sidered to be one of the main thinkers in French social psychology, we would like to review some of the major ideas of his work which highlight his intellectual consistency and creativity, and reflected real life issues. Indeed, one of his main concerns was to make use of the findings of social psychology to better understand social issues and help resolve them, including group discrimination, intergroup conflicts in institutions, unemployment, academic and professional orientation, employee evaluations, workrelated accidents, health protection behaviours, energy consumption, and environmental protection, to name but a few. His view of the role of people as social actors evolving and growing within a social context, aroused such enthusiasm that many students and colleagues followed in his footsteps to develop this stimulating approach to social psychology. In his early career, Jean-Léon Beauvois worked on language with Rodolphe Ghiglione. Refuting the dominant linguistic theories of Saussure (1916) and Chomsky (1968), according to which language can be analyzed as a distinct entity, Beauvois and Ghiglione (1981) believed that language could not be dissociated from the speaker or from the context in which it was used. They viewed language as a tool that people use to communicate with others and to decipher the world around them. Language is thus first and foremost determined by the communication structures in which individuals find themselves. This primary role attributed to the social aspect as a structuring feature of language and thought would feature strongly in all of Beauvois’s work.

ation practices. It serves a function in social life, where people act more as judges evaluating behaviours (their own as well as those of others) than as scientists seeking the ' objective' triggers which cause them (Beauvois, 1976;Pansu & Beauvois, 2004).

Social relationships generate two types of knowledge
In Beauvois's opinion, the knowledge that people have of their own personality and that of others is not a scientific knowledge. He provides two reasons. First, only the social relation of observation, which a scientist has with the object he or she is studying provides the scientific knowledge of that object (e.g., a biologist examining protozoa). This social relation of observation, which is not spontaneous, allows the biologist to grasp the intrinsic properties of the object of study (the protozoa). The second reason is that this type of scientific observation is not the one which is operative in social relationships such as those involved in education, work, consumption, etc. Indeed, these social relationships, because they attribute a specific position to each person (teacher or student, employer or employee, producer or consumer), produce a different type of knowledge: the evaluative knowledge.
According to Beauvois, unlike scientific knowledge, which is that of an observer describing an object of study without the intervention of social presuppositions, beliefs, opinions and prejudice, evaluative knowledge, which is the knowledge of ordinary people in everyday life, assigns a fundamental role to the function that objects and people have in the social system. Beauvois liked to illustrate these two types of knowledge by comparing the way a biologist, placed in a relation of observation, and a butcher, in a relation of consumption, might talk about the same piece of meat. While the biologist will speak of it as a muscle and will describe its elasticity and its contracting power, the butcher will speak of it as a fleshy, soft and gelatinous piece of meat. These two types of knowledge have just as much social reality as each other, but only scientific knowledge is descriptive because it provides information on the natural properties of the object. The butcher's knowledge is different: It communicates the social value of the piece of meat, i.e. what it is worth relative to others (e.g., a stewed meat) in a relation of consumption. For Beauvois, the type of knowledge of the academic psychologist, whose work is to analyze personality, is much closer to that of the butcher than that of the biologist. From his point of view, this was not a negative thing in any way, considering that evaluative knowledge is a type of social knowledge in its own right. What is a problem is to conflate evaluative knowledge with descriptive knowledge. This confusion between the two types of knowledge raises the issue of what we are studying, or indeed, what we think we are studying.
Awareness of the existence of these two types of knowledge is far from widespread in psychology. This is evidenced by the fact that very early, researchers in psychology addressed the study of judgement and attribution processes by presupposing that individuals operate as 'intuitive or naïve scientists' (Heider, 1958;Kelley, 1967). When this presupposition could not be confirmed, researchers turned to a softer cognitive conception defending the idea that when issuing judgements, especially in uncertain situations, people use heuristics, i.e. simple principles which allow them to offer a satisfying, although imperfect, solution to complex questions. Among the most important, are representativeness, availability and anchoring heuristics, updated by Tversky and Kahneman (1974), as well as simulation heuristics (Kahneman & Tversky, 1982). Since people are subject to cognitive biases (e.g., the halo effect: Thorndike, 1920;Kahneman, 2011;illusory correlation: Chapman & Chapman, 1967), researchers came to see the individual as a 'biased computer.' The underlying idea was that if people took the time and the means and if they were motivated enough, then surely they would produce scientific knowledge. For them, the only explanation for these cognitive biases came from cognitive deficits in individuals. Beauvois found such an explanation utterly unconvincing (Beauvois 1984b). For him, understanding the psychological functioning of individuals in their daily life is not a matter of descriptive knowledge, but of evaluative knowledge.
These two types of knowledge are fundamentally different. Each has its own logic: the former uses hypotheticaldeductive reasoning, while the latter is anchored in social structures. Each also has its own validity criteria: criterion of truth for the former, and criterion of social acceptability for the latter. In Beauvois's opinion, this view is based on the distinction between descriptive and evaluative knowledge. It arises from a discontinuous epistemology, which involves, as argued by Bachelard (1938), an epistemological break, and not a single gap on a scale continuum from science to common knowledge. Beauvois's originality thus resides in his plea for the recognition of common, evaluative knowledge, sometimes taken to be the lack of knowledge, and to give it the status of a true knowledge, even if it is not descriptive.

Personality traits are merely indicators of the social value of social agents
The weight of psychological realism Beauvois's opposition between descriptive and evaluative psychological knowledge can be better understood through his work on personality traits, i.e. the adjectives that are used to speak of others and of ourselves. Traditionally, personality traits have been seen as natural realities, intrinsic properties of a person (Beauvois & Dubois, 2016). For example, one might say of a person that he or she is shy, of another that he or she is easy-going, and of oneself that one is a fundamentally honest person. Because they are viewed as natural, these characteristics are supposed to differentiate people from one another. Beauvois called this view 'psychological realism' because it assigns a purely descriptive function to personality traits ( Beauvois & Dépret, 2008). These traits are assumed to reflect 'true' psychological characteristics of individuals and determine their behaviours . For example, we expect a person who has been described as 'honest' to notify the cashier at the supermarket if he or she was undercharged for an item. For Beauvois and Dubois (2016), however, personality traits are not what psychologists claim they are. They do not arise from descriptive knowledge. When we claim that Sophie is affectionate and loyal, we obviously express that she is gentle and keeps her commitments, but we mostly convey her value as a spouse within the context of a conjugal relationship, and as a mother within the context of a parental relationship. We understand from this evaluation that Sophie will not be easily seduced at a party, but she could be trusted to look after our children. In other words, while personality traits tell us little about who a person is, they do inform us about his or her value in a given register of social relations. Personality traits tell us what we can or should expect of others and/or we can or should do with them, in the context of romantic or interpersonal relationships, within the family, at school or in the workplace.

Two types of social value
People's social value (like objects' social value), provided by evaluative knowledge is expressed within two distinct domains: the domain of positive/negative affects that people arouse (he or she is liked vs disliked), and the domain of people's social utility in their production activity (he or she is a good or a bad student; he or she is a competent or incompetent employee). The former domain was called social desirability, and the latter, more rooted in social functioning, was called social utility. Beauvois carried out a series of studies showing that people in an evaluator position (teachers, trainers, psychologists, managers, business leaders) are more concerned with people's social utility than their social desirability. For example, when determining whether a pupil can advance to the next academic year, a teacher will mostly take into account the pupil's record as a good or bad pupil, not his popularity rating. Other studies have shown that people in more prestigious positions (good student, factory manager, doctor) distinguish themselves from people in less prestigious positions (bad student, worker, caregiver) mainly by their social utility. The former are seen and see themselves as more socially useful that the latter, relegating social desirability as less important.

The social anchoring of judgement norms
Other studies have shown that social utility is associated with the normativity of judgements, such as for example those related to internality (Dubois, 2003(Dubois, , 2006. For Beauvois (1984a;Beauvois & Le Poultier, 1986), internality involves a normative tendency which consists in thinking that people have within themselves the causes of what they do and what happens to them. This translates into an increased use of internal causal explanations (personality traits, motivations) rather than external ones (others, context), whether the event is positive or negative. Beauvois's originality in this domain is twofold. First, he expanded the concept of norm of internality, introduced by Jellison and Green (1981) to explain reinforcements (what people get), to also explain behaviours (what people do). He also highlighted the social anchoring of this norm in evaluation practices. Beauvois thought that internal explanations are neither an intrinsic characteristic of certain people (called 'internals') nor a cognitive bias (a fundamental error in attribution or an illusion of control). Such a preference for internal causal explanations is a consequence of a form of social learning, which when it is realized, leads to being appreciated by social evaluators (Pansu & Gilibert, 2002;Pansu, Bressoux & Louche, 2003). People who refer to internal explanations are judged as socially more useful than those who do not, or who do it only a little (Cambon, Djouari & Beauvois, 2006;Dompnier, Pansu, & Bressoux, 2007). The former are hired and promoted to positions of authority without hesitation, regardless of their social desirability, while the latter are not hired or promoted to such positions, even when they are highly socially desirable.

Social agents are not driven by their ideas, but by the context
This social determinism is also fundamental in Beauvois's conception of the psychology of commitment, which has found many uses in fields such as training, work, healthcare, marketing and even therapeutic practices (e.g., Beauvois & Joule, 1998). For Beauvois, what drives action cannot be found in the attitudes or motivations, nor even in the intentions of individuals. It can only be found in their previous actions in a context that did not involve forced compliance (e.g., agreeing to put a pro-environment sticker on their car). These previous actions are what drive an individual's commitment and allow him or her to perform future actions in the same vein, actions that he or she would never have done unprompted (e.g., distributing tracts in favour of more drastic pro-environmental actions every Sunday morning for a month). Beauvois thought that the way people act and think is a function of their previous actions, rather than a function of their convictions and ideas. People are thus good at rationalising, but they are not necessarily good rational actors. This view, and the manner in which it has been developed in commitment theory (Joule & Beauvois, 1987), has often been considered shocking. Indeed, it seems contrary to common sense and to the image we have of humans as depicted in the media, which puts forward the intrinsic determination of behaviours and peddles the image of free and autonomous people. Beauvois breaks with this tradition since for him, this freedom of social agents simply does not exist. In Western societies, people only claim to be free because they have been brought up with liberal and individualistic values, chief among which is surely the notion of 'freedom.' In moving away from a view of commitment, which gives the individual a decisive role, Beauvois and Joule attribute a central causal role to context in determining attitudes and behaviours. Of course, this requires that the context presents a certain number of characteristics, including that the person must feel free, not coerced into performing the action, that the action be realized publicly rather than in isolation, this action must have important consequences for the person, it must not be positively or negatively reinforced, and it must have been performed multiple times. An individual's commitment to an action thus depends exclusively on the conditions in which it has been performed, even if the individual believes that he himself is the sole originator of the action. Clearly, in commitment theory, Beauvois is very far from self-determination and internal control of behaviours.
This brief overview of Beauvois' major ideas makes it possible to grasp the originality of his ideas: Individuals do not exist, there are only social agents, whether they are students, employees, unemployed workers, business managers or retirees. It is as social agents inserted within social structures and involved in particular social relations (teacher/student; employer/employee; doctor/patient) that they produce evaluative knowledge about themselves, others, and the world. It is also as social agents that they behave, with the illusion of freedom, in particular social relationships within these social structures. Social organization, through the places and the roles that it gives to people, thus has a determinative causal role in the formation of attitudes, judgements, decisions and behaviours, and even of the 'self,' about which Beauvois liked to say 'Being oneself is surely the most ridiculous of all popular maxims.'