The Social Value of Persons: Theory and Applications

When people talk about an object in their environment, they have recourse not to scientific language, but to evaluative language. This evaluative language allows them to say what one does, can do or must do with that object (i.e., its functional or social value). For instance, when Mrs Smith talks about an orchid that someone has given her, she says, It's so delicate and refined! Its flowers come in fantastic shapes and yet it’s really easy to grow. When she talks about the meat she is cooking for dinner, she says, It's tender, juicy, and smells good. Statements like these make sense, but not scientific sense. These words are not the words of a botanist or a physiologist. This is a language used to say what one does, can do or must do with a flower or a cut of meat, something that is never expressed in descriptive or scientific language. A botanist might say, for example, that an orchid has three colorful sepals and three petals, that two of the petals are underdeveloped, that the third, on the contrary, is highly developed and forms the labellum, and so on. A physiologist might say that muscle is made up of thousands of cylindrical cells called muscle fibers, that each fiber is enveloped and separated from the others by a thin layer of connective tissue the endomysium -, and so on. Curiously, though, when Mrs Smith talks about her friends or neighbors, many social psychologists opine that the words she uses to describe those friends and neighbors are more like the descriptive or scientific words of the botanist or physiologist than the evaluative words she uses to talk about meat or flowers.

theorists felt it was important to reinstate human rationality in the social sciences, after decades of psychoanalysis and psychology of affects and motivations. However, such was their overpowering influence on social psychology throughout the second half of the twentieth century that we can legitimately ask whether they did not do more harm than good. These ideas led social psychologists to forget that human beings essentially are evaluative creatures (Kluckhohn, 1954). And yet, it can be argued that the approach that people have to objects (animate or inanimate) traditionally called "objects of knowledge" prevents them from knowing what those objects are from a descriptive point of view. Instead, it allows them to decide what one can or should do with them (Can I get involved in this project with him?) or what one can make them do (Can I persuade this pupil to embark on a lengthy course of study?). In other words, it allows them to evaluate the objects (Beauvois & Dubois, 2000, 2009. Thus, instead of being "objects of knowledge", objects are actually "objects of evaluation" to people. This conception rests on a different, noncontinuist epistemology, wherein science can only flourish if it breaks away from common sense (Bachelard, 1938(Bachelard, , 1953. The fact of seeing the science everywhere in the thought and in the judgment is linked to a dichotomous conception according to which the science opposes to the error, maybe to the madness. A noncontinuist position moves forward that the science is only one of the possible modes of knowledge of the objects which do not have all the rationality of the science. Common sense is certainly not irrational, but its rationality is different from that of science. Common sense cannot be judged on the criterion of truth value, but it can be judged on the criterion of social acceptability. The common sense's statements must be acceptable (versus unacceptable) by people rather than be true (versus false). This is the case when we use naive psychology and, in particular, personality traits to understand the people with whom we have dealings. In this conception, traits are regarded as genuine concepts, but concepts that are not descriptive (or scientific) but that are directly, genuinely evaluative.
The purpose of this chapter is to outline a set of theoretical formulations and some studies that rely on an evaluative conception of intuitive psychology (Beauvois, 1976(Beauvois, , 2011Beauvois & Dubois, 2000, 2009Dubois, 2006;Dubois & Beauvois, 2011) and, more specifically, of the intuitive psychology of personality traits. While this conception differs from most current theories, it can be likened to the formulations of researchers who differentiate between a paradigmatic manner of thinking and a narrative one (Zuckier, 1986).

Psychological realism: Traits and implicit personality theories
The continuist claim that the concepts handled by the naive, or intuitive, psychologist are descriptive or quasi-scientific has been referred to as psychological realism (Beauvois & Dépret, 2008). In psychological realism, the use of traits is regarded as a realist approach to a person's psychological nature and these traits are assumed to be intrinsic properties of the person, the homo sapiens sapiens, about whom the intuitive psychologist is talking. This idea is based on a more basic hypothesis (the lexical hypothesis; see Mollaret & Mignon, 2006) that traits are efficient tools for decoding the psychological reality. Obviously, psychological realism does not presuppose that the intuitive psychologist never makes mistakes. As such, psychological realism is less concerned with the functioning of an individual than with the nature of the concepts he/she uses, in particular personality and personality traits. Because www.intechopen.com The Social Value of Persons: Theory and Applications 309 traits, as quasi-scientific constructs, cannot be "seen" directly, they have to be apprehended via their visible behavioral manifestations. Accordingly, many models of social cognition (and, in particular, person memory;Srull & Wyer, 1989) stipulate that knowing about a person's psychological nature means being able to encode his/her behaviors as traits.
We use the term TB behaviors to designate the behaviors of a target person associated with a trait in memory (see Beauvois & Dubois, 2000). The well-known example of "pointing out a mistake made in one's favor" is a TB that allows us to attribute the honest trait to a target person. For convenience's sake, we consider that TB behaviors reflect knowledge that can be viewed as descriptive at the operational level.
In intuitive psychology, one important line of research concerns the dimensions underlying the psychological descriptions made by people in terms of traits. These dimensions are seen as the implicit personality theories of the intuitive psychologist (Beauvois, 1982;Bruner & Tagiuri, 1954;Rosenberg & Sedlak, 1972;Schneider, 1973). When people are asked to describe either themselves or others using personality traits, their psychological descriptions are generally framed by a very limited number of dimensions, either one (Kim & Rosenberg, 1980), two (Rosenberg & Sedlak, 1972) or five (Passini & Norman, 1966). Explanations for the theoretical foundations of these dimensions are often anchored in psychological realism, in that the dimensions are seen as referring to a psychological reality 1 . They are assumed to correspond to the structure of human personality, and the perceiver, as a quasi-scientific intuitive psychologist, therefore seeks to answer the question How can I characterize this person according to the basic dimensions of human personality? (Abele & Wojciszke, 2007;Judd, James-Hawkins, Yzerbyt, & Kashima, 2007;Suitner & Maass, 2008). Obviously, perceivers rapidly deduce the evaluative implications of the psychological realities they discover (Srull & Wyer, 1989). Moreover, psychological information is often processed precisely with the aim of arriving at these evaluative implications. However, the evaluations are reached on the basis of a presumed psychological reality by people who are explicitly or implicitly seen as potential intuitive scientists, even if they may, like all scientists, sometimes make mistakes. These errors do not invalidate the idea that the dimensions are genuine psychological dimensions that arise from people's actual nature.
By contrast, in the evaluative conception, dimensions are thought of as fundamentally social, and belonging to the registers of social values.

The evaluative approach: Traits are criteria of social value but not descriptive constructs
The evaluative approach rests on two major arguments: 1 This can be problematic when these dimensions are used in studies of stereotypes. The latter prompt the elaboration of a supposedly universal model: the Stereotype Content Model (see Echebarria Echabe, in this book). These presuppositions are problematic insofar as the truth value of stereotypes is often questioned in the psychosocial literature. The foundations of psychological realism can nevertheless be preserved by arguing that if the truth value of a stereotype is questionable, it is because that stereotype is the product of errors of judgment, which can be explained by social necessities. We may, for instance, think that the Chinese are smaller than the Indonesians. This is doubtless an error, but the fact remains that the size of a person is an authentic individual variable which allows us to express true or false judgments. After all, psychological realism does not imply that intuitive psychologists are accurate in all their judgments.

www.intechopen.com
Psychology -Selected Papers 310 -An empirical argument: the main personality traits, the central most frequently used traits (Asch, 1946), are known to be strongly evaluative and only weakly descriptive in meaning (Labourin & Lecourvoisier, 1986). We have also known for a long time that the dimensions of implicit personality theories are clearly evaluative: they set "good traits", such as intelligence and competence, against "bad traits", such as stupidity and laziness (for a brief review, see Cambon, 2006). This opposition between good and bad traits rests on a social point of view (people who behave intelligently are more highly paid than people who behave stupidly because the former are more socially useful than the latter). The defenders of psychological realism therefore have to answer the following puzzling question: Why are humans (Homo sapiens sapiens) the only creatures in nature whose intrinsic properties vary linearly with social utility or value? -A cognitive and epistemological argument: an analysis of the cognitive construction of the meaning of personality traits (Beauvois, 1976(Beauvois, , 1984Beauvois & Dubois, 2000) reveals that the social value of people's behaviors 2 (e.g., the social value of the behaviors of a student said to be intelligent or honest) is not a secondary, post hoc inference (intelligent => intelligent behaviors => individual's social value), but rather is inherent to the meaning of the traits in the first place (intelligence = a set of behavioral social values). A set of behavioral social values is the sole basis for the emergence of a personality trait. Most traits are constructed not to grasp an invisible individual characteristic, but to express the observed behavioral social value. This process is doubtful from an epistemological point of view. The mere connection between a behavior (or performance) and a personality-trait implies that the numerous variables (in particular situational variables) which, according to the determinist point of view, affect this behavior (or performance) are forgotten. Only the social value of behavior is retained and is "psychologized" through the use of an assumed causal trait (see Beauvois & Dubois, 2000, for a formal description of this process).
In other words, when a pupil is said to be intelligent, the intelligence trait tells us nothing other than: this pupil has displayed a behavior (or performance) characterized by a certain kind of social value. Good (or poor) behavior (or performance) is thus translated into traits. These traits convey the person's social value without any descriptive or scientific (determinist) basis.
In the above example, the value is named intelligence. Thus, personality traits intelligent, competent, helpful vs. weak, stupid, selfish are merely paraphrases of the social value of the behaviors or performances of the person concerned (Beauvois, 2011). This is the reason why traits tell us what we can or should do with people (the purpose of any evaluative activity), but say nothing about the individual variables that affected this behavior or performance.
We use the term OB behaviors (see Beauvois & Dubois, 1992, 2000 to designate the behaviors that others can or should display towards a person known either by a trait or by a set of behaviors endowed with some social value that is expressed by that trait. For instance, for honest, you can give him confidential information or you can lend him your office keys. By definition, OB behaviors reflect an evaluative knowledge. If the evaluative conception is correct, traits in psychological descriptions should directly activate OB behaviors in memory, as well as TB behaviors. In short, the evaluative conception merely assumes that, through evaluative practices, everyone learns to use the language of personality traits and hence, implicit personality theories with the purpose of grasping and expressing, in psychological terms, the main aspects of the social value of people's behaviors both the positive value (e.g., intelligent) and the negative value (e.g., stupid).
In the following three sections, we discuss a variety of empirical studies whose findings provide strong support for the evaluative approach in psychological description (for a more comprehensive review, see Beauvois & Dubois, 2000, 2009 It was important at the beginning of this trend of research to show that OBs provide a system for encoding personological information that is just as efficient as the TB system. To this end, Beauvois and Dubois (1992, Study 1) conducted an experiment whereby a list of 33 positive and negative traits was read out to a group of students who had to write down a behavior they considered to be characteristic of a person defined by each of those traits in turn (TB list). A second group had to write down a behavior that people might or should have towards that person (OB list). In a second phase, different students were shown the most frequent behaviors associated with each of the 33 traits on the TB or OBs lists. Then they were asked to select those they considered to be the most representative of each trait. They were also asked to rate their confidence after each choice on a 100-point scale. Results are given in Table 1 for positive and negative traits: the frequency hierarchies are virtually identical, as are the confidence curves for TBs and OBs. Similar results were obtained by Mignon and Mollaret (2002   In the two tables, no comparison of frequencies (Χ²) nor means (t test) inside the same rank reached the threshold of significance (.05). So, these results show that the OB register is no more ambiguous than the TB register 3 . They exclude the idea that the OB register is more idiosyncratic than the TB register. They seem hardly compatible with the idea that OBs are post hoc inferences made on the basis of TBs. One could, for example, claim that it is because I have stored honest => hands in wallets to the police (TB) in memory that I can infer that an honest person is someone to whom I can lend money (OB).
Beauvois and Dubois (1992, Study 2) used a semantic decision task precisely to reject this idea as far as main traits are concerned. Subjects were asked to indicate as quickly as possible whether or not a particular OB or TB behavior was characteristic of a trait that had just been presented. Traits were either strongly evaluative (traits whose evaluative component outweighed the descriptive component: e.g., honest, dynamic) or weakly evaluative (traits whose descriptive component outweighted the evaluative component: e.g., shy or talkative). The twenty OB and TB were chosen so that twelve of them (6 OBs and 6 TBs with comprable association frequencies) exemplify the given trait, six of them (3 OBs and 3 TBs) exemplify another trait and two of them (one OB and one TB) exemplify an antonym of the given trait. Reaction times were recorded. The significant behavior x traits interaction: F(1,17)= 13.30, p<.002 indicated that, for positive and negative responses alike, the OBs were judged as rapidly as TBs on strongly evaluative traits (1328ms vs. 1307 ms). On the other hand, they were judged more slowly on weakly evaluative traits (1647 ms vs. 1308 ms) 4 . These results allow us to contend that when highly evaluative traits are at stake (the most central traits), OBs are indeed stored and granted a status in long-term memory that is at least equivalent to that of TBs. In a subsequent semantic decision experiment, Beauvois, Dubois, and Tarquinio (1994) assigned half the subjects to a "social context" condition, where TBs and OBs were placed in a social setting that was relevant to the trait in question (at a party, he enjoys talking to everyone). Interestingly, in this condition, and always for the most evaluative traits, reaction times were shorter for OBs than for TBs, whereas no such difference was found in the "no social context" condition (i.e., the same condition as in Study 2 of Beauvois & Dubois, 1992). Again, these results are in line with the idea that the OBs for more central traits are activated without having to call upon TBs beforehand, in particular when there is a social context.

Person memory
Beauvois, Dubois, Mira, and Monteil (1996; see also Milhabet, 2004) described three studies exploring the person memory paradigm. Students listened to lists of behaviors comprising filler items, OBs and TBs said to characterize a single person. Four OBs and four TBs were provided in Studies 1 and 2, and six OBs and six TBs in Study 3. In all three studies, half the OBs and half the TBs were representative of a single trait that was either strongly evaluative (honest, dynamic, etc.), or weakly evaluative (shy, passionate, etc.) These traits varied from one experiment to another. In all three studies (see Table 2), the same significant behavior x trait interaction was observed: in study 1  Beauvois, Dubois, Mira, and Monteil (1996) For highly evaluative traits, subjects remembered more OBs than TBs, and vice versa for less evaluative traits. Results suggested that OBs are endowed with a memory structuring power that exceeds that of TBs. In the third experiment, the virtual social relation between target and participant was manipulated. The subjects were students, while the target was said to be either "a person", "a student" or "a teacher". The interaction was strongest, and accounted for a greater proportion of variance, when the target was one of the student's professors (i.e., power relationship) than when the target was a student or a person. Again, as in Beauvois, Dubois and Tarquinio's study (2004), the structuring power of the OBs was strongest when a social context was induced.

Implications
The above-described studies validate three implications of the evaluative conception. As far as the more central traits are concerned: a. the evaluative component is mobilized just as much as the descriptive component in the intuitive use of traits. In other words, OBs constitute a means of encoding the knowledge provided by the traits and are called upon either in conjunction with TBs or as an alternative to them; b. evaluative knowledge is not the product of inferences drawn from descriptive knowledge. In other words, from an operational standpoint, OBs are part of the meaning of a trait and do not presuppose TBs; c. the evaluative component of traits becomes more accessible and is more frequently used when the social context is made salient.
Others studies (see Beauvois & Dubois, 2000) have yielded further empirical evidence of the validity of the evaluative conception of trait meaning. In particular, they have shown that OBs and TBs have different effects on first impressions (zero-acquaintance paradigm): OBs are more effective than TBs as far as interpersonal differentiation (evaluation) is concerned, while TBs are better than OBs at allowing intraindividual analysis of targets (description). It has also been shown that OBs are just as good recall c u e s a s T B s i n p e r s o n m e m o r y a n d t h a t OBs define a universe that is a good approximation of implicit personality theories established on the basis of the traits themselves. Moreover, people whose job involves judging other people for the purposes of determining which actions to take towards them (e.g., social workers) exhibit a sociocognitive functioning which relies more on the evaluative component of intuitive psychology than on the descriptive one.
The choice of the word "evaluative" to refer to a register of knowledge may seem surprising. In the psychosocial literature, it is always associated with the affective or emotional register, which is why the evaluative process has attracted so little attention from social cognition theorists. This is the consequence of the dichotomous thinking that we mentioned above. Evaluation can rarely be considered as a scientific activity. It cannot therefore be considered as a competence of the intelligent thinking. Thus, evaluation is of the emotional register. And yet it seems clear that OBs give rise to genuine knowledge (from this point of view, they have the same status as TBs), namely the knowledge of what we can or should do with people. Of course, this knowledge is sometimes inferred from another, possibly more descriptive kind of knowledge (TBs). But for traits that are very common in psychological descriptions, namely the most central ones, which are more evaluative than descriptive in nature, the evaluation is supplied directly as knowledge of what we can or should do with people, that is, knowledge of their social value (Beauvois, 2011). People's social value cannot be a mere individual fact. Human resources managers cannot refer to their affective reactions to fix salaries or grant promotions. Nor can professors bank on their emotions when marking their students' work. Both can very exceptionally only justify their evaluation by the science. Now, their decisions must be understandable. Social evaluative knowledge, we said it, must be endowed with social acceptance. Obviously, the managers' and professors' decisions can still be disputed, but any contestation must be based on a piece of acceptable evaluative knowledge. We can see how the prevalent idea in social psychology that evaluation is derived from an emotional register has caused social psychologists to neglect an important aspect of the knowledge conveyed by naive psychology.

The evaluative conception and the two-factor model: The main factors extracted from psychological descriptions are the main dimensions of people's social value
We have said that when people are asked to describe either themselves or others using personality traits, the psychological descriptions they produce are generally framed by one, two or five main factors (or dimensions). The two-dimensions approach is the most stimulating one today (see Special Issue, European Journal of Social Psychology, 2008; see also Echebarria Echabe, in this book). The first dimension pertains to the quality of the target person's interpersonal relationships (e.g., warm, helpful, honest vs. cold, irritable, selfish), while the second refers to the energy and resources that the target person invests in his/her activities (e.g., intelligent, ambitious, active vs. lazy, weak, careless). These two dimensions have been given various names which, from a theoretical standpoint, may seem arbitrary: value vs. dynamism (Osgood, 1962), social good-bad vs. intellectual good-bad (Rosenberg & Sedlak, 1972), other-profitability vs. self-profitability (Peeters, 1992), morality vs. competence (Wojciszke, 1994(Wojciszke, , 2005, warmth vs. competence (Fiske, Cuddy, & Glick, 2007), communion vs. agency (Abele & Wojciszke, 2007), etc. Except for a few variations, however, these two dimensions always contain the same traits, at least as far as the positive pole is concerned.
According to psychological realism, these two dimensions are clearly derived from genuine psychological knowledge of what people are (i.e., from the standpoint of a virtually scientific or quasi-scientific descriptive psychology). They refer to a psychological reality in the form of psychological information that is extracted by the perceiver and/or derived from the dimensions of human personality.
This view is challenged by the advocates of the evaluative conception. They assume that both types of information are derived not from truly psychological knowledge of people, but rather from a social evaluation of people that is directly supplied by interchangeable traits (i.e., intelligent, capable, clever, gifted). In addition to the above arguments, there is an important historical argument: the two dimensions were identified by industrial psychologists studying the occupational value of men and women in the workplace (see De Montmollin, 1972;Tiffin & McCormick, 1965). These two dimensions therefore emerged from people's knowledge of two basic aspects of the social value of employees in organizations. In this context, but also in a more general way, the value of a person is generally apprehended through the observation of poor or good behaviors, judgments or performances. As we demonstrated earlier, poor or good behaviors (or performances, or judgments) are then given labels (assumed causal traits). Implicit personality theories, which are essentially semantic or conceptual in nature (Shweder & D'Andrade, 1979) are then activated, giving rise to descriptions that go far beyond what has actually been observed (intelligent and competent activate ambitious, ambitious activates dynamic; helpful activates nice, nice activates sociable, sociable activates honest, and so on.) This evaluative view led Beauvois (1995) to propose new names for the two dimensions and to emphasize their evaluative nature by presenting them from an evaluator's viewpoint: social desirability for the former, social utility for the latter.
Their definitions are set out below. First, however, we describe three sets of studies which have yielded strong evidence supporting this evaluative view.

Social desirability and social utility of professionals. Is a factory manager more competent than a hospital manager?
Cambon (2004) borrowed Ricardo's and Smith's opposition (echoed by Marx) between production (production of goods for the market) and maintenance (taking care of the producers) within a social structure. In several studies, he studied the psychological descriptions of different professionals known only by their occupation (e.g., a factory manager in the production sector vs. a hospital manager in the maintenance sector). These descriptions were made in terms of positive or negative traits indicating social desirability or social utility. Every description was analyzed using a utility index and a desirability index (number of positive adjectives minus number of negative adjectives). In studies where gender stereotypes and professional status were manipulated or controlled, Cambon observed that professionals working in the production sector were seen as more useful than desirable, whereas professionals working in the maintenance sector were often described as more desirable than useful (see Table 3 for the data of Study 4 in which the gender of the targets was controlled).

Production Maintenance
Social utility 1.06 0.31 Social desirability 0.10 0.63 Table 3. Mean desirability index and mean utility index for psychological descriptions of professionals working in production versus maintenance sectors (subjects: economics students) in Cambon (2004, study 4) These data were submitted to a 2 (professional sector: production vs. maintenance) x 2 (traits: social utility vs. social desirability) ANOVA. This analysis yielded the predicted interaction, F(1,49)= 28,56; p <.001. Such results are not easy to explain from the psychological realism perspective (who can say that the director of a hospital is less "competent" than the director of a factory?) On the contrary, they are typical of our evaluative approach and justify our quasi-economic idea of social utility. Adam Smith said that when a worker is hired, (economic) value is created, while when a domestic is hired, (economic) value is consumed. Economists have indeed often claimed that production activities are more typical of (economic) value than maintenance activities. We can understand why production professionals are more highly rated on the most "economic" dimension (i.e., the social utility dimension). This certainly does not mean that they are, from a purely "psychological" point of view, more competent, more dynamic than maintenance professionals, who are supposedly nicer and more honest. Rather, it simply means that, because they occupy different economic positions, they are not judged on the same social values. Since we do not use the same traits to express different values, production and maintenance professionals are described using different personality traits. Intuitive psychologists have no access to invisible psychological realities. They talk about what they see or think they are seeing in a social structure or in a given context. The words they use are embedded in that social structure or context. Consequently, intuitive psychologists always talk about a target and about the value of that target within the social structure or context.

A human psychological reality is not a prerequisite for the emergence of either dimension: Are some pet rabbits more competent than others?
The evaluative approach assumes that social desirability and social utility do not have a human psychological reality as empirical reference. Rather, they operate independently of any kind of empirical or even conceivable human psychological reality. The notion of "human psychological reality" is specific to psychological realism: a human psychological reality is any human object that can be thought of and studied via a descriptive scientific or quasi-scientific psychology. The advocates of the "man as scientist" axiom or of psychological realism will necessarily agree with this definition. They will contend that the two dimensions of social judgment are linked to two human psychological realities (two registers of personality-traits). We, however, take the view that they are two basic aspects of people's social value: the value of people in interpersonal contexts and the value of people in activity (work) contexts. It has been asserted (cf. Dubois & Beauvois, 2011) that these two dimensions exist independently of an empirical and even a conceivable human psychological reality, and are activated through implicit personality theories, solely via the retrieval of the described object's social value. This is why the subjects in Dubois and Beauvois' experiment (2011, Study 2) were told to pick personality traits from a list to describe not people but pet rabbits, after they had been given two pieces of information about the mere social value of those rabbits, namely price ([in]expensive) and love ([un]loved). No other information was given. The traits provided for the "description" were located at the positive and negative poles of either the social desirability dimension (likeable, pleasant, warm vs. aggressive, irritable, cold) or the social utility dimension (active, intelligent, dynamic vs. stupid, lazy, weak). Two principal component analyses were performed (PCA).
a. The first PCA (see Table 4) revealed two factors accounting for 50% of the variance. We considered only the first two factors because we wanted to verify that they were in accordance with those brought out in the literature. In addition, we observed an important drop in the explained variance between the second factor (15.85%) and the third factor (10.58%). These two factors were identical to those usually found in psychological description studies. The first factor (accounting for 35.02% of the variance) set high social-desirability traits (likeable, pleasant, warm) against low socialdesirability traits (aggressive, irritable, cold). The second factor (accounting for 15.85% of the variance) set high social-utility traits (dynamic, intelligent, active) against low socialutility traits (lazy, spineless, stupid, weak). b. A second PCA was then performed with price and love (see Table 4). The introduction of these two variables did not change the factor structure. The PCA also revealed two first factors accounting for more than 50% of the variance (35.78% for the first factor and 18.06% for the second factor) 5 . Taken together, these results confirmed the evaluative nature of the factors).
Love was associated with social desirability, price with social utility. The fact that the two classic dimensions of social judgment were obtained in conditions where no descriptive information was available but only information about the rabbits' social value (love: social desirability, and price: social utility) led us to question the descriptive basis of person description. We think that these results validate our claim that person description simply amounts to situating individuals on the two basic dimensions of social value rather than on the two dimensions of a true psychology of Homo sapiens sapiens (psychological realism), a true psychology which thus remains uncharted waters. These dimensions only tell us about the social value of individuals.

Social value of people and social value of everyday objects: From a person to a car
Recent studies (Cambon, 2007;Schiffler, Dubois, & Mollaret, 2010) have shown that the bidimensional evaluation of people bears some similarity to the bidimensional evaluation of everyday objects, such as watches, tables and cars. Readers who recall Osgood's writings will not be surprised (Osgood, Suci, & Tanenbaum, 1957): with his theory of semantic mediation, Osgood would certainly not have been a proponent of psychological realism 6 . When Schiffler et al. (2010) ran a PCA on descriptions of everyday objects, they came up with two dimensions. On the first one, attractive, nice, and amusing were contrasted with horrible, ugly and square. On the second one, luxury, extremely expensive and chic were set against cheap, bottom of the range and poor quality. The first dimension was named "agreeableness" and the second "market value". Results were similar to those obtained with pet rabbits. We can therefore legitimately wonder whether evaluative processes are characterized by a fundamental duality, in that, at least in our societies, these processes distinguish between pleasure derived from contact with an object and the economic value of that object. If this is indeed the case, the social desirabilty and social utility of human beings correspond to the agreeableness and market value of objects. This fundamental duality obviously only fits in with the evaluative conception put forward in this chapter.
Schiffler et al. conducted an experiment to confirm this fundamental duality. Subjects first had to make a decision concerning photos of target individuals' faces. This decision was manipulated through instructions steering them towards the social desirability of these targets (choose the person who has most friends), or their social utility (choose the person who has the highest salary). In the second phase of the experiment, the same subjects had to choose the most typical adjectives for 12 objects (armchair, car, pen, etc.). Adjectives were selected on the basis of the above-mentioned PCA. They were either descriptive or evaluative (either high in agreeableness: amusing, ugly, or high in market value: luxurious, downmarket). Phases 1 and 2 were presented as pertaining to two distinct experiments (priming paradigm). Schiffler et al. expected decisions about target individuals in the first phase to prime the choice of adjectives in the second phase. More specifically, they predicted that the social desirability instruction would drive the choice of agreeableness and that social utility instruction would drive the choice of market value. Results are given in Table 5.
The 2 (kind of priming: social desirability vs. social utility) x 2 (kind of adjectives: agreeableness vs. market value) ANOVA performed on the data (see Table 5 Table 5. Saliency indexes (means and standard deviations) as a function of kind of priming and kind of adjectives in Shiffler, Dubois and Mollaret (2010) However, even if social desirability and social utility are homologous with agreeableness and market value, we cannot bank on their strict equivalence. The fundamental evaluative duality takes diverse forms according to the type of object. The social desirabilty of human beings, for instance, seems to have at least two components, namely morality (honest, sincere) and sociability (helpful, warm). While sociability can be seen as the "human" shape of agreeableness, there seems to be no relevant equivalent of morality for a watch or a rabbit. This is why we can surmise that this fundamental duality results from people's evaluative experience, particularly through evaluative practices, which is then generalized to the evaluation of nonhuman objects.

Social utility is associated with social differentiation
Other research that is typical of the evaluative conception can be summed up in a single proposition: "worthful people" are seen as being socially useful rather than socially desirable. It has been shown that "worthful people" make normative judgments (see Dubois, 2003 andDubois, 2009 for the concept of social judgment norms). For example, "worthful people" have been found to be more internal, more self-sufficient and more optimistic than others. Several studies have shown that the more normative people are in their judgments, the more they are considered and consider themselves to be socially useful. In these studies, subjects were given questionnaires filled out by a person showing himself/herself to be either normative or counternormative in his/her answers. They were then asked to describe, usually with adjectives 7 , the person that was supposed to have filled out the questionnaire. Results showed that the internal targets were seen as more socially useful than the external ones (Cambon, Djouari, & Beauvois, 2006;Dubois & Beauvois, 2005). The self-sufficient targets were also seen as more socially useful that the non-self-sufficient targets (Dubois, 2005;Dubois & Beauvois, 2005), and the optimistic targets were likewise seen as more socially useful than the nonoptimistic or pessimistic targets (Le Barbenchon & Milhabert, 2005;Milhabet, 2011). This link between normativeness and social utility persists in self-descriptions. Cambon et al. asked students to fill out an internality questionnaire, then describe themselves using adjectives. They observed an almost linear relationship between the internality of these students and the social utility they assigned themselves.
This body of research is interesting because it shows that "psychological description", regarded as an evaluative -and as such attributing social utility or social desirability-, is a way of communicating a person's social value when this value is assessed on other criteria. The above-mentioned normativeness of judgment is not the only criterion of people's social value. Other criteria are linked to social structure, in particular the positions occupied by individuals within that structure as well as the profits associated with those positions. We have good reason to think that psychological descriptions vary with these criteria, as we saw with Cambon (2004)'s results. These criteria are discussed below in the organizational context.
We conclude this section with a hypothesis derived from previous research. Significant social differences have been shown to give rise to differences in social utility, but less so in social desirability (professionals working in the production sector are seen as more socially useful but not more socially desirable than professionals working in the maintenance sector; normative individuals are seen as more socially useful but not more socially desirable than counternormative individuals). We can therefore deduce that while social utility is used to corroborate some important social differences, social desirability is used to compensate for a lack of social valorization. Social desirability is thus used by people who are not socially valued as a means of self-promotion, in other words, to earn a judgment that is not too pejorative: I am certainly not dynamic and ambitious (understood: like worthful individuals), but I am honest and helpful (see Kervyn, Yzerbyt & Judd, 2011;Yzerbyt, Kervyn, & Judd, 2008, for compensation effects).

Social desirability and social utility in organizational contexts: Social utility corresponds to people's economic value
This section is divided into two parts. Part 1 deals with the role of social utility in the organizational context. Part 2 discusses the relevance of breaking social utility down into three components, each with a different status in an organization and, more generally, in society.

Recruitment or cooptation?
In a study by Dubois and Aubert (2010, Study 1), executives in charge of a team and who had considerable experience in assessment were told to put themselves in the position of recruiting a target peer described in terms of socially useful or socially desirable adjectives. As predicted (see Table 6, study 1), the socially useful target candidate (active, intelligent, organized, etc.) was judged by the vast majority to be more recruitable than the socially desirable target candidate (agreeable, honest, helpful, etc.): F(1,70)= 80.42, p<.001, η 2 = .535.
In a second study, Dubois and Aubert compared a recruitment condition to a co-optation condition. In this last condition, students were asked to coopt a target person to go on vacation. This target person was described using the same socially useful or socially desirable adjectives as those used in the recruitment condition. The 2(condition: recruitment vs. co-optation) x 2 (target: socially useful vs. socially desirable) between-subjects ANOVA performed on the data (see Table 6, study 2) showed a significant interaction effect, F(1,396)= 1552.445, p<.001, η 2 = .797). Whereas a similar result to that of Study 1 was observed in the recruitment condition (the socially useful target was the one most frequently recruited), an opposite result was found in the friend cooptation situation (the socially desirable target was this time the most frequently coopted).
The results of these two studies led to a simple idea that we have already touched on, namely, that social utility is a relevant selection criterion as far as economic value is concerned (i.e., in organizations and the workplace), but social desirabilty is a relevant selection criterion as far as interpersonal relations are concerned.

Recruitment for a job in production or a job in maintenance?
The above results nevertheless raise an important question. We have already seen (Cambon, 2004) that while social utility is the main value in production, this is not the case in maintenance. Dubois and Aubert's studies therefore needed to be replicated in the light of these two sectors. To this end, Dubois (2010a, Study 4) looked at ordinary employees' ratings of a target applicant's chances of being hired when the applicant was said to be either socially useful (active, intelligent, organized, efficient, hardworking) or socially desirable (agreeable, honest, open-minded, helpful, likeable). Vacant positions were a high-or low-ranking position in either the production sector or the maintenance sector of the economy (head of an automobile company for the high-status production job, milling machine operator for the low-status production job, artistic director at an opera house for the high-status maintenance job, and sports coach at a summer camp for the low-status maintenance job). A 2(types of trait descriptions: socially useful vs. socially desirable) x 2 (job sectors:production vs. maintenance) x 2 (job statuses: high vs. low) ) between-subjects ANOVA performed on the data (see Table 7) showed a significant types of traits x job sectors interaction effect, F(1,53) = 19.42, p < .001, η 2 = .113. The breakdown of this interaction indicated firstly that the type-of-trait effect was only found for the production jobs (F(1,53) = 36.03, p < .001, η 2 = .191): when presented as socially useful, both applicants applying for a production job were rated as more likely to be hired (M = 3.64) than when they were presented as socially desirable (M = 2.42). For the maintenance jobs, whether presented as socially useful (M = 3.30) or socially desirable (M = 3.35), the applicant's chances of being recruited were the same (F(1,53) = .06, ns). The fact that the second-order interaction was nonsignificant (F(1,53) = .70, ns) showed that the jobrank variable did not affect the type-of-trait by job-sector interaction. Such results clearly showed that social utility was the dominant value when the production sector was involved (whether for a job as the head of an automobile company or as a milling machine operator).
There was no such finding for applicants seeking a position in the maintenance sector (whether as an artistic director or a sports coach).

Three components of social utility 8
When we look at the actual meaning of the words, we find that the traits located at the positive poles of the two dimensions are not at all similar. Honesty and helpfulness, for instance, (two traits traditionally found at the positive pole of the first dimension) do not belong to the same register of meaning. Similarly, ambition and competence (two traits found at the positive pole of the second dimension) belong to two different registers, as far as their simple meaning is concerned. These same semantic problems are observed when we consider the negative poles. Why would a shy person (trait located at the negative pole of the second dimension) not be competent (trait located at the positive pole of this same second dimension)? Similarly, why would a liar (located at the negative pole of the first dimension) not be warm (located at the positive pole of this same first dimension)? This observation has led researchers to introduce a distinction between morality and sociability within the first dimension (Brambilla, Rusconi, Sacchi, & Cherubini, 2011). French researchers have proposed several differentiations within the second factor which have recently turned out to be very productive. While certain traits belonging to the second dimension (which we call "social utility" but many name "competence") directly evoke ideas of competence or incompetence (intelligent and capable at the positive pole, and stupid and out of his/her depth at the negative pole), other traits (e.g., courageous at the positive pole, and shy at the negative pole) have nothing to do with (in)competence. This is why the second factor has been broken down into the following three components.
1. effort (in an activity; i.e., persevering). Description of someone who is hardworking, extremely conscientious and always sees his or her work through to the end, even if it is difficult; 2. competence (in an activity; i.e., capable) Description of someone who has considerable abilities, masters techniques well and knows how to resolve problems; 3. ease (in an activity; i.e., competitive). Description of someone who is ambitious, feels at ease with the competition and is never temperamental.
We evoke just two studies here. Dubois (2010b) asked subjects to read the three above descriptions through carefully. The latter were said to be descriptions of three employees. The experimenter then informed the subjects of these employees' status: one was said factory manager, another was said foreman and the third, manual worker. Subjects had to match descriptions and status. In another group, instead of statuses, Dubois gave the salaries of these employees: one earned €6,000 a month, another earned €3,000, and the third €1,500. Similarly, subjects had to match descriptions and salaries. The results (Tables 8a and  8b) support the predictions in a very significant way for the statuses (for « factory manager » : Χ (2) ²= 29.4, p< .001, for « foreman » : Χ (2) ²= 22.4, p< .001, for « manual worker » : Χ (2) ²= 35, p<.001) and in an only marginally way for salaries, although the salaries associated with the three components were in the expected direction (for "€6,000": Χ (2) ²= 2.4, ns, for « €3,000" : Χ (2) ²= 4.2, ns, and for "€1,500": Χ (2) ²= 5,4, p <.07).
These results suggest that the highest status was associated with the "at ease in the activity" portrait and the lowest status was associated with the "effort" portrait and, to a lesser extent that the lowest salary was associated with the "effort" portrait.
These results prompted the author to wonder whether the observed hierarchy of the three components of social utility was specific to one particular organizational culture. A study by Mollaret and Miraucourt (n.d) has now provided an answer to that question. These researchers asked servicemen, public-sector workers and the employees of a large private company to say how much money (7-point rating scale) 75 traits would fetch a member of the organization to which they belonged. The traits were representative either of effort, competence, or ease. The researchers were thus able to calculate how much money the three components were worth in each organization. Ease was found to be the dominant value in the private sector, but not in the other two organizational models under scrutiny. Interestingly, when they asked people encountered in their own homes or in the street, to say what the same 75 traits might bring someone in a general way (without referring to any particular organization), they reproduced the results for the private sector. The results are set out in Table 9.  Table 9. Mean economic value (and standard deviations) for adjectives associated with ease, competence and effort worth in three organizational models and in a random sample in Mollaret and Miraucourt (n.d) "Ease" seemed to be the preferred value in private companies (F(2, 44)= 16.91, p <.001 ; η 2 = .43). The analysis of contrasts showed that subjects choosed significantly more "Ease" than "Competence" and "Effort" (undifferentiated). "Competence" seemed to be the most chosen value by public-sector workers (F(2, 44)= 6.65, p < .01 ; η 2 = .23). The analysis of contrasts showed that subjects choosed significantly more "Competence" than "Ease" and "Effort" (undifferentiated). "Effort" seemed the value prized by servicemen (F(2, 44)= 5.61, p < .01 ; η 2 = .20). The analysis of contrasts showed that subjects choosed significantly more "Effort" than "Ease" and "Competence" (undifferentiated). As we might have expected on the basis of an analysis of ideological processes in our society, the general public seemed to be imbued with the private-sector model (F(2, 44)= 9.67, p <.001 ; η 2 = .30). The analysis of contrasts showed that subjects choosed significantly more "Ease" than "Competence" and "Effort" (undifferentiated).
The research described in this section confirms the idea that social utility is the most relevant dimension in business organizations in the production sector and that it comes in the shape of a hierarchy indicating people's market value in these organizations. Ease is at the top of this hierarchy, with the highest status, and effort at the bottom, with the lowest status. To the best of our knowledge, social desirabilty has yet to give rise to studies that would allow us to organize morality and sociability in a similar hierarchy.
Taken together, these findings can help us to put forward new definitions.

Conclusion
The previous sections should allow readers to judge for themselves the contribution of the empirical bases of the following evaluative definitions of the two factors (the "big two", as they are sometimes called): -A person's social desirability is his or her reputation for arousing positive (vs. negative) affects in others and for going along with (vs. going against) other people's motivations. It has at least two components, namely morality and sociability (or agreeableness); -A person's social utility is his or her reputation for being able (vs. unable) to occupy different positions in social organizations, positions ranged from the least high to the highest. It has at least three components, ranging from the least valued to the most valued, namely effort, competence and ease; -A person's reputation for being more or less desirable and more or less useful is the product of the social evaluative practices to which the person has been subjected. When an evaluation is communicated in the form of personality traits, these traits activate in memory the behaviors that can or should be adopted towards that target person, in other words, what could or should be done with him/her. There are good reasons to think that people internalize the main aspects of their reputation.
The above evaluative conception can be summed up in a basic idea whereby the concepts of naive psychology are not intended to describe human beeing from a psychological, scientific or quasi-scientific point of view, but instead to evaluate them as social actors or agents. The dimensional structure of implicit personality theories does not reflect the structure of human psychology, but rather the two main registers of social values that people realize (or should realize) through their behavior (e.g., depending on whether they are hardworking or lazy). We do not claim that people are stupid or irrational. The fact of not being centred on the intrinsic, descriptive properties of objects and people must not be considered as an indication of stupidity. On the contrary, we claim that people behave as fine analysts when they deal with things that are important to them, namely, the value of people and things, even if the intrinsic nature of those people and things is not their main preoccupation 9 . The register of the social values is not a register which is used by default. It is not established by imaginary whims. This register is made by realities. People refer to these realities to hire somebody, to pay him/her, to associate him/her with projects of work or leisure activities 9 Moreover, if our subjects are perfectly capable of mobilizing their I.P.T. and to produce both dimensions of the social value of the persons when we incite them to psychologize by asking them to describe pet rabbits with personality traits (Dubois & Beauvois, 2011, study 2, see above) , equivalent subjects, placed in quite a different situation, a situation of purchase on the market (Dubois & Beauvois, 2011, study 3), recognize that nobody would pay a high price for a rabbit because it is intelligent. In this situation, the subjects deal with the social value of a rabbit, without psychologizing. They say that we can pay a high price for a rabbit if it is beautiful and if it is of pure race. What corresponds to the reality of the market. and even to marry him/her. In numerous domains of the life and the social life (agriculture, crafts, interpersonal relations…), people learnt at first t o handle this register of s ocial realities without adopting a scientific way of thinking. Science came much later, when it came: at this beginning of XXI° century, it is not sure that science came in psychology of individual differences. Indeed, what is presented as individual differences in personality are in fact differences in the social value of people. We can even wonder if a scientific psychology of the individual differences will not have to give up the concept of "personality" which is maybe too much filled with evaluative perception of people.
We conclude with the two main advantages of this evaluative conception.
The evaluative conception does not rely on any of the assumptions about the validity of the psychology used by laypeople as a tool of descriptive or scientific knowledge, assumptions which, since the fifties, and despite Heider (1958), have often been criticized by social psychologists either directly or indirectly. In particular, this conception does not require naive psychologists (1) to be aware of the psychological nature of their own personality or that of their target, (2) to have spontaneous access to that nature, or (3) to speak about that nature as an intuitive scientist would speak (which they never do for other objects of knowledge, except obviously when the person who speaks is a scientist in the exercise of his/her work).
The evaluative conception considers people in terms of their concrete social existence. The research reported in this chapter supports the idea that the evaluative approach is indeed derived from a psychology that is possibly scientific, but definitely social. People are evaluated on a daily basis in social situations which involve them as agents or social actors. They are not evaluated as human beings. We do not need to rely on what is, for us, a theoretical fantasy, namely the existence of a universal subject who, in a social vacuum, attempts to pin down the authentic nature of objects (and more particularly of persons) by relying solely on his/her cognitive abilities which predispose him/her to be an intuitive scientist. When people judge a person, they always consider this person in the social relationships which determine the social values which serve as reference to their judgment: this person is either a pupil or a worker or a teacher, or an artist… However, the dominant thinking in academic psychology well and truly considers the person judging (the teacher) and the judged person (the pupil) as if they were two homo sapiens sapiens. In fact, they never are homo sapiens sapiens but they are persons inserted into social situations involving roles and particular objectives. In brief, they are actors or social agents. In addition, describing a homo sapiens sapiens is not the same as judging a social agent or actor. Recognizing that we evaluate actors or social agents and not human beings would lead to question important postulates of the research on the "personality". Among these postulates, the most important, which is also the basis of what social psychologists have called the fundamental error , i s t h a t p e o p l e a l w a y s e x p r e s s t h e i r psychological nature to explain what they do. Accept the idea that they express especially their position of actor or social agent led to more concrete and less essentialist theoretical views.