LSE Research Online

Book Section

Sandra Jovchelovitch

Social representations, public life and social

construction

Originally published as:

Deaux, Kay; Philogene, Gina; Eds., Representations of the Social.
London, England: Blackwell Publications, 2001. © 2001 Blackwell
Publishing.

You may cite this version as:

Jovchelovitch, Sandra (2001). Social representations, public life and
social construction [online]. London: LSE Research Online.
Available at: http://eprints.lse.ac.uk/2649
Available in LSE Research Online: August 2007

LSE has developed LSE Research Online so that users may access research output of
the School. Copyright © and Moral Rights for the papers on this site are retained by the
individual authors and/or other copyright owners. Users may download and/or print
one copy of any article(s) in LSE Research Online to facilitate their private study or for
non-commercial research. You may not engage in further distribution of the material or
use it for any profit-making activities or any commercial gain. You may freely distribute
the URL (http://eprints.lse.ac.uk) of the LSE Research Online website.

This document is the author’s final manuscript version of the book section,
incorporating any revisions agreed during the review process. Some differences
between this version and the publisher’s version may remain. You are advised to
consult the publisher’s version if you wish to cite from it.

http://eprints.lse.ac.uk
Contact LSE Research Online at: [email protected]
Sandra Jovchelovitch
Institute of Social Psychology, LSE

Social Representations, Public Life, and Social

Construction1
Sandra Jovchelovitch

Introduction

In this chapter I want to advance some ideas that are concentrated around one main proposition: social representations are a form of
symbolic knowledge intrinsic to public life. Public life, I argue, is the place in which they are generated, develop, meet other
representations, change and, if the social and historical conditions so determine, die. In order to develop this argument, I shall proceed in
two steps. First, I shall address the links between social representations, as a specific type of social knowledge, and the public sphere, as a
specific type of social space. There are various types of social knowledge and not all are social representations - think, for
instance, of science and collective representations. In the same manner social spaces vary in form and configuration and not all of them
can be described as a public sphere - think, for instance, of the family. By addressing what is specific to the public sphere I hope to show
what is specific to social representations and how they differ from Durkheim's notion of collective representations. This first step should
make clear the conception I wish to advance about the ontological status of representations, and provide the basis for the second part of
my discussion where I address the problem of construction, and more specifically, the limits of construction.
I argue that construction in social representations theory is directly related to the symbolic function of representations. It is through a
careful assessment of the symbolic register that we can best understand the constructivism of social representations. At the same time, and
paradoxically to some, I argue that precisely because symbolic knowledges such as social representations are social, cultural, and histori-
cal they cannot fully construct reality. It is, I argue, in the properties of a socially constructed knowledge, dependent on history and
culture, that we find the reasons which explain why symbolic knowledge cannot be taken as the measure of all
reality. In short, it is because knowledge itself is socially constructed that it does not fully construct reality. It constructs partial realities, it
constructs human reality, but these are not, and should not, be understood as the whole of reality. While advocating the constructed
character of all human experience, I argue that a strong constructivist perspective recognizes that reality goes far beyond what we can
make of it.

Traditional and Detraditionalized Public Spheres: From Collective to Social Representations

Under which conditions can a type of knowledge such as social representations emerge? This question directs us to two essential, and
interrelated, dimensions of the production of social knowledge: the problem of genesis and the problem of context of production. In more
than one way, this question can be answered through a consideration of Moscovici's original work on psychoanalysis (Moscovici, 1976a)
and his earlier accounts of why the Durkheimian concept of collective representations was transformed into the social psychological
concept of social representations (Moscovici, 1988, 1989).
For Moscovici, as for Durkheim before him, the genesis of social knowledge is to be found in the social context. The link between
knowledge and the social context of its production is a problem that permeates the theory of social representations from its inception.
Moscovici's oeuvre is, in many ways, an attempt to provide an answer to the most fundamental assumption established by the work
of Durkheim and his followers, even the most reticent ones: when societal conditions change so does social knowledge. This assertion is
crucial to the theoretical edifice of the theory of social representations and to all those traditions of thought in psychology, sociology and
anthropology which, deeply influenced by phenomenological traditions, sought to establish that knowledge - any form of knowledge, from
science to common sense - is bound to the social context of its production.
From this perspective, the usage of "social context" is more than an abstraction or an added variable in a research program. Indeed, the
link between social knowledge and social context demands an understanding of what gives form to a social context, what makes one social
context different from another and how these differences produce variety in social psychological phenomena. It poses the need to unravel
theoretically and empirically how the structural features of a social context are decisive in accounting for the genesis, development, and
transformation of any type of social knowledge. Thus once we accept that social knowledge is shaped by social context, we face two new
requirements. The first is to inquire in depth, and conceptualize, the features of a social context, and the second is to ask what happens to
knowledge when a social context undergoes change.
The difference between collective representations, as described by Durkheim, and social representations, as proposed by Moscovici,
can be explained precisely by a careful assessment of the social contexts which Durkheim and Moscovici studied and sought to
understand. From collective to social representations the change is clearly related to transformations in social context. Moscovici himself

1
This chapter was published in K. Deaux & G. Philogène (Eds), (2001), Representations of the Social, Blackwell: New York
Sandra Jovchelovitch
Institute of Social Psychology, LSE

provides an illuminating statement in this regard in his reply to Jahoda in the pages of the European Journal of Social Psychology
(Moscovici, 1988). There he makes clear why the concept moved from collective to social representations. He writes:
In our days; therefore, collective representations as it used to be defined no longer is a general category but a special kind of
representations among many with different characteristics. It seems an aberration, in any case, to consider representations as ho-
mogenous and shared as such by a whole society. What we wished to emphasize by giving up the word collective was this plurality
of representations and their diversity within a group. . . . In effect, what we had in mind were representations that were always in the
making, in the context of interrelations and actions that were themselves always in the making. (p. 219)

More recently, Moscovici has also stated:

I prefer to use only "social" because it seems to me that a cultural belief or a ritual in modem times is secularized and embedded or
constituted in the society, a society in which conversations and communications between individuals become historically more
important than in more traditional societies. (Moscovici & Markova, 1998, p. 400)

In both quotations Moscovici refers to the more fluid dynamic of modem societies, where worldviews and practices are contested and
negotiated, and the space for a homogenous, unquestioned, and single view of the world is very limited indeed. In opting for "social" and
dropping "collective" Moscovici is acutely guided by the problem of plurality and the renewed importance of communication in modem
societies. As he points out, it would be an "aberration" to think today of representations consensually shared by a whole society. Societies
have changed and very few, if any, remain traditional in the sense described by Durkheim.
What has changed then in the social contexts considered by Durkheim and Moscovici? It is my contention that the crucial difference
between the social contexts considered by Durkheim and Moscovici's refers to changes in the public sphere, and more specifically, to the
issue of tradition versus detraditionalization in public spaces. Indeed, to understand the societal features which are key to the production of
social representations and, at the same time, undermine collective representations we need to examine the process of detraditionalization
in the public sphere. This process can be better understood through the concept of the public sphere, as developed by Habermas (1989,
1992) in his seminal work on the topic.
Habermas has described in detail the concept and phenomenon of the public sphere (1989). His account remains to this date the most
comprehensive analysis of this new category of capitalist society. In it, Habermas extensively discusses the emergence, development, and
transformation of the bourgeois public sphere in Europe and defines it as a space where citizens meet and talk to each other in a fashion
that guarantees access to all. It is a social space where:

1. All members, at least ideally or in the 'Nord of the law, meet as peers and discuss and decide in conditions of equality (the first
article of any bourgeois constitution refers to the equality of all men before the law);
2. Where the arguments of authority are replaced by the authority of arguments;
3. Where access is open to all and visibility guides the procedures and;
4. Where rational dialogue and nothing else establishes the legitimacy and subsequent authority of a proposition.

According to Habermas this set of assumptions is ushered in by the achievements of an age that gradually freed itself from tradition and
unquestioned historical orderings and sought in rational debate and democratic dialogue the response for matters of common concern.
Historically, this can be located around the set of ideas and events, which stemming out of the European Enlightenment, came to define
what is called Modernity. Centralized sources of authority, expressed by the power of both Church and State, gradually lost ground to the
emergence of an informed public of citizens who together construct a public sphere. The public sphere and its principles are produced
by - and in turn, help to produce - a social space where argumentation and rational dialogue are the key entry points to deal with
difference in perspectives. In it, participants are recognized by the quality of what they have to say and not by the authority or wealth of
their position.
It is obvious that these assumptions have never been fully realized. Habermas has been extensively criticized for failing to realize that
the liberal model of the public sphere rested, in fact, in a number of important exclusions, of which women and workers were the most
flagrant (Landes, 1988; Eley, 1992). Other commentators on Habermas work have noted, however, that the principles under which the
public sphere sought to function remain normative ideals, which can help us to question and challenge the quality of public (Holub, 1991;
Calhoun, 1992).
Another stream of criticism has pointed out, correctly in my view, that other public spheres need to be recognized, even if they are not
guided by the principles described by Habermas (Fraser, 1990). While agreeing with the importance of the principles of the Habermasian
concept of the public sphere as normative ideals, I believe that it is important to recognize that there are public spheres that do not
conform to them. There are other "publics" that also need to be recognized. It would be incorrect, in my view, to call a public space
"public sphere" only if it is guided by the mostly unrealized principles described by Habermas 2 . Think, for instance, of a market place in
traditional societies or of ritualized ceremonies, such as carnivals, and other types of religious celebration. They are certainly different

2
It is important to note, however, that unrealized principles are not necessarily unrealisable principles. Normative principles have
precisely the function of guiding and constraining action and historical change.
Sandra Jovchelovitch
Institute of Social Psychology, LSE

from the liberal model dominant in the West. They are, nonetheless, public spaces in so far as they constitute a space that (a) is shared by
all members of the community and (b) is the stage for issues related to the common life of the community. The principles that organize the
sharing of that space and the location of people in it are just other than the ones adopted by the liberal model of the public sphere.
It is here, I suggest, that we need to introduce the problem of tradition and detraditionalization. By considering traditional and
detraditionalized public spheres, it is possible to do justice to the fundamental differences in the knowledge they produce while retaining
the nonetheless fundamental proposition that all social knowledge is produced in a public space. These differences become clear when
considering the set of assumptions underlying public spaces in traditional societies. The principles of detraditionalized public spheres
stand in clear opposition to the basic rules of so-called traditional societies where:

1. The authority of some few sacred people defines the legitimacy of worldviews and constrains the access of members;
2. Where secrecy guarantees the sacred and;
3. Where inequalities in status structure the display of views by some members and the silencing of others.

In this type of traditional society what kind of social knowledge is possible? As Durkeim (1996, first published 1898) and Durkheim &
Mauss (1963) showed, what emerges from this type of social context are collective representations 3 . They are a type of everyday
knowledge which all members share and which operates as a full binding force; they are produced through conditions of strong asymmetry
between participants; they are strongly bound to, and dependent upon, ritual, and the conditions for their change are minimal. Collective
representations are a type of knowledge resistant to experience, argumentation, and logical proof, which relies mainly, if not purely, on the
social bond and its subjective value. They have the force of a social fact in the Durkheimian sense and fulfill functions of social
integration and reproduction while guaranteeing a strong solidarity between those who share them.
To advance slightly the second part of this chapter, which will tackle the problem of construction, in this type of knowledge reality is its
representation. Here, the social knowledge constructed by a community in the form of collective representations constructs, in the sense of
fully defining for participants, all reality. It is the social world that shapes and circumscribes all that exists, and the community, with
its peculiar hierarchy, is the fundamental source of authority. In this type of society the subjective of the social (i.e., everything related to
the power of the social bond and the feelings of allegiance and obedience it entails) has primacy over the objective (what happens to be the
case) and there is very little scope for individual variation.
We still find instances of this primacy in the way some young people find they "need" to conform to what is considered to be the "right
thing to do" by their communities. Individual aspirations, which are shaped by today's multitude of options circulating in public spaces,
clash with the rules of more traditional communities, where members are expected to live by, and comply to, the rules established
traditionally in ritualized practices and collective representations. Here one can understand why Durkheim has been criticized for his
tendency to speak of society as a homogenous unity (Giddens, 1971; Moscovici, 1988). The claustrophobic nature of his overtotalizing
conception of the social is evident, for social conflict in Durkheim is always related to the oppositions between the individual and the col-
lective. The notion that society can also be conceptualized as a system where there is tension and contradiction between different groups,
or collectives, is quite absent from his sociology (Giddens, 1978).
Before I proceed any further it is important to note that the descriptions above correspond more or less to ideal types and that neither
traditional societies nor detraditionalized public spheres are fully immune to the influence of their opposite. Traditional public spheres,
however, are much more resistant to the introduction of novelty/difference and, indeed, the predominant form of everyday knowledge
(collective representations) that they produce is especially apt to fulfill the function of resisting novelty and the transformations it may
entail. Traditional societies call upon the power and emotional dimension of the social bond to reproduce the knowledge they believe to
be right and needed to perpetuate their way of life. In detraditionalized public spheres, on the contrary, both strong elements of
tradition and strong challenge to tradition exist side by side. These diverse tendencies meet, clash and are constantly negotiated in the
public sphere (Beck, Giddens, & Lash, 1994; Heelas, Lash, & Morris, 1996; Thompson, 1995).
This dynamics was particularly clear in my own research on the health beliefs of the Chinese community in England (Jovchelovitch &
Gervais, 1999; Gervais & Jovchelovitch, 1998). We found that the older generation of Chinese people holds on to the collective
representations which express the culture, the traditions and the identity of the Chinese people. They also expect their offspring to follow
on their values and representations and rely heavily on the power of family hierarchy and community bonds to see this through. The
youngsters, in their turn, are tom between the expectations of their traditional community and the reality of a detraditionalized, Western,
public sphere where they were born and grew up. Accordingly, they expressed a mixed representational field made of the clashes and
negotiations resulting from their experience as BBCs (British Born Chinese).
The experience of the Chinese, as that of so many other diasporic communities, sharply exemplifies the antinomies of detraditionalized
public spheres, where fluidity and multiplicity in knowledge constitute the dominant way of life. It is in this type of detraditionalized

3
To acknowledge that collective representations are not the only form of social knowledge which exists in detraditionalized societies,
would require of Durkheim to conceive of the social as a field of tensions, malleable to the impact of novelty and the transformations
made possible even by minorities. This conception is absent from Durkheim's sociology and present in Moscovici's social psychology.
This is all the more clear in the context of Moscovici's theory of active minorities, and in his proposition that the function of social
representations is to deal with novelty and familiarize us with the strange.
Sandra Jovchelovitch
Institute of Social Psychology, LSE

public sphere, I suggest, that we find the conditions under which social representations emerge. They are a form of social knowledge that
comes into being in a social arena characterized by the mobility, and even more important, the diversity of social groups, a high degree of
reflexivity propitiated by the multiple encounters of different traditions, the massive and widespread circulation of information through the
development of the mass media (and more recently, the World Wide Web) and last, but not least, the liberal principles of equal access to,
and full visibility in, the public sphere. If social knowledge in all its forms is bound to the social world in which it is produced then, per
force, social representations cannot be equated to collective representations. Social representations at times encompass, at times reject, and
most of the time derive from, collective representations, but they cannot be immediately equated to them. A new public sphere calls for,
and produces, a new type of social knowledge.

Social Representations and the Appropriation of the New

As stated above, contrary to collective representations, which tend to appear as the dominant form of social knowledge in traditional
public spheres, social representations are symbolic forms typical of more contemporary, detraditionalized public spheres. They are a new
type of social knowledge: a social knowledge particularly suitable to cope with the new, and ontologically and epistemologically bound to
the macrosocial developments of our time. These developments have lead to, among other phenomena, a clouding in the processes of
making sense of the world, to a complexification in the "nesting of representations" to use an image proposed by Rommetveit
(personal communication, 1997). This means that communities of people now must handle the diversity of realities that constitutes their
life horizons and find new symbolic strategies to make sense of them (Jovchelovitch, 1997, Gervais & Jovchelovitch, 1998). Some of
these realities are utterly new, some are grounded in well-established traditions, and some belong to the merging of realities that takes
place in highly mediated societies. In any case, sense is attained, sustained, and transformed in a manner that transcends the limitations
traditionally imposed by context.
Context today goes beyond place and depends on a number of spatial and temporal displacements that upset our traditional
dependence on locale. This, of course, does not erase the importance of locales in the production of sense; it can, in fact, provoke a
recrudescence in the importance conferred upon the local (as in the revival of local identities vis-à-vis the threat of globalization).
However, locale alone is no longer sufficient to explain the formation of knowledge and how human life is coping with the increasing
otherness embedded in the various knowledges that shape local experience and ways of life. This otherness derives from a new regime of
objects in the social arena; an arena that has enlarged and shrunk at once, and brings the interplay between near and remote, familiar and
unfamiliar, same and other to new degrees of radicalization.
It is by taking into account these transformations in the public sphere that we can fully appreciate the innovative character of
Moscovici's theory of social representations and what, in my view, constitutes its breaking ground as a social psychological theory. When
Moscovici formulated the initial postulates of the theory, the signs of a detraditionalized, late modern, public sphere were still in an
incipient form. However, his empirical study on the transformations psychoanalysis underwent as it moved from one social group to
another, provided the evidence to ground the production and transformation of social knowledge in the vivid dynamics of the social
milieu.
La psychanalyse, son image et son public (Moscovici, 1976a) can also be read as a study about the manner in which
psychoanalysis is appropriated and resignified as it moves from one social group to another and penetrates different lifeworlds, different
horizons, different identities, and different projects. In short, what happens with psychoanalysis when it leaves the enclosed walls of
clinical practice, the narrow circles of training and expert societies, and falls, as it were, in the real world of everyday life? What type of
process is at stake when a specific type of knowledge in this case scientific knowledge 4 - circulates through the social fabric far beyond
the source of its production? The answer was clear to Moscovici: this knowledge changes. It changes in the same way it changes the
people who have changed it in the first place. Thus, psychoanalysis was not only what its "owners" would make of it, but it became a
different phenomenon as it penetrated the lifeworlds of urban liberal professionals, catholics, and communists, the social groups
Moscovici studied.
This difference, or internal disjunction in the representation of a social object, can only occur because it takes place in a society where
worldviews are removed from one single source of authority and legitimization and confront each other in more or less equal terms in the
public sphere. Think, for instance, what would happen if the psychoanalytic institution could determine what every single social group
should think of it? As in a photocopying line, social knowledge would be a mere reproduction (although not even a copy is a perfect
reproduction). However, there is no one single institution in a detraditionalized public sphere capable of imposing its authority without
some form of contest. The process of institutionalizing specific representations in these societies is entangled in a complex network of
competing interests, representations, and powers. Diversity of worldviews engenders representational systems that express variety and
reminds us that the phenomenon of social representations is as much about shared and consensual symbolic codes as it is about
contradictory and unresolved ones.

4
It is interesting to note that psychoanalysis in France has always been treated as a scientific theory. This is not the case in Britain where
psychoanalysis has struggled to prove its epistemological credentials as a scientific theory. In this discrepancy we have yet another case of
how cultural traditions are deeply intertwined with the production and acceptability of what comes to constitute scientific knowledge.
Sandra Jovchelovitch
Institute of Social Psychology, LSE

The diversity and plurality of the public sphere produces a great deal of unfamiliarity, or strangeness. In fact, it is the strangeness
introduced by differences and contradictions that sets into motion the processes of anchoring and objectification, through which the
unfamiliar is tamed and made familiar. Anchoring is nothing but the attempt to settle a new, and therefore strange, meaning into the
established geography of symbols of a community. Objectification, in its turn, gives to novelty a concrete, almost "natural," face.
Anchoring and objectification are triggered to "settle accounts," as it were, with new meaning, to make it known, to reduce its threat or the
fears it generates, to make it a part of the "us" of a community of people (Gaskell, this volume; Philogène, 1999). Or, as Moscovici has put
it, to make the unfamiliar familiar (Moscovici, 1984b). In this sense it is crucial to acknowledge that social representations are forms of
knowledge dependent on the introduction of the "strangeness" of the other and its novelty/difference (Jovchelovitch, 1998c). Note that this
other can be any empirical other, from a different social group in the same community, to a "stranger" in Schutz's sense (Schutz, 1944),
who comes from far away and upholds different cultural values. The other, by its sheer condition of being other, upsets one's way of
thinking and conceiving of the world and disrupts the familiarity of the taken-for-granted.
It is in order to tame the strangeness of this unfamiliarity that a new social production of knowledge starts, leading to the formation of
social representations. Since they draw on the unfamiliarity produced by the other as a condition for their formation, we can say that social
representations are as much about constructing meaning as they are about constructing bridges that link the diversity of perspectives
constitutive of public life. In this sense, social representations are forms of knowledge structured to handle difference and have close links
to the quality of public spaces. Indeed, they are themselves indicators of this quality. Public spaces that allow difference and dissent to
coexist, be negotiated and reworked through dialogue and equality of access are, par excellence, the space in which social
representations are produced. Totalitarian societies5 , alternatively, militate against the production of everyday knowledges such as social
representations. Representations produced in the bubble of social life, where different worldviews are allowed to propose their project and
sustain the claims they make about reality and what the future should be, are potentially dangerous to whoever wants to sustain the su-
premacy of a single worldview, sure of itself and of the "truth" it proposes.
Thus what accounts for the actuality of Moscovici's study on representations of psychoanalysis, some 40 years after its publication, is
precisely the contemporary character of the very questions Moscovici posed then. These questions have not aged. What happens with
knowledge - any form of social knowledge - as it moves context and penetrates the lives of different social groups and is reworked by a
variety of modes of communication and interaction, remains one of the most pertinent and challenging questions opened to the social
sciences. It is a question as much related to the possibility of communication between different worldviews, different ways of knowing,
thinking, and living, as it is to the potentials embedded in this communication and the quality of our public spaces. In my view, it also
binds the theory of social representations to a theory of modernity, insofar as the distinctive phenomenon it theorizes - social
representations - are bound to a modern, detraditionalized, public sphere.
This argument could be taken even further if one considers Moscovici's notion or cognitive polyphasia, something I have discussed in
more detail elsewhere (Jovchelovitch, 1998b). Moscovici introduced the concept in his study about the reception or psychoanalysis in
France. He round strong evidence that different types or rationality were involved in the construction of psychoanalysis in France; these
rationalities were dependent on the context of their production and intended to respond to different aims. The striking finding, however,
was that contrary to well established interpretations of cognitive phenomena, the different rationalities did not correspond to different
groups, in different contexts; on the contrary, they were capable of coexisting side by side in the same context, social group, and mutatis
mutantis, the same individual. I do not have the space to expand on this notion here, but it suffices to say that social representations are
described by Moscovici as sociocognitive structures guided by cognitive polyphasia, capable of encompassing different logics, which
social actors draw on as a function of different contexts and practical demands. Further evidence has been gathered by a number of
empirical studies in the field of social representations (see Gervais & Jovchelovitch, 1998; Wagner, Duveen, Themel, & Verma, 1999).
The notion of cognitive polyphasia, which corresponds both to the polysemic nature of meaning and to the contradictory demands of a
detraditionalized public sphere, consolidates the critical stance of social representations theory towards the view of a rationality identical
to itself, a view dear to most theoretical developments in the history of social psychology (see Farr, 1996).
In the foregoing I have argued that social representations are forms of social knowledge bound to the public sphere. Moreover, I have
shown that the type of public sphere in which social representations emerge stands in opposition to traditional public spheres such as those
studied by Durkheim. In both cases, there is the underlying assumption that social knowledge is organically linked to the social context of
its production. The issue posed to the theory and to the development of its empirical program is to qualify this proposition. It can do so by
investigating more precisely the ways in which transformation in the structure and form of public life entails transformation in the
structure and form of the social knowledge which is there produced.
I hope I have shown that, in the study in which the theory of social representations originated, Moscovici was already grappling with a
form of public sphere fundamentally different from the traditional societies Durkheim sought to understand. Durkheim's collective
representations possess the status of a social fact in its "hard" sense: even though they are produced by social actors, they acquire a char-
acter that is external to human action and condition all individuals of the societies in which they are present to think according to the
categories they propose. Contrary to traditional societies that keep their knowledges firmly controlled by sacred individuals, sacred rituals,

5
It is important to make a clear distinction between authority and totalitarianism. Totalitarian regimes try to prevent the production of
everyday knowledges by force, whereas traditional societies operate by the authority and power of the social bond itself The latter derives
its power from the fact that it is recognized; the former needs force and, frequently, state violence to exert its effects.
Sandra Jovchelovitch
Institute of Social Psychology, LSE

and sacred objects, modern societies are characterized by the reflexivity associated with social representations. Nothing is "naturally
accepted," everything is under question, or as Marx would say, "all that is solid melts into the air. "
Thus my analysis, starting from the question I put at the beginning of this section, has in a sense come full circle. Under which
conditions can a type of knowledge such as social representations emerge? The tentative answer I provide is that they can only emerge
under societal conditions which have escaped from unquestionable historical orders, where the belief in a pregiven order of things has de-
clined or, at least, lost some of its authority, and where voices of authority become decentered, resting either on a variety of voices or, as
in the case of most industrialized western societies, on self Social representations therefore need the bubble of conversation, conflict
between social groups and encounters between different perspectives to come into being. They need the process of decentration of
worldviews (Habermas, 1991), a process which challenges the status of traditions and sets into motion new forms of social knowledge.
These social representations, while struggling to come to terms with the provisional unfamiliarity of the new, also provide the means
through which communication and action can be established.

Social Representations, Social Construction, and Reality

Let me now turn to the problem of construction. There has been much debate about what is "construction," and what construction entails
in social psychology and in the social sciences in general. It is not my intention here to trace the various shades of this debate, something
which has been effectively done by Danzinger (1997) in a recent review. My intention is rather to relate the problem to the issues I
addressed in the first part of this chapter and to state what I believe to be the constructivist character of the theory of social representations.
In the avant-propos to the second edition of La Psychanalyse, son image et son public (1976a), Moscovici stressed that his original
ambition with that work was to redefine the concepts and problems of social psychology, taking the phenomenon of social representations
as a starting point and reaffirming its symbolic function and power to construct reality. It is, I argue, the careful analysis of the symbolic
function that clarifies the constructivist nature of the theory of social representations. The analysis of the symbolic function also elucidates
the public, cultural and historical dimensions of knowledge.
Traditional empiricist notions consider the central function of cognitive systems to submit to an a priori reality, a reality that both
precedes and ignores the labors of cognition. By copying the features of this pregiven reality as closely as possible, mental representations
may produce a reproduction that differs as little as possible from it. This type of conceptualization defines representation as an individual
process of mentally copying the outside world. The symbolic function of representational activity, that is, the fact that a representation is a
medium for investing the outside world with meaning, is mainly ignored. Ignoring the symbolic function has thrown subjects of research
into a state of perpetual error (usually called "bias") where, from the perspective of the researcher, subjects do not remember correctly, do
not perceive correctly or correctly know what they are talking about. It is a conceptualization so deeply rooted in psychology that it
became mixed with the phenomenon of representation and contributed to produce the neobehaviorism expressed by some extreme forms
of discursive psychology (for an example, see Ibanez, 1994). Ironically, in being reluctant to theorize representation or cognition, critical
psychologists allow traditional cognitivism to have the ultimate say about these phenomena (Jovchelovitch, 1996).
Now it is precisely the symbolic dimension of representations that inspires researchers in the field of social representations and
demarcates its difference from traditional cognitivism. The analytic focus is on what people mean as they engage in the task of making
sense of the world in which they live and communicate with others about it. The analysis of meaning is not primarily concerned with
whether a meaning is "right" or "wrong" in relation to any given reality. On the contrary, the analysis of meaning is concerned with the
expressiveness of utterances, images, rituals, or any other representational act: it is concerned with the symbolic function of the
representational act and its power to construct what is real to a group of people. It is through the symbolic function that we can make sense
of the fact that the same social object acquires different shades of meaning to different people in different contexts and times. It is also the
symbolic function that allows "errors" and "biases" to be considered as expressive symbols worth of analysis and interpretation. Moreover,
the insertion of temporal and spatial considerations, which correspond to the vital problems of historical and contextualized
understandings, is directly derived from the acknowledgment of the symbolic register. This has firmly established the theory of social
representations in what Geertz (1993b) calls the interpretative turn in the social sciences. Inspired by Moscovici in La
Psychanalyse, son image et son public (1976a, first published 1961) and Jodelet in Madness and Social Representations (1991),
researchers in the field continue to pursue the symbolic dimension.
Following the constructivism opened up by the work of developmental psychologists such as Piaget (1954) and Vygotsky (1978,
1994), the theory understands that reality, and knowledge of reality in whichever form it takes, are not immediately given to us; that
between the world standing out there, in itself, and any possible knowledge we can have of it, there must be a process of bridging, of
mediation, of active symbolic construction. Things in themselves mean nothing; people must represent them and make them signify,
that is, they must give them a meaning using a symbol. Through the work of representations our social world becomes a multitude of
symbols which define what is real for us and allow the world of things to come into existence, as it were. It is the triangulation between
people (in plural because human reality is made of a multitude of perspectives; it is not an individual person who constitutes human life),
objects and the work of representations that produces the symbolic register in which we live. Humans need to construct in self and
cognitive development, in work, in institutions, in culture, the conditions which make them human, and define, to a large degree, the
worlds in which they live.
Social construction accounts for the complex and difficult, painful but liberating, process of producing culture and all that culture
Sandra Jovchelovitch
Institute of Social Psychology, LSE

entails, that is, a cultured human being, cultured societies, cultured knowledges, and cultural artifacts. Such is the power of symbols: to
construct realities, to institute them, to prescribe what should be and what should not be accepted, and to produce extremely concrete and
real consequences in people's lives. This, however, does not mean that the material world or our biological make up is not important. We
are both cultural and natural creatures, and our cultural being is deeply intertwined with the sorts of bodies we have and the kind of world
to which they belong. Culture transforms nature, but does not do away with it and, in fact, it could not itself exist without it. Processes of
cultural production, or processes of social construction, are dependent on an outside, objective and natural world, from which they draw
the materials without which nothing can be constructed. To understand this substantive materiality of social life - expressed both in
natural elements and in objects, machines, rooms, walls, other people, etc. - permits us to understand how the material world offers
resistance to one, and the degree to which it frames and encloses the possibilities of human action and construction.
In this sense, reality is, in itself, much larger than what we socially construct. It not only contains dimensions which continually escape
the knowledge we have of it, but it also contains dimensions which, even though constructed by humans, confront them as objective
matter. Symbolic knowledge, as perhaps the most crucial process of cultural production, plays a pivotal role in what we can know of
reality and in shaping the meaning we invest in it. But knowledge, as a representational system, as a symbolic system, cannot be equated
with reality because this is analogous with saying that reality is our representation. Symbols, however, represent reality and reshape it;
they produce meaning out of it. They are not it. Between knowledge and its object, or between representations of reality, and reality itself,
there is a difference, a disjunction, which can be clearly identified and explained if one takes seriously the historical, cultural, public, and
symbolic dimensions of knowledge. I believe that a clear appreciation of these dimensions can contribute both to dissipating some
problematic views in the field of social representations itself (see Wagner, 1996, for an example of what I believe is a problematic view;
see also Wagner, 1998 for an interesting elaboration of his previous position) and to replying to the excesses of some streams of post-
modern psychology.

The historical dimension of knowledge

Knowledge cannot be immediately equated with reality because knowledge is historical. To understand the historicity of knowledge we
just need to observe how it changes over time. This can be observed both in biographies and in cultural history. It can also be observed
through historical empirical evidence, both in relation to the history of sciences (Canguilhem, 1991; Lakatos & Musgrave, 1970) and to
the history of mentalities (Aries, 1960; Blumenfeld-Kosinski, 1990; Camporesi, 1995, Young, 1995). Should we have thought that
knowledge is reality, the sciences would not have evolved and forms of lay thinking about, and knowing of, reality would not have
changed. Take, for instance, representations of black people, of women, and of colonial peoples. These three groups of people have been
historically represented as inferior and these representations have, to a large degree, defined their identity and conditions of being. These
representations are real enough, but to say that black people, women, and colonial peoples are really inferior means to
acknowledge that they can only be what the eye of a dominant beholder makes of them. This is certainly not the case, vide the social
movements and struggles of so many excluded peoples. Knowledge therefore suffers the transformations of history and remains bound
to temporal displacements that give to it a provisional character.

The cultural dimension of knowledge

Knowledge cannot be equated with reality because knowledge is cultural. To understand the cultural dimension of knowledge we just need
to observe how it compares across contexts. Human societies are not homogenous and they produce different cultural assumptions that
frame reality in different ways. This means to say that what is said here is different from what is said there, that there is a number of
"taken-for-granteds" which vary across place, and that people in the plural, and not just one person, inhabit the world (Arendt, 1958).
Should we advocate the coincidence between knowledge and reality, reality would be either pure fragmentation or pure isolation, since
every culture has the right to advocate its reality as the "real one." There is a wealth of evidence coming from the social sciences that
shows how different cultures produce different knowledges and operate under different assumptions concerning reality (Bhabha, 1994;
Geertz, 1993a; Levy-Bhrul, 1910/ 1985; Said, 1993, 1995; Todorov, 1992). Knowledge therefore is open to cultural variation and remains
bound to the spatial displacements that give to it a limited characters 6

The public dimension of knowledge

Knowledge cannot be equated with reality because knowledge is public. To understand the public dimension of knowledge we just need to
observe how it changes in relation to the different interests and projects associated with different social groups. Society is a public field of
tensions and differences, where social groups struggle to propose their views and the projects they hold for the future. Plurality in the

6
Please note that I am not referring here to scientific knowledge, which is a type of social knowledge which intends to transcend the
limitations imposed by time and space. Sometimes it manages to do so, sometimes it does not, whereas everyday knowledges such as
social representations are always bound to time and space. The theory of gravity applies independently of time and context (on Earth),
whereas social representations of women, biotechnology, or health are always bound to time and context.
Sandra Jovchelovitch
Institute of Social Psychology, LSE

social field coupled with power differentials in access to resources shape the knowledge formation of both individual people and social
groups. The public sphere is an arena where knowledges compete, clash, and are renegotiated, making clear the distinction between
knowledge and its object. In societies with decentered conditions of legitimation, that is to say, where not only one, but diverse centers of
power claim authority and legitimacy to construct knowledge about reality, it becomes clear that knowledge cannot be immediately
equated to the object it intends to capture. In fact, it is in the guiding principles of detraditionalized public spheres, where plurality can be
fully expressed, that we see more clearly the clear distinction between representations and reality.

The acknowledgment of the historical, cultural, and public dimensions of knowledge entails, therefore, the understanding that knowledge
is a limited and incomplete enterprise, and that for us, the real will always be an unfinished business. It is precisely this precarious nature
of knowledge that propels the knower towards the as yet unknown dimensions of reality and therefore sustains the very possibility of
development in knowledge and expansion of what is real to us. As much as knowledge constructs what is real for us, it does not apprehend
the whole of reality, a domain ever open to our tentative efforts to capture it. In many respects, this position coincides, albeit in a different
wording, with Wagner's (1998) discussion of the differences between a domesticated world and "somethings," i.e., things that do not
deserve to be called objects. The "something," although phenomenally there, is not yet named, or symbolically "domesticated" by the
representational labor of a group of people. It "has to do with what Searle (1995) calls 'brute facts,' and it is useful as a reminder that there
are many things beyond socially constructed worlds" (Wagner, 1998: 306). This is something Durkheim has also shown convincingly in
his attack on pragmatism (Durkheim, 1983, first published 1955).
To recognize that the unknown is constitutive of our efforts to know goes hand in hand with the human struggle for accuracy in
cognition, since it is not true that "everything goes." There is an almost obvious problem in stating that "reality is what I, or we, know," in
the same way that there is an obvious problem in stating that if reality is what each of us can make of it, every representation of reality is
as good as any other. As much as social representations theory struggles to rehabilitate lay knowledge and bring into focus its power to
signify, it also needs to keep an acute awareness of its limitations, because no knowledge is completely immune to distortion, to
misunderstanding, to barbarities. Knowledges can reveal and mislead, emancipate and oppress, and we need to distinguish between a
symbolically structured order and the order referred to, as the only condition of critique.
Thus we need to recognize that not every story can be accepted as history and not every representation is accurate towards its object.
Conceptions that defend the "everything goes" deny the arduous and painful struggle of so many human groups to reach a precise
knowledge and to overcome the various distortions which entangle it. To say that truth and falsity are socially constructed does not mean
to say that we can just throwaway the idea of truth, and even less, the idea of falsity. For centuries, to go back to the example I used above,
humans have constructed representations of others which are a clear expression of the necessity of truth. Although the consequences of
these representations are extremely real to all those involved in the set of practices they originate, it is only the notion of truth which can
put them where they belong, which is the realm of falsity.

Concluding Remarks

In this chapter I have expanded the discussion which I initiated some time ago about the link between social representations and the public
sphere (Jovchelovitch, 1995, 1998a, 2000). Drawing on one of the most fundamental premises of the theory of social representations, i.e.,
that social knowledge is bound to the social context of its production, I have discussed the discontinuities between Durkheim's concept of
collective representations and Moscovici's concept of social representations. I have shown that social representations are forms of
knowledge bound to detraditionalized public spheres, where novelty plays a crucial role and worldviews compete in more or less equal
terms in the social arena.
Having established the differences between social and collective representations in relation to a transformed public sphere, I addressed
the issue of social construction. I argued that, although we construct knowledge, we do it in relation to a reality which permanently
escapes from our making. Knowledge constructs what is real for us, which is not necessarily real for other people and other historical
periods. Thus knowledge is limited, incomplete, and unfinished; it can never grasp the wholeness of reality. There is always an unknown
dimension in reality to be explored and known. I explained this apparent contradiction between knowledge of reality and reality itself
through a discussion of the historical, cultural, and public dimensions of knowledge. Finally, I argued that the distinction between
representations and reality needs to be understood and preserved as fundamental to sustain the very idea of critique.
These issues are not final and certainly there is much more to consider in the study of social representations. They are important,
however, and I believe they grow in importance if we take into account the globalizing tendencies of late modernity. Meanings,
knowledges, and ways of life travel today freed from the spatial and temporal restrictions which traditionally have characterized our
relations to times and places. This freedom, which we owe especially to the development of the mass media of communication, produces
a new reflexivity which allows social representations to clash, to compete, to intermingle, and to appropriate new sense. But it also brings
to the center of our discussion what happens in those communities which lack the material, cognitive, and symbolic resources to propose
and to defend their representations and way of life.
It is not a matter of denying the increasing processes of communication and circulation of inforrnation; here there is no return and the
world the twenty-first century will see will have expanded and shrunk in a dimension without precedents. But this process puts into
sharp focus concern and solidarity with the diversity of the peoples on our planet and the hard realization of the invisibility of so many
Sandra Jovchelovitch
Institute of Social Psychology, LSE

peripheral communities. Different and distant lives are not unconnected lives, and the history of all cultures is also the history of cultural
borrowings. Sometimes the emphasis on our differences obscures the fact that what we have in common - and "we" here means
anybody, wherever they come from - is precisely the multiple contexts that make each human experience unique in its own right.
Defending this diversity needs to be today, just as it has been for almost 300 years, central to the project of an enlightened universalism.

Notes
Research for this paper was conducted while I was a Directeur d'Études Associé at the École des Hautes Études en Sciences Sociales in
Paris. I would like to thank M. Maurice Aymard, at the Maisons des Sciences de I'Homme, for providing such a hospitable and congenial
environment in which to work. This paper has benefited greatly from discussions I have had over the years with Martin Bauer, Gerard
Duveen, George Gaskell, Robert Farr, Pedrinho Guareschi, and Wolfgang Wagner. I am also grateful to Kay Deaux and Gina Philogène
for their very helpful comments in a previous version of this paper. I am indebted to Professor Serge Moscovici, who has patiently listened
to my ideas and concerns and generously gave me his time and knowledge while I was conducting research on these issues.