Finalement, finalement,
Il nous fallut bien du talent
Pour être vieux sens être adulte
…

Et plus le temps nous fait cortège
Et plus le temps nous fait tourment,
Mais n’est-ce pas le pire piège
Que vivre en paix pour des amants.
Bien sûr tu pleures un peu moins tôt,
Je me déchire un peu plus tard
Nous protégeons moins nos mystères;
On laisse moins faire le hasard
On se méfie du fil de l’eau
Mais c’est toujours la tendre guerre

1 – Introduction

In this article I intend to show the relation between a proposed theoretical model, the construct of collusion, and the experimental procedure that I developed in the years leading up to this proposal; the development of such research enabled the conceptual foundation to be laid for the proposed model.
I am referring to the model of collusion, proposed by Carli and Paniccia (1981) at the beginning of the 1980s and later refined and improved so as to create the basis for analysis of the demand and more generally to put forward a psychoanalytic theory of the social bond and living together.
The unit of analysis in this model is the relationship, not the single individual. The model is based on collusive affective symbolization, i.e. on the affective symbolization concerning the same context, which creates the bond between those sharing the context. We have repeatedly underlined that the context is “cultural” and not structural1; the co-occurrence of affective symbolizations does not require, therefore, the co-presence of those who are part of the collusive phenomenon within a definite time and space. Collusion is a phenomenon that is found in groups, organizations, cultural dimensions; it is a phenomenon underpinning social relations.
What I would like to examine in more depth is the experimental experience in the domain of social and clinical psychology, which made it possible to lay the basis for the formulation of this construct.

2 – Cinema participation

In 1970 Ancona and Carli published their research on the different effects of a film stimulus (Cronaca familiare2 by Zurlini), which previous research had shown to be particularly effective in increasing the need for achievement measured with McClelland’s Test of Imagination3, when the film was watched in groups or alone.
A phenomenon that emerged clearly was what we then called “cinema participation”: a low participation was hypothesized when viewing alone or in those who adopted an attitude of isolation from others in the group viewing (these tended to be field-independent in the Embedded Figure Test or H. A. Witkins’ EFT). With the hypothesis of low cinema participation the annulling of the film’s psychodynamic effect could be explained, that is, the lack of increase in the need for achievement, in the two conditions mentioned. On the other hand, a participation at the highest levels was hypothesized in group viewing, but only in people with a high capacity for emotional integration with others (field-dependent subjects in Witkins’ EFT): here there was a significant increase in the need for achievement after seeing the film. As can be seen, we hypothesized the cinema participation to explain data that depended on an individual variable (the single subjects’ field dependence or independence) but at the same time on a contextual variable, i.e. the condition of viewing film “alone” or “in a group; the individual variable (field dependence – field independence) was not able to totally explain such a different effect of the film if one saw it in a shared cinema or if one saw it in the same room but alone and with only one seat in an otherwise empty hall. At the time we talked about an effect defined as “the assumption of the presence of others showing the same motivational dynamics” at the viewing of the film.
The experimental set-up of the research, which followed in the footsteps of previous research in the areas of the psychological effects of watching films, was clear: viewing the film was, and this was realised even then, a “social” event; cinemas were a meeting place for collective enjoyment. The psychological experiment, however, envisaged that the experimenter’s unit of analysis was the single individual. Also McClelland’s Test of Imagination, though setting out to measure the motivational level of the social bond, was a test with strictly individual administration, initialling and scoring. The same was true of Witkins’ test. Experimental evidence, too, concerned single individuals, though in the hypothesis that as general laws they could cover all the individuals responding to specific characteristics. It was on the basis of these premises that the research projects just mentioned were carried out.
I remember one autumn evening in the mid-sixties: after a talk by Leonardo Ancona at the ‘Istituto Filmologico’ in Milan4 on the psychological effects caused by viewing a film, Cesare Musatti approached him as he was talking to our group of students at the school of specialization in Clinical Psychology, and commented that he would be curious to see how people seeing the film alone would have reacted. There was obvious interest in a research project that set out to analyse the effects of watching the “television” film: at the time television was becoming more and more important in the lives of Italians. Musatti’s idea struck both Ancona and myself: it brought into question and challenged, if you like, the implicit individualist paradigm supporting the research into film viewing conducted up to that moment. We decided to take up Musatti’s suggestion and to set up an experimental situation in which single people, in a cinema setting that underlined their “solitude”, would see one of the films that had proven capable of arousing deep changes in the spectators’ motivational dynamics. We arranged to have all the seats removed from the cinema of the ‘Film Institute’, leaving in the centre of the room only one seat for the “subject” who would see Cronaca Familiare. Albinoni’s adagio for strings and organ in G-minor (the same that Pasolini was to use the year after Zurlini, in the film La rabbia of 1963) accompanied the ending of the film, and at the moving sound of that music, I rushed into the room to ask the “subject” to take part in some psychological tests. The same experiment was later repeated with about 60 people who saw the film all together in the same room of the Institute. These spectators too were subjected to the battery of tests established in the experiment plan.
The results of the research created great debate among both us experimenters and our fellow psychologists. Musatti’s casually expressed suspicions were confirmed: on the other hand, it was not a matter of a difference between those who saw the film alone or those who saw it in a traditional situation, in a room with many other people. The latter, too, could fail to undergo the effect of the film. The field independent subjects did not show the emotional responses shown by the field dependent subjects, in the “group vision” experimental mode5. There were multiple variables involved: the film stimulus with the emotional involvement that that particular film entailed for the “single spectator”; the way of viewing the film, alone or in a group; the field dependent or field independent cognitive style of the single people. The departure point for all these variables was the single individual. What the research revealed, however, could not be explained solely with the aspects related to the single individual. For this reason we were “forced” to hypothesize a phenomenon that we called “cinema participation” and that we defined as the capacity to “assume the presence of others” while viewing the film, to make sense of something that did not belong to the single individuals but that also depended on the characteristics of the single individuals, specifically their cognitive style. The construct of collusion was as yet lacking: with such a construct, “cinema participation” could have been explained more simply as the shared emotional symbolization of the film stimulus on the part of those with greater “propensity” for this, being field dependent. We shall shortly return to this relation between cognitive style and collusion.
The fact remains that the “cinema participation” had effects that were hard to comprehend solely with individualist psychology. In order to explain the effect of the film on the motivational state in the group viewing alone, though limited to field-dependent subjects, a construct was needed that was connected to the symbolic emotional bond between people and context; the context was obviously represented by the film stimulus and at the same time by the “group” situation of the film viewing. This can explain our interest in research projects which in the mid-‘60s6, had made it possible to look critically at a psychology strictly and a-critically anchored to the “single” individual and to his psychological traits of a psychometric type. These traits did not take into account the context in which the individual lived out his experience and which characterized the experience itself. Hence the need to look to models and constructs that could explain the symbolic relation between individuals in a specific clearly defined context.

3 – Assumed Similarity and the “generalized other”

Let us now discuss a research project from 1968. The subject of the research is the construct of Assumed Similarity (A.S.) proposed by Fiedler. The A.S. between two people, who we will call A and B, indicates the distance between the way person A sees himself and the way A thinks person B sees himself. Fiedler had put forward an A.S. measurement tool consisting of a series of adjectives, followed by a five-step assessment scale. A is asked to use the scale to describe himself, and then to use the same scale to describe how he thinks person B sees himself (for the sake of simplicity A can be given the following instruction: “try to gues how B would use the scale to describe himself”). The A.S. is an index that is computed as the square root of the sum of the squares of the differences between the self description scores and those of the anticipated self description of the other person, for every single adjective on the scale. Research in analysis shows that A.S., when applied to small groups of people that know each other, leads to a result not foreseen by Fiedler, and interesting in its regularity. Think of a group of 6 people who know each other due to work or study habits, for example. Each group member’s Assumed Similiarity to all the others can be measured, obtaining 30 measurements of Assumed Similarity (n x n-1). Now, by analysing the variance it was shown that for 130 “groups” of various kinds (companies, schools, military, friends) for a total of about 800 subjects, each subject was characterized by a stable A.S. score, with little variation, towards all the other subjects in the group. Let us try to be more specific. The experimental situation is that of a small group, where the individuals are used to working together, with strong bonds linking them within a specific context. If the A.S. of each group member with respect to all the others is measured, it emerges that each member tends to situate the others in general at a constant distance; the differences therefore only concern the way the single individuals situate “all the others” at a greater or lesser distance. This way of seeing all the other components of the group with a constant A.S. score made it possible to talk about the “generalized other”; an “other” who was situated at a greater or lesser distance. In the same research it was also shown that the more a person tends to perceive the “generalized other” as being different from him/herself (high A.S. index), the more he/she will be capable of effective social behaviour, for example correctly anticipating the sociometric behaviour of the group members towards him/her.
What does this research tell us in the perspective we are adopting? It tells us that in the A.S. sphere, we do not differentiate between others by adopting different degrees of Similarity; instead, we tend to situate all the members of the small groups in which we work, study and enjoy free time and friendship, at a single, constant A.S. level: we perceive others not in different terms, but at a stable degree of “distance”, seeing them as the “generalized other”. Every person uses a different A.S. level for the generalized other. From this it emerges that for each of us, in the small groups that we frequent for a variety of reasons, the generalized other is situated at a greater or lesser interpersonal distance, assuming the A.S. as an indicator of interpersonal distance. In other words, we perceive to what extent, in our view, the other sees himself near or far from the way we see ourselves. This fact could be said to indicate a sort of measurement of the perception of “otherness” in our social life. We hypothesize that underlying the assumption of similarity there is an emotional symbolization of the generalized other, experienced emotionally as “other from self” and therefore situated at a specific distance, as a function of our mode of relating socially. Now, this stable A.S. towards the generalized other can be seen as an aspect motivating the collusion dynamic, as a personal “code”, different from person to person, through which one colludes with the others in the groups in which one is involved in living together. Let us consider the previous research and the different way the spectators “in the group” responded to the film stimulus and modified their motivational state. It could now be said that the people with a high A.S. towards the generalized other are also those that are more sensitive to the film stimulus as a collusive response; while those with a lower A.S., therefore with greater interpersonal distance between themselves and the generalized other, are less sensitive to the modification of their motivational state after being exposed to the film stimulus. While the first research underlines the collusion dynamic, this second research allows us to differentiate the way of participating in the collusion itself in different individuals, in relation to the symbolization of the context in terms of greater or lesser interpersonal distance. But it also enables collusion to find its experimental grounding, in that the stable A.S. towards the “generalized other” is simply the outcome of the emotional symbolization of the other, in social relations.

4 – Boss and underboss

Let us look at the question posed by the third research project: the film stimulus can, as we have seen, modify the motivational level creating responses that are consistent with the emotional content of the film; can the same film stimulus influence institutional relations, relations between people who share specific contexts, creating new dynamics and new modes of relating? More specifically, can a film stimulus change the dynamics of relating with the authorities?
Pirio Esposito and I (Carli & Esposito, 1971) studied the psychological effect manifested on exposure to a stressful film stimulus7 by groups of professional military trainees, characterized by a strong identification with the authorities controlling their training and their promotion from training to profession. The groups, studied with double blind methodology, were made up of young professional military trainees in an academy that trains one of the components of the armed forces in our country. The hypothesis was that in the groups8 that saw the stressful film without the presence of the authorities, there would be an attenuation both of the defensive cohesion within the group and of the idealization of authority. The phenomena were studied with the tools of Fiedler’s Assumed Similarity and with the test of reaction to frustration, Rosenzweig’s P.F.T. (modified by me and Trentini, personifying the frustrating figure and the frustrated one as a function of the different aims of the research; here the frustrating figure was the authority organizing the training, the frustrated one was the trainee). In short, the research allowed the finding that “the stressful film, in the groups without structural Authorities, is able to destructure the process of identification with authority, measured by Fiedler’s A.S., due to a deterioration of the authority image that went beyond the threshold of personal and social acceptance by the group members. This process of attenuation of the identification with authority involves a parallel destructurating of the group’s internal defensive cohesion and favours relations based on acceptance of “otherness” (Pagès, 1968) and on “a greater social realism” (p. 282). These findings are also confirmed by the Picture Frustration Test, where exposure to the film without the presence of the authorities reduces the trainees’ identification with the authority figure, seen as punitive, aggressive and intolerant; the trainees therefore de-idealize the violent, rigid authority with which they previously tended to identify, and take their distance, showing a reaction to frustration that is more capable of overcoming the frustrating obstacle. Let us look at the variables considered in the research. We are in a context of training for military life, i.e. a context where the relations with the authorities have a particular importance, not only for the usual functions of coordination and of communication of knowledge, but above all as a figure of identification on which to model one’s social formation and to anticipate one’s function and image in the future system of living together. What is this authority? The image attributed to the authorities by the young trainees is that of an aggressive figure uninterested in the other and his needs, punitive and rigid in his judgements and in his punitive reactions. There is the tendency to identify with this image, creating a strong defensive cohesion in horizontal relations, organized around the identification with the punitive authorities. Obviously we are talking about relations, not about people as single individuals. But this description uses an “individualist” category (identification) to make sense of a phenomenon that pertains not to the individual but to relating. On the other hand, “defensive cohesion” is a psychological dimension that can be placed on the side of relations rather than on that of the individual. But the latter is a post hoc category describing an event deriving from experimental data (reduction of the interpersonal distance between all the members of the experimental group), rather than a model that can make sense of the event considered.
Let us now think of the experimental variable: the viewing of a film, critical of authority, with or without the presence of formal authorities. Why does the viewing of a critical film without the presence of the authorities prove to be capable of producing changes in the relations between members of the experimental group and in the representation of authority? The research enables us to record such change in the area of Assumed Similarity between group members towards the authorities, but there are no categories to make sense of the finding. This is because the data reflects the individual dimension, although all the research deals with changes in the relations between people, not with changes in the people themselves. The phenomena recorded at the time were difficult to explain without the idea, which had not yet been elaborated, of collusive symbolization of the context itself. The context was represented by the group subjected to the stressful film without the presence of the formal authorities. The plan of the experiment, before obtaining the results, talked about the importance of the context for an understanding of the event examined, the modification of the relations with the authorities by means of participating in the film. While in the first research mentioned above, the subjects of the experiment were indiscriminately “common” people, though classed by sex, age and social class, this research is related to a specific context where the subjects of the experiment as a whole and their membership of a specific military academy makes it research into the effect of  viewing a film tied to a specific context. Above, I defined the context as a set of affective collusive symbolizations underlying the phantasmatic relations that allow a specific historical experience. Research like that currently being analysed have made a decisive contribution to the establishment of this definition. Let us think about what happens in the research: a group of trainees sees a film critical towards the authorities, without the authoritarian figure both during the film and during the compilation of the test afterwards. The film leads to a modification of the collusive symbolizations of authority thanks to its highly critical content, in that it represents an absolutely intolerable, violent authority; these modifications are shared also thanks to the absence of authority. This facilitates the collusive identification of the film’s violent authority with the violent authority of the collusive context, usually shared in the formative experience. And the historical experience? This is an important point. The research was in fact a research-intervention, although at the time we were not conscious of this. In effect, collusive identification with a violent, rigid authority was based on the phantasmatic relations symbolizing authority, in some respects independent of how authority really presents itself in the life of the academy. We were dealing with a typical case of collusive phantasmatic anticipation of reality, with the effect of actually conditioning the authority figure. This research, while limited to a few experimental groups, made it possible to question the phantasmatic representation of authority, to de-idealize it and to create a more critical collusive situation that can make people think about the function of authority. In the medium term, this changed the attitude towards the authorities also in those participating in the research, and led those in charge of the academy to ask for a psychological intervention to make the change that had occurred in the trainee group become stable and repeatable during the years of training. The research gave rise to a demand for psychological consulting and intervention in the real life of the academy.

5 – Signal Detection Theory

From 1967, I studied Moreno’s sociometric test as an indicator of group phenomena and not of strictly individual phenomena. I used Signal Detection Theory9 as a model for the reading of the data produced by the sociometric test, no longer in individual but in cultural dimensions referring to the whole school class. With Mosca (Carli & Mosca, 1980) I put forward the indexes of cohesion and of exploration efficiency, indicators of dynamics concerning the ways of relating in the class. Numerous research projects, including those carried out abroad, followed the model proposed, which confirmed the two indexes as effective indicators of the development process of the school group, and which was adopted in numerous assessments of the psychology interventions in the class group.
Signal Detection Theory was examined in depth also in its “social” implications, distinguishing the two factors that, in the theoretical construct under analysis, make up the decision in a situation of uncertainty: d’ and criterion.
This is the diagram showing the possible outcomes of the decision in a situation of uncertainty. The hypothesis shown in the diagram envisages a YES/NO decision about the presence or absence of a signal amid background noise; think of a radar operator on a military ship in wartime, with the highly luminous screen in front of him: the decision concerns the presence or absence of the signal that indicates the arrival of an enemy plane on the luminous screen. If the radar operator says YES and the enemy plane really is present, there is a HIT (the target is hit); if he says YES and the plane is not present but there is only the luminosity of the screen, there is an “error” of FALSE ALARM. If he says NO and the plane is not present, there is a CORRECT REFUSAL of the signal; if he says NO but the plane is present, there is the “error” MISS.

The optimal decision strategy depends firstly on the degree of uncertainty of the decision, that is, on the higher or lower overlapping of the curves of probability defining on the one hand, the ‘noise’ (or background interference: the luminousness of the screen, for instance), on the other hand, the signal plus the ‘noise’. The value of d’ indicates the distance between the median points of the two probability curves, therefore their degree of overlap. This is an element characterizing “objective” conditions of uncertainty. The other factor comes from the criterion, i.e. from the point on the x-axis, on the plane with the two curves just mentioned, where the decision-maker decides on the presence of the signal. In view of the situation of uncertainty, it is not possible to make a decision that does not involve an error. In the strategy adopted, one must see if the reduction of FALSE ALARMS is favoured, by adopting a restrictive criterion for the YES answer, or if one wants to reduce the probability of a MISS error occurring, by adopting a restrictive criterion for the NO answer.

The two strategies have a different degree of efficacy if they are adopted in situations of different levels of uncertainty, as in the case of the two figures above. With the same criterion about FALSE ALARMS, restrictive in this case, it can be seen that when the curves are apart (first figure), the probability of F. A. can be reduced without greatly diminishing one’s chance of exploring the signal curve. On the other hand, in the second case, the same criterion keeps the probability of FALSE ALARM low, but “pays” for this with an only partial exploration of the signal distribution curve. This means that when uncertainty is high, the adoption of a criterion that keeps the probability of F.A. low, does not permit learning, i.e. exploration, trying out and testing oneself in signal analysis, and accepts FALSE ALARMS.
If d’ is a cognitive factor, as it were, characterizing the degree of uncertainty that each decision-maker experiences in decisions, the second is a social factor, based on the cultural assumption of the point beyond which it is acceptable to commit the error of FALSE ALARM and before which it is acceptable to commit the error of MISS. In various research projects (see for example: Carli, Canarecci, & Frare, 1973) it was experimentally demonstrated that the scholastic performance of primary and middle school pupils depended, for a small part, on the individual’s cognitive skills, and for the most part on the relationship between teachers and the class group. In particular, the teacher - class group relationship was important because it could influence the decision about the value of the criterion adopted in situations of uncertainty. In an intervention carried out over five years (from the mid-seventies to the end of the decade) in about twenty middle schools in Rome, a drastic reduction in the cases of “scholastic maladjustment” was achieved by training the teachers to change the model of communicating the criterion: the teachers learnt to accept low values of the criterion (with greater probability that the pupils would commit FALSE ALARM errors) in the early stages of learning, to then adopt higher criterion values in the later stages (Carli, Guerra, Cannizzo, & Daini, 1980). These criterion values were lower when the discrimination point shifted to the left on the x-axis, and higher when the point shifted towards the right, always with reference to the two curves of probability of noise and signal (plus noise) distribution.
For the teachers this meant accepting that in the early stages of learning the pupils could make errors, could “explore” the area of knowledge, receiving feedback on their exploration rather than a negative evaluation of the errors committed. Feedback, or if you like, verification, took the place of evaluation. This entailed a great change in the scholastic relationship: for instance it meant questioning the identification between what a person “does” and what a person “is”: accepting that a pupil can explore and make mistakes, means accepting a process of progressive learning through trial and error; it means giving the teacher the role of facilitator of the exploration, casting off that of implacable judge in search of errors. It was the entire collusive system in the scholastic relationship that could change, into a relationship that agrees to deal with pupils from the most disadvantaged social classes and therefore not ready for learning without errors. This entailed hours of work with the teachers, during which the emotional symbolizations of the pupils’ performances were discussed, with greater tolerance for errors and for trials, encouragement of exploration, and not with disincentivation from the teachers especially in the early phases of learning.
This was an example of how experimental research in psychology could give the intervention a direction, providing teachers with a measuring stick by which to assess the consequences of their decisions for their pupils.

6 – Conclusions

Fiedler’s Assumed Similarity, Signal Detection Theory, Rosenzweig’s Picture Frustration Test, Sociometric Test, McClelland’s Test of Imagination: it was these, along with other tools and theoretical models, that oriented my research from the mid-‘60s right through the ‘70s. The main aim was to go beyond the individualist paradigm, which had proven so weak in explaining experimental phenomena, starting from the difference between the group and individual response to the stimulus of film. The whole experimental complex underlines that there must necessarily be dimensions concerning interaction in social group relations; without a hypothesis on these relational aspects, a great deal of experimental evidence cannot be explained or understood. Think of “cinema participation”: how can one explain the experimental relevance of confining the effect of a film stimulus on the spectators’ motivational state only to high participation, i.e. seeing the film with other spectators, and only to those who denote a high field dependence? In groups of military trainees, when a stressful film is seen what causes the change in the indexes of defensive cohesion and of identification with the authorities? In a sociometric test, what makes the indexes of cohesion and exploration efficiency in one school class different from another? What factor inhibits or facilitates exploration in the early stages of learning, making learning enjoyable for those who are gifted but come from disadvantaged social and cultural classes, thus reducing the rejection of learning classified post hoc as “scholastic maladjustment”?
Think of the Assumed Similarity construct: the basic element is the social bond, though the evidence is still tied to individual dimensions. To stay within Fiedler’s theorization, it is obvious that social relations will be different if we think the other sees himself in a similar or different way from how we see ourselves. But how to explain the experimental data that each of us, in groups or organized structures, has the tendency to place others in general at a constant distance, be it nearer or farther; and that this tendency to see the generalized other as being more or less similar to how we see ourselves, has a very strong influence on the efficiency of groups and their development dynamics?
It seemed impossible to answer these questions, with the experimental research data of the time. Something was missing. But all the data led in the same direction: in the relationship there must be “something” that could not be related to single individuals; the point was not to consider the relation as the outcome of individual characteristics relating to each other; this option, traditional in scientific psychological research, showed all its limits in the experimental situations that were gradually being defined. It is when we approached psychoanalysis, and in particular the works of Ignacio Matte Blanco on the mind’s characteristics and its way of being unconscious, that the solution to the problems raised by the experimental work gradually took shape. The principles of generalization and symmetry put forward by the Chilean psychoanalyst10 and the resulting investigation of the mind’s way of being unconscious did not concern the “individual mind”. It is the unconscious of Freud’s first topography, revisited and examined in depth by Matte Blanco, that is the grounds for the new mode of organizing and trying out the idea of affective symbolization, which was being put forward in the same years by the psychoanalyst Franco Fornari. Reading the works of these two great thinkers, the intense debate with them, and the fact that their proposals offered the chance to make an “emotional” study of social relations, gave rise to my hypothesis on the notion of collusion. Collusion enabled us to explain a great deal of data from psychosocial research and many aspects of the phenomenology emerging from systematic experimental studies of small group interaction. While, on the other hand, collusion could explain experimental data, the construct of collusion also had to find experimental backing. Here, moreover, a change of level was necessary in experimental studies and models. It was in non-narrative language, in the texts produced within the relationship, “collusively” if you like, that the collusion had to be sought. Hence the creation of the idea of Local Culture as a collusive dynamic characterizing specific institutional dimensions; examination of the texts produced in the Local Cultures led to the formulation of the research methodology that we have called Emotional Text Analysis (Carli & Paniccia, 2001) and that we believe to be the most suitable experimental methodology so far for showing and measuring the collusive processes in a relationship.
We have thus arrived at the present day. I have recalled the experimental work, albeit in brief and only considering some of the numerous works that were produced during the sixties and seventies; this work was the foundation for the formulation of the notion of collusion. Just as important as the experimental work of those years was my participation, intense and rewarding, in the theoretical and methodological movement of French psychosociology. I am thinking of the equivocal way in which, at the beginning of the sixties, one talked about the “group” in psychology. The word ‘group’ made no distinction between the experimental group, the analysis group, the discussion group, the group of psychologists, economists or doctors, the phantasmic group, the real group, the middle management group in a firm or the executive group, and many other things. At the time T Groups were organized with people who at least in theory did not know each other and who, under the guidance of a trainer, spent ten or fifteen days together in a residential experience of interaction, with marked emotional connotations. French psychosociology strongly opposed the possibility of transferring what was learnt in such an experience to one’s own usual organizational and social life; it advocated, instead, psychosocial interventions in organizations wanting to deal with their internal cultural problems. It was within the psychosocial intervention that the notion of collusion became useful and capable of making sense of the psychological function.
I remember one summer afternoon in 1976, in Palermo, at the home of Gigliola Lo Cascio, having a meeting with the teaching assistants I was working with at the Psychology Institute in via Divisi. We were discussing the poverty of the conceptual tools in general psychology, when one talked about social relations and wanted to capture the emotional dimension. It was during that discussion that I suggested the term collusion, to indicate the emotional and symbolic relation that characterizes social relations seen from a psychological viewpoint. I later regretted using a term which in our language indicates above all a criminal agreement between people. But when I thought of the word, its etymology was present in my mind, cum ludere, meaning ‘to play together’, and also the various declinations that the verb ludere assumes in combination with different prefixes: illusion, allude, delude… and collude. It is interesting to notice that in all these variations of the word, ludere seems to indicate a social dimension, and at the same time to be closely tied to the symbolic emotional dimension.

References

Ancona, L., & Bertini, M. (1963). Aggressivity Discharge Effect through Films Causing strong Emotional Stress. Ikon, 19 (57), 7-27.

Ancona, L., & Carli, R. (1970). La dinamica della partecipazione cinematografica [The dynamic of cinematographic partecipation]. Contributi dell’Istituto di Psicologia U.C.S.C., 30, 21-45.

Ancona, L., & Croce, M.A. (1967). Dinamica psichica e dinamismo cinematografico [Psychical dynamic and cinematographic dynamism]. Arch. Psicol. Neurol. Psich., 28, 85-106.

Atkinson, J.W. (1964). An Introduction to Motivation. Pinceton N.Y: Van Nostrand (trad it. La motivazione, Il Mulino, 1973).

Beretta, A. (1968). TDS: la teoria della detezione del segnale [TDS: Sign’s detection theory]. Firenze: OS.

Carli, R. (1968). La Similarità Assunta come indice di adattamento sociale [The assumed similarità as index of social adaptation]. Annali di Psicologia, 1, 1-3.

Carli, R., & Esposito, P. (1971). Contributo sperimentale allo studio della dinamica di gruppo e delle relazioni con l’autorità nella partecipazione cinematografica [An experimental contribute to the study of group’s dynamic and of the relationship with the authority in the cinematographic partecipation]. Arch. Psicol., Neurol.,  Psich., 32, 231-285.

Carli, R., Canarecci, P., & Frare, M. (1973). Livello scolare, classe sociale e dinamica decisionale [Scholar level, social class and decisional dynamic]. Arch. Psicol., Neurol., Psich., 34 (6), 541-570.

Carli, R., Guerra, G., Cannizzo, G., & Daini, S. (1980). Aggiornamento degli insegnanti: Una proposta di intervento psicosociale [Improvement of teachers: a proposition of psychosocial intervention]. Firenze: La Nuova Italia.

Carli, R., & Mosca, A. (1980). Gruppo e istituzione a scuola [Group and institution at school]. Torino: Boringhieri.

Carli, R., & Paniccia, R. M. (1981). Psicosociologia delle organizzazioni e delle istituzioni [Psychosociology of organizations and institutions]. Bologna: Il Mulino.

Carli, R., & Paniccia, R. M. (2001). L’Analisi Emozionale del Testo [Emotional text analysis]. Milano: FrancoAngeli.

Fiedler, F.E. (1954). Assumed similarity measures as predictors of team effectiveness. J. Abn. Soc. Psychol., 49, 381-388.

Fiedler, F.E. (1958). Leaders attitudes and group effectiveness. Urbana: University of Illinois Press.

Matte Blanco, I. (1975). The Unconscious as Infinite Sets. An Essay in Bi-Logic.  London: Duckworth & Company (trad. it. L’inconscio come insiemi infiniti, Einaudi, Torino, 1981).

McClelland, D.C., Atkinson, J.W., Clark, R.A. & Lowell, E.L. (1953). Analysis of imaginative stories for motivational content. In D.C McClelland, J.W Atkinson, R.A Clark, E.L. Lowell, The achievement motive. Century psychology series (pp. 107-138). East Norwalk, CT, US: Appleton-Century-Crofts.

Moreno, J. L. (1943). Who shall survive? A new approach to the problem of human interrelations. Washington: Nervous and mental disease Publ. Co.

Rosenzweig, S., Fleming, E.E., & Clark, H.J. (1947). Revised Scoring Manual for the Rosenzweig Picture-Frustration Study. St. Louis: Saul Rosenzweig Pbl.

Rosenzweig, S., & Kogan, K.L. (1949). Psychodiagnosis; an introduction to tests in the clinical practice of psychodynamics. N.Y.:Grune Stratton.

Rosenzweig, S.(1992). PFS: Picture-frustration study: manuale integrato delle tre forme per adulti, bambini e adolescenti. Firenze: OS.

Pagès, M. (1968). La vie affective des groups [Emotional groups life]. Paris: Dunod.

Witkin, H.A., Dyk R.B., Faterson, M.F., Goodenough, D.R., & Karp, S.A.. (1962). Psychological Differentiation. New York: Wiley (trad. it. La Differenziazione psicologica, Roma, Bulzoni, 1962).

Notes

* Full Professor of Clinical Psychology, Faculty of Psychology1, "La Sapienza" University of Rome, Italy. President of the degree course in "Clinical Intervention for persona, group, institutions", Ordinary Member of the Italian Psychoanalytic Society and of the International Psychoanalytic Association. Top

1. A definition of context, consistent with what I am saying, could be: the set of affective collusive symbolizations underlying the phantasmatic relations through which a specific past  relational experience is lived. Top

2. Family chronicle. Top

3. Previous research had shown that the cinema viewing of specific films that could determine a strong emotional involvement in the spectator, could produce great changes in the spectators’ motivational state, in the period immediately after the viewing. Ancona and Bertini (1963) had found that watching a film with a high motor dynamic content, if experienced with great personal participation, led to a “cathartic discharge” of aggressivity. Ancona and Croce (1967) had specified the relation between the content of the film and the effects on the spectator’s psychic dynamics. The emotional content of the film had specific effects on motivational dynamics, measured using McClelland’s model: the film contributed to a sort of reduction of the motivational level involved in watching the film and at the same time to an increase in the complementary need.  The film Ivan the terrible by Ejzenstejn (1942), in which the dynamics of power were strongly involved, led to a significant reduction in the spectators’ need for power, straight after seeing the film, and at the same time to an increase in the need for affiliation; the film The guns of Navarone by Thompson (1961) led to a reduction in the need for achievement and at the same time, again an increase in the need for affiliation; the film Cronaca familiare by Zurlini (1962) led to a reduction in the need for affiliation and at the same time a significant increase in the need for achievement. Top

4. The ‘Istituto Filmologico’ (Film Institute) began its activity in 1960 in Milan, under the auspices of the Milan Provincial authorities, the  ‘Centro Nazionale Prevenzione e Difesa Sociale’,  the State University and the Catholic University of Milan. When Agostino Gemelli died, Musatti, who had been co-founder with Gemelli and Cohén-Seat, suggested calling it the ‘A. Gemelli Institute’; on Musatti’s death, those in charge of the Institute decided to call it the ‘A. Gemelli and C.  Musatti Institute of Research into Communication’. The Institute publishes the journal Ikon – Forme e processi del comunicare, progeny of the publication Révue Internationale de Filmologie. Top

5. It is interesting to notice that the term “group” is used here as a synonym of “set of people who watch a film together in the same room”. There is no reference to the relations between the people, or to the dynamics characterizing their relations.  And yet the fact that this term was used at that time seems to foreshadow an implicit relation between the spectators, that can configure the film as a stimulus for shared emotional responses. Top

6. The research we are talking about was published in 1970, but was carried out in 1966-1967. Top

7. This is a 15’ short film entitled Il Padrone, in which there is the dramatic representation of a game typical of the southern Italian culture, that of the padrone (boss) – sottopadrone (underboss): in the film, the boss character sadistically humiliates a young countryman whose wife he has previously stolen, violating the most traditional values, personal and family honour, in the presence of a consenting group of townsmen, which increases the social violence of the frustrating situation. The film is shown to the groups participating in the research without the cathartic ending of the original version.
The game, of Sicilian origin, is called patrónu j sóttu, and is recorded in medieval documents as the game ad passatellam. The spectators crowd around the players with enough red wine paid for all of them placed on the table with glasses. After the cards are dealt and the boss chosen, the latter chooses the underboss and the game begins. The boss has absolute power over the distribution of wine while the underboss only has the power conferred on him by the boss; the latter can cheerfully offer wine to the whole group, ask the underboss’s opinion in filling the glasses, give wine only to some players, withholding it from others who are then considered ùlumi i.e. left dry; the boss can also drink all the wine himself with the underboss, or force one player to drink it all, until he is blind drunk, amidst the sniggers of the crowd. Being made ùlumu is an offence, almost an insult that is tolerated and accepted because those are the rules of the game, strict and to be respected unless one wants to lose one’s personal honour, but not to be forgotten. It is a game and remains such, like life in the hands of fate; after all, there is always the probability that the next time the cards will fall differently, and that today’s “dry one” will be tomorrow’s boss. Top

8. The experiment envisaged that some groups would see the stressful film just described, others a film presenting the history and landscape of a region in the south of the country. Within these two experimental situations (stressful film – non stressful film) the experimental groups were further subdivided: some participated in the experimental situation in the presence of formal authorities, some without this presence. Top

9. At the time, Angelo Beretta (1968) published an interesting, detailed text applying SDT to psychology. Thanks to studies conducted with Beretta, I was able to use SDT in many areas of clinical and social psychology, including the reading of sociometric data as indicators of the social relations in a specific group, i.e. the school group. Top

10. Matte Blanco was Chilean by birth and culture, but he had trained in psychoanalysis with frequent, long stays in England and the United States. At the time of his most important work (1975) Matte Blanco was living and working in Rome. Top