Introduction: The Power of Ideas
1 A SOCIAL PSYCHOLOGY OF KNOWLEDGE
Imagine you are looking at an outline map of Europe, with no features marked on it except for the city of Vienna near the centre, and to the north of it the city of Berlin. Where would you then locate the cities of Prague and Budapest? For most people who have grown up since the end of the Second World War both these cities belong to the eastern division of Europe, while Vienna belongs to the West, and consequently both Prague and Budapest should be to the east of Vienna. But now look at a map of Europe and see the actual locations of these cities. Budapest, to be sure, lies further east, downstream along the Danube from Vienna. But Prague lies in fact to the west of Vienna.
This small example illustrates something of the phenomena of social representations.
Our image of the geography of Europe has been reconstructed in terms of the political division of the Cold War, in which the ideological definitions of East and West have come to be substituted for the geographical ones. We can also observe in this example how patterns of communication in the post-war years have influenced this process and stabilized a particular image of Europe. Of course, in the West there has been a fear and anxiety about the East which antedated the Second World War, and which persists even today, a decade after the fall of the Berlin Wall and the end of the Cold War.
But where this representation of a divided Europe in the post-war years had its most powerful influence was in the eclipse of the old image of Mitteleuropa, of a Central Europe embracing the heartlands of the old Austro-Hungarian Empire and stretching northwards towards Berlin. It was this Central Europe, which was dismembered by the Cold War which also ideologically repositioned Prague to the east of 'Western’ Vienna.
Today the idea of Mitteleuropa is again being discussed, but perhaps the sense of the eastern “otherness” has marked the image of Prague so clearly that it may take a long time before these new patterns of communication reposition the city back to the west of Vienna.
As well as illustrating the role of communication and influence in the process of social representation, this example also illustrates the way in which representations become common sense. They enter into the ordinary and everyday world, which we inhabit and discuss with our friends and colleagues, and they circulate in the media we read and watch. In short, representations sustained by the social influences of communication constitute the realistics of our daily lives and serve as the principal means for establishing the affiliations through which we are bound to one another.
For more than four decades Serge Moscovici, together with his colleagues, has advanced and developed the study of social representations, and this collection brings together some of the central essays drawn from a much larger body of work which has appeared over these years.
Some of these essays have appeared in English before, while some are translated here for the first time. Together they illustrate the ways in which Moscovici has elaborated and defended the theory of social representations, while, in the concluding interview with Ivana Marková, he provides the main elements of the history of his own intellectual itinerary.
At the heart of this project has been the idea of constructing a social psychology of knowledge, and it is within the context of this wider project that his work on social representations needs to be viewed.
What, then, might a social psychology of knowledge look like? What terrain would it seek to explore, and what would be the key features of this terrain? Moscovici himself introduced this theme in the following way:
There are numerous sciences, which study the way in which people handle, distribute and represent knowledge. But the study of how and why people share knowledge and thereby constitute their common reality, of how they transform ideas into practice – in a word, the power of ideas – is the specific problem of social psychology. (Moscovici, 1990a, p. 164)
Thus, from the perspective of social psychology, knowledge is never a simple description or a copy of a state of affairs. Rather, knowledge is always produced through interaction and communication, and its expression is always linked to the human interests, which are engaged. Knowledge emerges from the world in which people meet and interact, the world in which human interests, needs and desires find expression, satisfaction or frustration.
In short, knowledge arises from human passions and, as such, is never disinterested; rather, it is always the product of particular groups of people who find themselves in specific circumstances in which they are engaged in definite projects (cf. Bauer and Gaskell, 1999). A social psychology of knowledge is concerned with the processes through which knowledge is generated, transformed and projected into the social world.
2 A LA RECHERCHE DES CONCEPTS PERDUS
Moscovici introduced the concept of social representation in his pioneering study of the ways in which psychoanalysis had penetrated popular thought in France. However, the work in which this study is reported, La Psychanalyse: son image et son public, first published in French in 1961 (with a second, much revised edition in 1976), remains untranslated into English, a state of affairs which has contributed to the problematic reception of the theory of social representations in the Anglo-Saxon world.
Of course, an English version of this text would not in itself resolve all the differences between Moscovici’s ideas and the dominant patterns of social-psychological thinking in Britain and the USA, but it would at least have helped to reduce the number of misunderstandings of Moscovici’s work which have added a penumbra of confusion to discussions of these ideas in English.
More than this, however, the lack of a translation means that a predominantly monolingual Anglo-Saxon culture has not had access to a text in which the key themes and ideas in the theory of social representations are presented and elaborated in the vital context of a specific research study.
When these ideas are put to work in structuring a research project and in ordering and making intelligible the mass of empirical data which emerge, they also take on a concrete sense which is only weakly visible in more abstract theoretical or programmatic texts.
But if Moscovici’s work has been obscured in the Anglo-Saxon world, the concept of social representation itself has had a problematic history within social psychology.
Indeed, Moscovici entitled the opening chapter of La Psychanalyse 'Social representation: a lost concept’, and introduces his work in these terms: Social representations are almost tangible entities. They circulate, intersect and crystallize continuously, through a word, a gesture, or a meeting in our daily world.
They impregnate most of our established social relations, the objects we produce or consume, and the communications we exchange. We know that they correspond, on one hand, to the symbolic substance which enters into their elaboration, and on the other to the practice which produces this substance, much as science or myth corresponds to a scientific or mythical practice.
But if the reality of social representations is easy to grasp, the concept is not. There are many good reasons why this is so. For the most part they are historical, which is why we must entrust historians with the task of discovering them. The non-historical reasons can all be reduced to a single one: its 'mixed’ position at the crossroads between a series of sociological concepts and a series of psychological concepts.
It is at this crossroads that we have to situate ourselves. The route certainly has something pedantic about it, but we can see no other way of freeing such a concept from its glorious past, of revitalizing it and understanding its specificity. (/1976, pp. 40-1; my translation)
The primary point of departure for this intellectual journey, however, has been Moscovici’s insistence on a recognition of the existence of social representations as a characteristic form of knowledge in our era, or, as he puts it, an insistence on considering 'as a phenomenon what was previously seen as a concept’ (chapter 1, p. 17).
Of course, to develop a theory of social representations implies that the second step of the journey must be to begin to conceptualize this phenomenon. But before turning to this second step, 1 want to pause for a moment at the first step and ask what it can mean to consider as a phenomenon what was previously seen as a concept, since what might appear as a small aperçu in fact contains some characteristic Moscovician tropes.
First of all, there is a boldness in this idea which is not inhibited from expressing a conclusive generalization, a generalization which also has the effect of radically separating Moscovici’s conception of the aims and scope of social psychology from the predominant forms of the discipline.
More precisely, here Moscovici is affiliating himself with a strand of social-psychological thinking which has always been a minority or marginal strand within a discipline dominated in this century first by behaviourism, and more recently by a no less reductive cognitivism, and throughout this time by a thoroughgoing individualism.
Yet in its origins social psychology was formed around a different set of concepts. If Wilhelm Wundt is mostly remembered today as the founder of experimental psychology, he is also increasingly recognized for the contribution his Voelkerpsychologie made to the establishment of social psychology (Danziger, 1990; Farr, 1996; Jahoda, 1992).
For all its faults, Wundt’s theory nevertheless clearly situated social psychology at the same crossroads between sociological and psychological concepts indicated by Moscovici. Far from opening a productive line of researched theory, Wundt’s work was soon eclipsed by a growing mainstream of psychological thought which rejected any association with the 'social’ as compromising the scientific status of psychology.
What Danziger (1979) has termed the 'positivist” repudiation of Wundt served to ensure the exclusion of the social from the theoretical purview of the emerging social psychology.
At least, this was the case in what Farr (1996) has called its 'psychological’ form, but, as he also notes, a 'sociological’ form has persisted as well, stemming largely from Mead’s work, in which Wundt’ s Voekerpsycho1ogie was a major influence (and one should add that a concern with the social is also characteristic of Vygotsky’s psychology; cf. chapters 3 and 6). Indeed, Farr has gone on to suggest that Durkheim’s (/1974) radical separation of 'individual’ from 'collective’ representations contributed to the institutionalization of a crisis for social psychology, which persists today.
Throughout the last century, whenever 'social’ forms of social psychology have emerged we have witnessed the same drama of exclusion which marked the reception of Wundt’s work. A 'compulsion to repeat’ masks a kind of ideological neurosis, which has been mobilized whenever the social has threatened to invade the psychological. Or, to shift from a Freudian to an anthropological metaphor, the purity of the social has consistently represented a polluting danger to escientific psychology.
Why has it proved so difficult to establish a social psychology which can embrace both the social and the psychological? Although in the quotation above Moscovici suggested that this was a question for historians, he himself has contributed something to clarifying this enigma, as several of the texts collected here bear witness (see chapters 1, 2, 3 and 7).
In a major historical essay, The Invention of Society, Moscovici (/1993) offers a further set of considerations by addressing the complementary question of why psycho logical explanations have been seen as illegitimate in sociological theory. Durkheim formulated this idea explicitly in his aphorism that 'every time that a social phenomenon is directly explained by a psychological phenomenon we may be sure that the explanation is false’ (/1982, p. 129).
But, as Moscovici shows, this prescription against psychological explanation not only runs like a connecting thread through the work of classical writers on modem social theory, it is just as surreptitiously contradicted in these very same texts.
For in producing social explanations for social phenomena, these sociologists (Weber and Simmel are the examples analysed by Moscovici alongside Durkheim) also need to introduce some reference to psychological processes to provide coherence and integrity to their analyses. In short, in this work Moscovici is able to demonstrate through his own analyses of these founding texts of modem sociology that the explanatory frame required for making social phenomena intelligible must include psychological as well as sociological concepts.
Yet the question of why it has been so difficult to achieve a stable theoretical frame embracing both the psychological and the social remains obscure. To be sure, there has been just as much hostility to ‘sociologism’ among psychologists as there has been towards 'psychologism’ among sociologists.
To say that social psychology as a mixed category represents a form of pollution remains descriptive as long as we do not understand why the social and the psychological are considered to be exclusive categories. This is the heart of the historical conundrum, and it retains its distinctive power even today. While it would be naïve to pretend to offer a clear account of its origin, we can glimpse something of its history in the opposition between reason and culture which, as Gellner (1992) argues, has been so influential since Descartes’s formulation of rationalism.
Against the relativism of culture, Descartes proclaimed the certainty stemming from reason. The argument for the cogito introduced a scepticism about the influences of culture and the social which has been difficult to overcome. Indeed, if Gellner is right in seeing in this argument an opposition between culture and reason, then any science of culture will be a science of unreason. From here it is a short step to becoming an unreasonable science, which seems to be the reputation earned by every attempt to combine sociological and psychological concepts in a 1 mixed’ science. Yet it is just such an 'unreasonable science’ which Moscovici has sought to resuscitate through a return to the concept of representation as central to a social psychology of knowledge.
3 DURKHEIM THE AMBIGUOUS ANCESTOR
In seeking to establish a 'mixed’ science centred on the concept of representation, Moscovici has acknowledged an enduring debt to the work of Durkheim. However, as we saw above, Durkheim’s formulation of the concept of collective representations has proved an ambiguous inheritance for social psychology. The effort to establish sociology as an autonomous science led Durkheim to argue for a radical separation between individual and collective representations, and to the suggestion that the former should be the province of psychology while the latter formed the object of sociology (interestingly, in some of his writings on this theme Durkheim toyed with the idea of calling this science 'social psychology’, but preferred 'sociology’ in order to eliminate any possible confusion with psychology; cf. Durkheim,
It is not only Farr who has noticed the difficulties, which Durkheim’s formulation has carried for social psychology. In an earlier discussion of the relation between Durkheim’s work and the theory of social representations, Irwin Deutscher (1984) also wrote of the complexity in taking Durkheim as an ancestor for a social-psychological theory.
Moscovici himself has suggested that in preferring the term 'social’ he wanted to emphasize the dynamic quality of representations as against the rather fixed or static character which they have in the Durkheimian account (cf. chapter 1, where Moscovici illustrates the way in which Durkheim. used the terms 'social’ and. 'collective’ interchangeably).
In commenting on this point further in his interview with Marková in chapter 7, Moscovici refers to the impossibility of sustaining any clear distinction between the 'social’ and the 'collective’.
These two terms do not refer to distinct orders in the arrangement of human society, but neither is it the case that the terms 'social representation’ and 'collective representation’ only mark a distinction without establishing a difference. In other words, Moscovici’s social psychology cannot simply be collapsed into a variant of Durkheimian sociology.
How then should we understand the relation of social representations to the Durkheimian concept? From a social-psychological perspective one might be tempted to think that the resolution of this ambiguity could be sought through a clarification of the terms 'individual’ and 'collective’ as they are used in Durkheim’s argument.
However, it is by no means clear that such an endeavour could successfully reclaim some theoretical space for a social psychology, particularly since, as Farr (1998) points out, the question is rendered problematic by the recognition of individualism as a powerful collective representation in modem society.
A more productive approach can be seen through a further reflection on Durkheim’s argument itself. Durkheim was not simply concerned to establish the sui generis character of collective representations as one element in his effort to sustain sociology as an autonomous science. His whole sociology is itself consistently oriented to what it is that holds societies together, that is, to the forces and structures, which can conserve or preserve the whole against any fragmentation or disintegration.
It is within this perspective that collective representations take on their sociological significance for Durkheìm; their constraining power helps to integrate society and conserve it. Indeed, it is partly this capacity for sustaining and conserving the social whole which gives collective representations their sacred character in Durkheim’s discussion of The Elementary Forms of Religious Life (1912/1995).
Moscovici’s social psychology, on the other hand, has been consistently oriented to questions of how things change in society, that is, to those social processes through which novelty and innovation become as much a part of social life as conservation or preservation. I have already alluded to his interest in the transformation of common sense in his study of social representations of psychoanalysis. It is in the course of such transformations that anchoring and objectification become significant processes (cf. chapter 1).
A clearer statement of this focus of Moscovici’s work can be found in his study of social influence (1976) which, in fact, carries the title Social Influence and Social Change. The point of departure for this study was his dissatisfaction with models of social influence which apprehended only conformity or compliance. If this were the only process of social influence, which existed, how would any social change be possible?
Such considerations led Moscovici to an interest in the process of minority influence or innovation, an interest he has pursued through a series of experimental investigations. It is this concern with innovation and social change which also led Moscovici to see that, from a socialpsychological perspective, representations cannot be taken for granted, nor can they serve simply as explanatory variables.
Rather, from this perspective it is the construction of these representations which is the issue to be addressed, hence his insistence both on considering 'as a phenomenon what was previously viewed as a concept’ and in emphasizing the dynamic character of social representations against the static character of collective representations in Durkheim’s formulation (a more extended discussion of this point by Moscovici can be found in chapter 1).
Thus, where Durkheim looks to collective representations as the stable forms of collective understanding with the power of constraint which can serve to integrate society as a whole, Moscovici has been more concerned to explore the variation and diversity of collective ideas in modem societies.
This diversity itself reflects the lack of homogeneity within modem societies, in which differences reflect an unequal distribution of power and generate a heterogeneity of representations. Within any culture there are points of tension, even of fracture, and it is around these points of cleavage in the representational system of a culture that new social representations emerge.
In other words, at these points of cleavage there is a lack of meaning, a point where the unfamiliar appears, and just as nature abhors a vacuum, so culture abhors an absence of meaning, setting in train some kind of representational work to familiarize the unfamiliar so as to re-establish a sense of stability (cf. Moscovici’s discussion of unfamiliarity as a source of social representations in chapter 1).
Cleavages in meaning can occur in many ways. It can be very dramatic, as we all saw as we watched the fall of the Berlin wall and felt the structures of meaning which had held a settled view of the world since the end of the war evaporate. Or again, the sudden appearance of new and threatening phenomena, such as HIV/AIDS, can be the occasion for representational work. More frequently social representations emerge around enduring points of conflict within the representational structures of culture itself, for example, in the tension between the formal recognition of the universality of the 'rights of man’ and their denial to particular groups within society. The struggles which have ensued have also been struggles for new forms of representations.
The phenomenon of social representations is thus linked to the social processes woven around differences in society. And it is in giving an account of this linkage that Moscovici has suggested that social representations are the form of collective ideation in conditions of modernity, a formulation which implies that under other conditions of social life the form of collective ideation may also be different.
In presenting his theory of social representations, Moscovici has often drawn this contrast (cf. chapter 1), and at times suggested that it is a major reason for preferring the term 'social’ to Durkheim’s collective’. There is an allusion here to a complex historical account of the emergence of social representations which Moscovici sketches in only very lightly and, without attempting to provide a more detailed or extensive account, it will be helpful in understanding something of the character of social representations to draw attention here to two related aspects of this historical transformation.
Modernity always stands in relation to some past, which is considered as traditional, and while it would be mistaken (as Bartlett, 1923, saw so presciently) to consider premodern – or traditional – societies as effectively homogeneous, the central thread in Moscovici’s argument about the transformation of the forms of collective ideation in the transition to modernity is concerned with the question of legitimation.
In pre-modern societies (which in this context means feudal society in Europe, although this point may also be relevant to other forms of premodem society) it is the centralized institutions of Church and State, Bishop and King, which stand at the apex of the hierarchy of power and regulate the legitimation of knowledge and beliefs.
Indeed, within feudal society the very inequalities between different social layers within this hierarchy were recognized as legitimate. Modernity, in contrast, is characterized by more diverse centres of power, which claim authority and legitimacy, so that the regulation of knowledge and belief is no longer exercised in the same way. The phenomenon of social representations can, in this sense, be seen as the form in which collective life has adapted to decentred conditions of legitimation.
Among the new forms of knowledge and belief, which have emerged in the modem world, science has been an important source, but so too, as Moscovici reminds us, has common sense. Legitimacy is no longer guaranteed by divine intervention, but becomes part of a more complex and contested social dynamic in which representations of different groups in society seek to establish a hegemony.
The transition to modernity is also characterized by the central role of new forms of communication, originating with the development of the printing press and the diffusion of literacy. The emergence of new forms of mass communication (cf. Thompson, 1995) has both generated new possibilities for the circulation of ideas and also drawn wider social groups into the process of the psychosocial production of knowledge.
This theme is too complex to deal with adequately here, except to say that in his analysis of the different forms of representation of psychoanalysis in the French mass media, Moscovici (1961/1976) shows how propagation, propaganda and diffusion take the forms they do because different social groups represent psychoanalysis in different ways, and seek to structure different kinds of communication about this object through these different forms.
Each of these forms seeks to extend the influence of a particular representation, and each of them, also claims its own legitimacy for the representation it promotes. It is the production and circulation of ideas within these diffuse forms of communication that distinguishes the modem era from the pre-modem, and helps to distinguish social representations as the form of collective ideation distinct from the autocratic and theocratic forms of feudal society. Questions of legitimation and communication serve to emphasize a sense of the heterogeneity of modern social life, a view which has helped to give research on social representations a distinct focus on the emergence of new forms of representation.
4 SOCIAL REPRESENTATIONS AND SOCIAL PSYCHOLOGY
The reception of the theory of social representations within the broader discipline of social psychology has been both fragmentary and problematic. If one looks back to the 'golden era’ of social psychology, one can see a certain affinity between Moscovici’s work and that of such predecessors as Kurt Lewin, Solomon Asch, Fritz Heider or, perhaps the last representative of that era, Leon Festinger – an affinity rather than a similarity, for while Moscovici’s work shares with these predecessors a concern with analysing the relations between social processes and psychological forms, his work retains a distinctive quality, just as these authors differ from one another.
Nevertheless, it is not hard to imagine the possibility of a productive conversation based on this affinity. But it is difficult to imagine such a productive conversation across the discipline of social psychology existing today, where the predominance of information processing paradigms and the emergence of varieties of 'post- modernist’ forms of social psychology have increased the segmentation of the field.
Moscovici (1 984b) himself has suggested that contemporary social psychology continues to exhibit a kind of discontinuous development of changing and shifting paradigms, 'lonely paradigms’ as he describes them. Within this flux each paradigm appears more or less disconnected from its predecessors and leaves little trace on its successors.
In this context it has been the common fate of theoretical interventions in social psychology to flicker briefly before passing into a kind of shadow land at the margins of a discipline which has shifted its centre towards the next paradigm, leaving little time for ideas to be assimilated and turned to productive use. From this point of view there is something remarkable in the persistence of the theory of social representations over a period of forty years.
In spite of its problematic relation to the shifting terrain of the mainstream of the discipline, the theory of social representations has survived and prospered. It has become not only one of the most enduring theoretical contributions in social psychology, but also one that is widely diffused across the world.
In his discussion of paradigms in social psychology, Moscovici goes on to argue that: concepts that operate at great depths seem to take over fifty years to penetrate the lowest layers of a scientific community. This is why most of us are only now beginning to sense the meaning of certain ideas that have been germinating in sociology, psychology and anthropology since the dawn of the century. (Moscovici, 1984b, p. 941)
It is this constellation of ideas that forms the focus for some of the essays in this collection (see especially chapters 3 and 6 and the interview in chapter 7), and within which the theory of social representations has taken shape.
To appreciate the distinctiveness of Moscovici’s contribution it is important to recall first of all what his social-psychological innovation reacted against. The cognitive revolution in psychology initiated in the 1950s legitimated the reintroduction of mentalistic concepts which has been proscribed by the more militant forms of behaviourism which had dominated the first half of the twentieth century, and subsequently, the idea of representations has been a central element in the emergence of cognitive science in the past two decades.
But from this perspective, representation has generally been viewed in a very restricted sense as the mental construction of an external object. While this has allowed the development of an informational calculus in which representations have been central terms, the social or symbolic character of representations has rarely figured in such theories.
To return for a moment to the example of the map of Europe, while contemporary forms of cognitive science might recognize the displacement of Prague in popular representations, it has no concepts through which to grasp the significance of this displacement nor the influences of the social processes, which underlie it. At best, such a displacement would appear as one of the many 'biases’ in ordinary thinking, which have been documented in theories of social cognition.
But whereas such theories in social psychology have discussed 'biases’ as examples of how ordinary thought departs from the systematic logic of science, from the point of view of social representations they are seen as forms of knowledge produced and sustained by particular social groups in a given historical conjuncture (cf. Farr, 1998).
Thus, whereas the 'classical’ forms of cognitive psychology (including social cognition, which has become the predominant contemporary form of social psychology) treat representation as a static element of cognitive organization, in the theory of social representation the concept of representation itself is given a more dynamic sense, referring as much to the process through which representations are elaborated as to the structures of knowledge which are established.
Indeed, it is through its articulation of the relation between process and structure in the genesis and organization of representations that the theory offers a perspective in social psychology distinct from that of social cognition (cf. Jovchelovitch, 1996). For Moscovici the source of this relationship lies in the function of representations themselves. Echoing earlier formulations by McDougall and Bartlett, Moscovici argues that 'the purpose of all representations is to make something unfamiliar, or unfamiliarity itself, familiar (cf. chapter 1, p. 37).
Familiarization is always a constructive process of anchoring and objectification (cf. chapter 1) through which the unfamiliar is given a place within our familiar world. But the same operation, which constructs an object in this way is also constitutive of the subject (the correlative construction of subject and object in the dialectic of knowledge was also a characteristic feature of Jean Piaget’s genetic psychology and Lucien Goldmann’s genetic structuralism).
Social representations emerge, not merely as a way of understanding a particular object, but also as a form in which the subject (individual or group) achieves a measure of definition, an identity function which is one way in which representations express a symbolic value (something which also lends Moscovici’s notion of familiarization an inflection which is distinct from McDougall or BartIett).
In the words of Moscovici’s long-term colleague Denise Jodelet, representation is 'a form of practical knowledge [savoir] connecting a subject to an object’ (Jodelet, 1989, p. 43, my translation), and she goes on to note that 'qualifying this knowledge as “practical” refers to the experience, from which it is produced, to the frameworks and conditions in which it is produced, and above all to the fact that representation is used for acting in the world and on others’ (Jodelet, 1989, pp. 43-4, my translation).
Representations are always the product of interaction and communication, and they take their particular shape and form at any moment as a consequence of the specific balance of these processes of social influence.
There is a subtle relationship here between representations and communicative influences, which Moscovici identifies when he defines a social representation as: a system of values, ideas and practices with a twofold function: first, to establish an order which will enable individuals to orientate themselves in their material and social world and to master it; and secondly to enable communication to take place among the members of a community by providing them with a code for social exchange and a code for naming and classifying unambiguously the various aspects of their world and their individual and group history. (1976, p. xiii)
This relationship, between representation and communication may well be the most controversial aspect of Moscovici’s theory, and in his own work it is most clearly expressed in the second part of his study of La Psychanalyse, the analysis of representations in the French media that 1 described above (and this is one point where an understanding of the theory of social representations has been most seriously hampered by the lack of an English translation of this text, as Willeim Doise (1993) has noted; this section of the book has rarely figured in Anglo-Saxon discussions of the theory).
In relation to cognitive psychology it is not difficult to see why this conception should be controversial, since the enduring weight of the idea of psychology as a natural science focused on processes removed from the polluting influence of the social has made the idea that our beliefs or actions may be formed out of such influences all but unthinkable.
Of course, Moscovici’s is not the first psychology to propose such a theme. Freudian psychoanalysis, for instance, has sought the origins of thoughts in libidinal processes, which, especially for the school of object relations, reflect the child’s early experiences in the world of others (cf. Jovchelovitch, 1996).
Mead too could be said to have made a similar argument in his analysis of the development of the self (cf. Moscovici, 1990b). But Moscovici’s work does not address the libidinal origins of our thoughts (though Lucien Goldmann (1976) has made a suggestive parallel between the organization of psycho- analytic and social constructions), nor is he primarily concerned with the interpersonal sources of the self.
His main focus has been to argue not simply that collective ideation is organized and structured in terms of representations, but that this organization and structure is both shaped by the communicative influences at work in society and at the same time serves to make communication possible.
Representations may be the product of communication, but it is also the case that without representation there could be no communication. Precisely because of this interconnection representations can also change, the stability of their organization and structure is dependent on the consistency and constancy of the patterns of communication, which sustain them.
Changing human interests can generate new forms of communication resulting in innovation and the emergence of new representations. Representations in this sense are structures, which have achieved stability through the transformation of an earlier structure.
If the perspective offered by the theory of social representations has generally been too sharply contrasted with the mainstream of the discipline for a constructive dialogue to emerge (although an interest in it is beginning to emerge in the United States; cf. Deaux and Philogene, 2000), what has been both more surprising and more disappointing has been the reception of the theory among those currents of social-psychological thought which have been its neighbours in this marginal shadowland.
With some notable exceptions (e.g. Billig5 1988, 1993; Harré, 1984, 1998, which have entered into a dialogue of constructive engagement from rhetorical and discursive perspectives) most commentaries from outside the mainstream have been antagonistic or even hostile to the theory of social representations (see, for example, the catalogue of objections in the recent contribution from Potter and Edwards, 1999).
There is no space here to give a systematic account of all the criticisms levelled at Moscovici’s work, but a focus on some key themes will not only serve to give a flavor of the issues raised, but also to elaborate a little further some of the central characteristics of the theory itself.
In one sense, as I mentioned earlier, Moscovici’s work formed part of the European perspective in social psychology, which emerged in the 1960s and 1970s. However, looking back at this work now one also notices the differences within this 'European’ outlook. For example, the collection edited by Israel and Tajfel (1972, a work often cited as a primary source for the European view, and for which chapter 2 of this collection was Moscovici’s contribution) appears now to be characterized as much by the diversity of their views as by a common critical spirit among the contributors.
Some of the strongest criticism of social representations has come from Gustav Jahoda (1988; see also Moscovici’s 1988 reply), who, belongs to the same generation of social psychologists as Moscovici, and who has made his own contribution to the 'European’ tradition. For Jahoda, far from helping to illuminate the problems of social psychology, the theory of social representations has rather served to obscure them. In particular, he finds the theory vague in the construction of its concepts, a charge which has been an important theme in discussions of social representations, surfacing again recently in a more sympathetic commentary by Jan Smedslund (1998; see also Duveen, 1998).
Vagueness, of course, is largely a matter of a point of view. Where one writer finds a theory to be so lacking in precision as to present nothing more than a series of chimeras, for other writers the same theory can open new pathways for considering old problems.
Thus Jahoda suggests that, shorn of its rhetoric, the theory of social representations contributes little which is not already contained in the traditional social psychology of attitudes. But, as Japars and Fraser (1984) have shown, while the original formulation of the concept of social attitudes in the work of Thomas and Znaniecki (1918-20) may have some important similarities with the concept of social representations, the concept of attitude has itself undergone a considerable transformation in subsequent socialpsychological theories.
In this transformation the idea of attitude has been stripped of its social and symbolic content and origins. In contemporary social psychology, attitudes appear as individual cognitive or motivational dispositions, so that the idea of an inherent connection between communication and representation has evaporated. If research in social representations has continued to employ some of the technology of attitude measurement, it has sought to 'frame these attitudes as part of a broader representational structure (see also the discussion – of the relations between attitudes and representations in the interview in chapter 7).
From another perspective, the more radical strands of discourse theory in social psychology (e.g. Potter and Edwards, 1999) have objected to the idea of representation itself as being a lingering attachment to 'modernist’ cognitive psychology. From this point of view all social psychological processes resolve themselves into the effects of discourse, and the fleeting achievements and reformulations of identity, which it sustains.
It is the activity of discourse alone which can be the object of study in this form of social psychology, and any talk of structure and organization at the cognitive level appears as a concession to the hegemony of information-processing models (and it matters little for these critics that the theory of social representations has always insisted on the symbolic character of cognition; see also Moscovici’s comments in the interview in chapter 7).
Here the vagueness of social representations is held to be its insufficiently radical departure from a 'mentalistic’ discourse, but as Jovchelovitch (1996) has observed, the rush to evacuate the mental from the discourse of social psychology is leading to the re-creation of a form of behaviourism.
Whatever its critics might suggest, the theory of social representations has appeared sufficiently clear and precise to support and sustain a growing body of research across diverse areas of social psychology. Indeed, from a different point of view one could argue that research on social representations has contributed as much as if not more than other work in social psychology to our understanding of a wide range of social phenomena (such as the public understanding of science, popular ideas of health and illness, conceptions of madness, or the development of gender identities, to name but a few).
Nevertheless, the insistence with which the charge of vagueness has been levelled against the theory deserves some further consideration. Some sense of what is intended by this characterization of the theory can be identified by considering some of the central research studies it has inspired. In addition to Moscovici’s own study of representations of psychoanalysis, Denise Jodelet’s (/1991; see also chapter 1) study of social representations of madness in a French village offers a second paradigmatic example of research in this field.
Methodologically these two studies adopt quite different approaches (indicating the importance of what Moscovici has referred to as the significance of 'methodological polytheism’). Moscovici employed survey methods and content analysis while Jodelet’s study is based on ethnography and interviews. What both studies share, however, is a similar research strategy in which the initial step is the establishment of a critical distance from the everyday world of common sense in which representations circulate.
If social representations serve to familiarize the unfamiliar, then the first task of a scientific study of representations is to make the familiar unfamiliar in order that they may be grasped as phenomena and described through whatever methodological techniques may be appropriate in particular circumstances. Description, of course, is never independent of the conceptualization of phenomena, and in this sense the theory of social representations provides the interpretative framework both for making representations visible and for rendering them intelligible as forms of social practice.
The question of vagueness can be seen to be largely a methodological issue, since it refers primarily to what different social-psychological perspectives render visible and intelligible. In this respect different perspectives in social psychology operate with different criteria and conditions. Armed with the conceptual apparatus of traditional social psychology, one will struggle to see anything other than attitudes, just as the discursive perspective will uncover only the effects of discourse in social-psychological processes. Each of these approaches operates within a more or less hermetically sealed theoretical universe.
Within each perspective there is a conceptual order which brings clarity and stability to the communication within it (each perspective, we could say, 'establishes its own code for social exchange’). What lies outside a particular perspective appears vague, and the harbinger of disorder. This, of 9, course, is no more than an expression of the enduring crisis in the discipline of social psychology, which continues to exist as a set of 'lonely paradigms’.
Recognizing this state of affairs by itself confers no special or privileged status on the theory of social representations. What gives Moscovici’s work its particular interest, and the reason why it continues to command attention, is that his work on social representations forms part of a broader enterprise to establish (or re-establish) the foundations for a discipline which is both social and psychological.
5 TOWARDS A GENETIC SOCIAL PSYCHOLOGY
From this point of view it is important to situate Moscovici’s studies of social representations within the context of his work as a whole, since it is as part of a wider contribution to social psychology that this work remains of capital significance. I have already alluded to the sense in which his work has expressed a critical and innovative spirit in relation to the discipline, and in this sense it also contributed to a wider critical reappraisal of the dominant forms of social psychology, which began in the 1960s and was for a time associated with a distinctively European perspective on the discipline (something of this critical spirit is evident in many of the chapters in this collection, but particularly in chapter 2 and the interview in chapter 7).
What has marked Moscovici’s contribution as innovative is that it has not been limited to a negative critique of the weaknesses and shortcomings of the predominant forms of social psychology, but has always rather sought to elaborate a positive alternative. In this respect, it is also important to recognize that while the theory of social representations has been one centre of this theoretical endeavour, Moscovici’s work has ranged more widely across social psychology, encompassing studies of crowd psychology, conspiracy and collective decisions, as well as the work on social influence.
In all of these contributions one finds the same inspiration at work, a particular form of what we might describe as the 'social psychological imagination’. If Moscovici’s work can be seen as offering a distinct perspective on social psychology, it is a perspective which is broader than what is connoted simply by the term social representations, although this term has often been taken as emblematic of this perspective.
Moscovici himself has only rarely ventured into efforts at articulating the interconnections between these different areas of work (though the interview in chapter 7 offers some significant thoughts). In part this reflects the fact that each of these areas of work has been articulated through different methodological procedures. His studies of social influence and group processes, for instance, have been rigorously experimental, while his study of the crowd drew on a critical analysis of earlier conceptualizations of mass psychology.
In part it may also reflect the sense that these studies focus on different levels of analysis, from face-to-face interaction through to mass communication and the circulation of collective ideas. Yet all these studies seem to be 'pregnant’ with the ideas, which have been articulated around the concept of social representations so that a focus on this concept can indicate something of his underlying perspective.
In this respect the essay on Proust in chapter 5 offers an illuminating study of the intricacies of the relations between influence and representation. Another example is his critical analysis of Weber’s discussion of the Protestant ethic in The Invention of Society (Moscovici ([ 1988]/1993).
What is apparent in both of these essays is that influence is always directed at sustaining or changing representations, while conversely, specific representations become stabilized through a balance achieved in a particular pattern of influence processes.
Here, as in the studies of decision-making in groups, it is the relationship between communication and representation, which is central.
In his book on social influence, Moscovici (1976) identified the perspective he described as a 'genetic social psychology’ to emphasize the sense that influence processes emerged in the communicative exchanges between people. The use of this term 'genetic’ echoes the sense it was given by both Jean Piaget and Lucien Goldmann.
In all of these instances, particular structures can only be understood as the transformations of earlier structures (Cf. the essay on themata in chapter 4). In Moscovici’s social psychology, it is through communicative exchanges that social representations are structured and transformed. It is this dialectical relationship between communication and representation which is at the core of Moscovici’s 'social psychological imagination’, and is the reason for describing this perspective as a genetic social psychology (cf. Duveen and. Lloyd, 1990).
In all communicative exchanges there is an effort to grasp the world through particular ideas and to project those ideas so as to influence others, to establish a certain way of making sense so that things are seen in this way rather than that way. Whenever knowledge is expressed it is for some purpose; it is never disinterested.
When Prague is located to the east of Vienna a certain sense of the world and particular set of human interests is being projected. The pursuit of knowledge returns us to the hurly-burly of human life and human society; it is here that knowledge takes shape and form through communication, and at the same time contributes to the shaping and forming of communicative exchanges. Through communication we are able to affiliate with or distance ourselves from others.
This is the power of ideas, and Moscovici’s theory of social representations has sought both to recognize a specific social phenomenon and to provide the means for making it intelligible as a social psychological process.
NB. Where two dates are given for a text, the first indicates the date of the original publication, and the second that of a later edition or an English translation.
Ableson, R. P., Aronson, E., McGuire, W. J., Newcomb, T. M., Rosenburg, M. J. and Tannenbaum, P. H. (eds) (1968). Theories of Cognitive Consisteney. Chicago: Rand MeNally.
Abric, J.-C. (1976). Jeux, conflits et représentations sociales. Thèse de Doctorat ès Lettres. Université de Provence, Aix-en-Provence. – (1984). A theoretical and experimental approach to the study of social representations in a situation of interaetion. In R. Farr and S. Moscovici (eds), Social Representations. Cambridge: Cambridge University Press,, 169-83. – (1988). Coopération, compétition et représentations sociales. Cousset: Del Val.
Abric ' L-C., Faucheux, C., Moscovici, S. and Plon, M. (1967). Approche et évitement dans des jeux à motivation mixte. Psychologie Franpise, 12, 277-86.
Ackennann, W. and Zygouris, R. (1974). Représentation et assimilation de la connaissance scientifique. Bulletin de CERP, 22 (1-2).
Alexander, L. N., Zucker, L. G. and Brody, C. L. (1970). Experimental expectations and autokinetic experiences: consisteney theories and judgemental convergence. Sociometry, 28, 108-22.
Allport, F. (1924). Social Psychology. Boston: Houghton Mifflin.
Anderson, H. N. (1968). A simple model for information integration. In R. P. Ableson, E. Aronson, W. J. McGuire, T. M. Newconib, M. J. Rosenburg and P. H. Tannenbaum (cds), Theories of Cognitive Consistency. Chicago: Rand McNally.
Ansart, P. (1988). Le concept de représentation en sociologie. In L. Marbeau and F. Audigier (eds), Seconde rencontre internationale sur la didactique de l’histoire et de la géographie. Paris: I.N.R.P.
Arendt, H. (1970). Men in Dark Times. London: Jonathan Cape.
Aristotle (1994). Posterior Analytics, trans. Jonathan Bames. Oxford: Clarendon Press.
Asch, S. (1952). Social Psychology. Englewood Cliffs, NJ: Prentice-Hall.
Back, K. W. (1964). La domaine de la psychologie sociale. Bulletin du C.E.R.P., 13, 21-33.
Barbichon, G. and Moscovici, S. (1965). Diffusion des connaissances scientifiques. Social Science Information, 4, 7-22.
Bartlett, F. C. (1923). Psychology and Primitive Culture. Cambridge: Cambridge University Press. – (1932). Remembering – A Study in Experimental and Social Psychology. Cambridge: Cambridge University Press.
Baucr, M. and Gaskell, G. (1999). Towards a paradigm for research on social representations. Journal for the Theory of Social Behaviour, 29, 163-86.
Bem, D. J. (1965). An experimental analysis of self-persuasion. Journal of Experimental Social Psychology, 1, 199-218.
Berger, P. L. and Luckmann, T. (1966). The Social Construction of Reality. Garden City, NY: Doubleday.
Bergson, H. (11932]/1935). The Two Sources of Morality and Religion, trans. R. Ashley Audra and Cloudesley Brereton. London: Macrnillan.
Berlin, I. (1981). Against the Current. Oxford: Oxford University Press.
Bíllig, M. (1987). Arguing and Thinking: A RhetoricalApproach to Social Psychology. Cambridge: Cambridge University Press. (1988). Social representation, objectification and anchoring: a rhetorical analysis. Social Behaviour, 3, 1- 16. (1993). Studying the thinking society: social representations, rhetoric and attitudes. In G. Breakwell and D. Canter (cds), Empirical Approaches to Social Representations. Oxford: Oxford University Press, 39- 62.
Blum, L. (1982). Souvenirs sur l’affaire. Paris: Gallimard.
Bower, T. (1977). The Perceptual World of the Child. London: Fontana.
Brehm, J. W. (1966). A Theory of Psychological Reactance. London: Academic Press.
Broch, H. (1979). Massenwahntheorie. Frankfurt: Suhrkainp.
Brown, R. (1965). Social Psychology. New York: Free Press.
Butor, M. (1960). Répertoire, vol. I. Paris: Editions de Minuit.
Carrier Duncan, J. (1985). Linking of thematic roles in derivational word formation. Linguistic Inquiry, 16, 1-34.
Carroll, J. S. and Paine, J. W. (1976). Cognition and Social Behaviour. Hillsdale, NJ: Lawrence Erlbaum.
Cassirer, E. (1946). The Myth of the State. New York: Doubleday.
Chapanis, N. P. and Chapanis, A. (1962). Cognitive dissonance: five years later. Psychological Bulletin, 61, 1-22.
Charle, C. (1977). Champ littéraire et champ du pouvoir: Les écrivains et l’affaire Dreyfus. Annales, 32, 240-64.
Chombart de Lauwe, M. J. (197 1). Un Monde autre: l’enfance, de ses représentations à son mythe. Paris: Payot.
Chomsky, N. (1975). Reflections on language. New York: Pantheon. – (1982). Some Concepts and Consequences of the Theory of Govermnent and Binding. Cambricige, MA: MIT Press. – (1993). Language and Thought. Wakeficid: Moyer Bell.
Coch, L. and French, J. R. P. (195 3). Overcoming resistance to change. In D. Cartwright and A. Zander (eds), Group Dynamics: Research and Theory. Evanston, IL: Tow Peterson, 336-50.
Cocking, J. M. (1982). Proust. Cambridge: Cambridge University Press.
Codol, J. P. (1974). On the system of representations in a group situation. European Journal of Social Psychology, 4, 343-65.
Codol, J. P. and Flament, C. (1971). Représentation de structures sociales simples dans lesquelles le sujet est impliqué. Cahiers de Psychologie, 14, 203-18.
Collins, B. E. and Guetzkow, H. (1964). A Social Psychology of Group Processes for Decision Making. New York: John Wiley.
Corbin, A. (1977). Le péril vénérien au début du siècle: prophylaxie sanitaire et prophylaxie morale. Recherches, 23, 63-71.
Comford, F. M. (1912). From Religion to Philosophy. London: E. Amold.
Culicover, P. (1988). Autonomy, predication and thematic relations. Syntax and Semantics, 21, 37-60.
Danziger, K. (1979). The positivist repudiation of Wundt. Journal of the History of the Behavioural Sciences, 15, 205-30. – (1990). Constructing the Subject. Cambridge: Cambridge University Press. – (1997). The varietics of social construction. Theory and Psychology, 7, 399416.
Davy, G. (193 1). Sociologues dhier et d’aujourclhui. Paris: Alcan.
Deaux, K. and Philogene, G. (eds) (2000). Social Representations: Introductions and Explorations. Oxford: Blackwell.
De Rosa, A. M. (1987). The social representations of mental illness in children and adults. In W. Doise and S. Moscovici (eds), Current Issues in Euroean Social Psychology, vol. 2. Cambridge: Cambridge University Press, 47-138.
Delacampagne, C. (1983). L’Invention du racisme. Paris: Fayard.
Deleuze, G. (1970). Proust et les signes. Paris: Presses Universitaires de France.
Delhorbe, C. (1932). L’affaire Dreyfus et les écrivains francais. Paris: Attinger.
Deutsch, Morton (1969). Organizational and conceptual barriers to social change. Journal of Social Issues, 25, 5-18.
Deutsch, Morton and Krauss, R. M. (1965). Theories in Social Psychology. New York: Basic Books.
Deutscher, I. (1984). Choosing ancestors: some consequences of the selection from intellectual traditions. In R. Farr and S. Moscovici (cds), Social Representations. Cambridge: Cambridge University Press, 71-100.
Doise, W. (1969). Intergroup relations and polarization of individual and collective judgements. Journal of Personality and Social Psychology, 12, 136-43. – (1992). L’ancrage dans les études sur les représentations sociales. Bulletin de Psychologie, 45, 189-95. – (1993). Debating social representations. In G. Breakwell and D. Canter (eds), Empirical Approaches to Social Representations. Oxford: Oxford University Press, 157-70.
Doise, W. and Palmonari, A. (eds) (1990). Létude des représentations sociales. Paris: Delachaux et Niestle.
Duby, G. (1961). Histoire des mentalités. In C. Sameran (ed.), L’histoire et ses méthodes. Paris: Gallimard.
Durkheim, E. (1895/1982). The Rules of Sociological Method, trans. W. D. Halls. London: Maemillan. – (1898/1974). Individual and collective representations. In E. Durkheim, Sociology and Philosophy. New York: Free Press, 1-34.
Durkheim, E. C 1912/1995). The Elementary Forms of the Religious Life, trans. Karen Fields. New York: Free Press.
(l955/l983). Pragmatism and Sociology, trans. J. C. Whitehouse. Cambridge: Cambridge University Press.
(1975). Textes, vol. 2. Paris: Editions de Minuit.
Duveen, G. (1598). The psychosocial production of ideas: social representations and psychologic- Culture and Psychology, 4, 455-72.
Duveen, G. and Lloyd, B. (eds) (1990). Social Representations and Development Of Knowledge. Cambridge: Cambridge University Press.
Eiser, J. R. andd Stroebe, W. (1972). Categorization and Social Judgement. London: Academic Press.
Ellison, R. (19065). Invisible Man. Harmondsworth: Penguin.
Emler, N. and Dickinson, J. (1985). Children’s representations of economic inequalities. British Journal of Developmental Psychology, 3, 191-8.
Evans-Pritchaírd, E. E. (1937). Witchcraft, Oracles and Magic among the Azande. Oxford: Oxford University Press.
(1964).,Social Anthropology and Other Essays. New York: Free Press.
(1965). -Theories of Primitive Religion. Oxford: Oxford Univesity Press.
Farr, R. M. (1 977). Heider, Harré and Herzlich on health and illness: some observations on the- structure of 'représentations collectives’. European Journal of Social Psychology, 7, 491-504.
(1978). On the varieties of social psychology: an essay on the relations between psychology and other social sciences. Social Science Information, 17, 503-25.
(19 8 1). On the nature of Imman nature and the science of behaviour. In P. Heelas and A. Lo&-ce (eds),
Indigenous Psychologies: The Anthropology of the Self London: Academic Press.
(1984). Social representations: their role in the design and execution of laboratory experiments. In R. M. Farr and S. Moscovici (eds), Social Representations. Cambridge:~ Cambridge University Press, 125-47. – (1993). Gommon sense, science and social representations. Public Understanding of Science, 2, 189-204.
(1996). -The Roots of Modern Social Psychology. Oxford: Blackwell.
(1998). From collective to social representations: Aller et Retour. Culture and Psychology-, 4, 275-96.
Farr, R. M. ar—xd Moscovici, S. (eds) (1984). Social Representations. Cambridge: Cambridge University Press.
Faucheux, C- (1970). Cross-cultural Research in Social Psychology. Unpublished manuscript, Social Science Research Council.
Faucheux, C. and Moscovici, S. (1960). Etudes sur la créativité des groupes: táche, structure des communications et réussite. Bulletin du C.E.R.P., 9, 11-22.
(1968). Self-esteem and exploitative behaviour in a game against chance and nature. Journal of Personality and Social Psychology, 8, 83-8.