The process of collective identity
Defining collective identity
The concept of collective identity was introduced in my previous contributions
to the analysis of contemporary social movements (Melucci 1988,
1989, 1994), and has already stimulated a promising discussion
(Bartholomew and Mayer 1992; Gamson 1992a; Mueller 1994; Proietto
1995). In recent sociological debates we are witnessing a renewed interest
in cultural analysis which corresponds to a shift towards new questions
about how people make sense of their world, how they relate to texts, practices,
and artifacts rendering these cultural products meaningful to them
(see Swidler 1986, 1995; Wuthnow et al 1984; Wuthnow 1987; Wuthnow
and Witten 1988; Clifford 1988; Alexander 1990; Alexander and Seidman
1990). The contributions of social psychology in terms of scripts (Schank
and Abelson 1977; Abelson 1981), social representations (Farr and
Moscovici 1984; Moscovici 1988), the rhetorical construction of arguments
and thoughts (Billig 1991, 1992, 1995), as well as the reflections on
the discoursive construction of identity (Gergen 1982, 1985, 1989, 1991;
Shotter and Gergen 1989; Shotter 1993b; Harre and Gillett 1994) are also
part of this shift towards a better understanding of the human capacity to
construct meaning and to make sense of action.
The present interest in culture and meaning is paralleled by a growing discussion
on the topic of identity, both at the individual and collective level,
which crosses different disciplinary fields in social sciences (for an introduction
see Hirsch 1982; Weigert et al 1986 Berkowitz 1988; Abrams and Hogg
1990; Burkitt 1991; Breakwell 1992; Barglow 1994). The interest is focused
on critical issues such as the continuity-discontinuity of identification processes
and the multiplication of the facets of identity in contemporary
society (Berger et al 1973; Parfit 1984; Elster 1985; Taylor 1989; Gergen
1991; Strathern 1991; White 1992; Burke 1992; Melucci 1996).
68
The process of collective identity 69
These new questions raised by the recent reflection on culture, identity
and meaning are paralleled by the increasing evidence of the weaknesses of
traditional sociological theories when confronted with contemporary
social movements. So far, the study of social movements has been divided
among those who continue to work under the premises of the dualistic
legacy discussed in chapter 1. As a result, we are still struggling to bridge
the gap between behaviour and meaning, between 'objective' conditions
and 'subjective' motives and orientations, between 'structure' and 'agency'.
Explanations based on 'structural determinants' on the one hand and
'values and beliefs' on the other can never answer the questions of how
social actors come to form a collectivity and recognize themselves as being
part of it; how they maintain themselves over time; how acting together
makes sense for the participants in a social movement; or how the meaning
of collective action derives from structural preconditions or from the sum
of the individual motives.
The development of a new interest in culture, and the related attention
to hermeneutics, linguistics (Barthes 1970, 1975), and to the many
methodological warnings issuing from ethnomethodology and cognitive
sociology (Garfinkel 1967; Cicourel 1974, 1982) have also made more
evident the low level of epistemological awareness and self-reflexivity typifying
traditional research on collective phenomena (see also chapter 20).
With few exceptions (such as Eyerman and Jamison 1991; Diani and
Eyerman 1992; Johnston 1995), research on social movements has up to the
present been informed by a widespread 'realistic' attitude toward the
object, as if collective actors existed in themselves as unified ontological
essences, readily offered for the comprehension of the researcher through
reference to some underlying structural condition or upon sorting the
motives behind the various behaviours. The position of the observer is of
course that of an external eye, as objective as possible, and very little attention
is paid to questions such as how the relationship of the researcher to
her/his field contributes to its construction, even if we can see signs of a
turning point on these matters and of an increasing epistemological awareness
(on social movements, see Johnston and Klandermans 1995;
Darnowsky, Epstein and Flacks 1995).
A thorough rethinking of the concept of collective identity is necessary
to productively confront the dualism between structure and meaning. The
concept, as we will see, cannot be separated from the production of meaning
in collective action and from some methodological consequences in considering
empirical forms of collective action (see chapter 20). This strategic role
of the concept in dealing with the questions that are arriving to the forefront
of contemporary sociological debates probably explains the parallel inter-
70 Theory of collective action
est in both cultural analysis and collective identity. By asking the question
of how individuals and groups make sense of their action and how we may
understand this process, we are obliged to shift from a monolithic and metaphysical
idea of collective actors towards the processes through which a collective
becomes a collective. A processual approach to collective identity
helps to affect such a theoretical and methodological shift. But the concept
risks being incorporated in recent social movements studies (see Taylor and
Whittier 1992; Friedman and McAdam 1992; Hunt et al. 1994) in a reified
fashion, as a new passepartout that simply substitutes the old search for a
core 'essence' of a movement: without, that is, taking into account its theoretical
and methodological implications. In the present chapter, I want to
outline a processual approach to collective identity, relying on the constructivist
view of collective action developed in this book. Such an
approach, moreover, carries important epistemological consequences for
the way the observer/observed relation is construed in social research, and
it affects the research practices themselves, as we shall see in chapter 20.
The question of how a collective actor is formed at this point assumes a
decisive theoretical importance: what was formerly considered a datum (the
existence of the movement) is precisely that which needs to be explained.
Analysis must address itself to the plurality of aspects present in the collective
action and explain how they are combined and sustained through
time. It must tell us, therefore, what type of 'construct' we are faced with in
the observed action and how the actor himself is 'constructed.'
Action andfield:a definition
I call collective identity the process of 'constructing' an action system (see
chapter 1). Collective identity is an interactive and shared definition produced
by a number of individuals (or groups at a more complex level) concerning
the orientations of their action and the field of opportunities and
constraints in which such action is to take place. By 'interactive and shared'
I mean that these elements are constructed and negotiated through a recurrent
process of activation of the relations that bind actors together.
(i) Collective identity as a process involves cognitive definitions concerning
the ends, means, and the field of action. These different elements,
or axes, of collective action are defined within a language that is shared
by a portion or the whole of society, or within one that is specific to a
group; they are incorporated in a given set of rituals, practices, cultural
artifacts; they are framed in different ways but they always allow some
kind of calculation between means and ends, investments and rewards.
The process of collective identity 71
This cognitive level does not necessarily imply unified and coherent
frameworks (unlike cognitivists tend to think: see Neisser 1976;
Abelson 1981; Eiser 1980); rather, it is constructed through interaction
and comprises different and sometimes contradictory definitions (see
Billigef*/. 1988; Billig 1995).
(ii) Collective identity as a process refers thus to a network of active relationships
between actors who interact, communicate, influence each
other, negotiate, and make decisions. Forms of organization and
models of leadership, communicative channels and technologies of
communication are constitutive parts of this network of relationship.
(iii) Finally, a certain degree of emotional investment is required in the
definition of a collective identity, which enables individuals to feel
themselves part of a common unity. Collective identity is never entirely
negotiable because participation in collective action is endowed with
meaning which cannot be reduced to cost-benefit calculation and
always mobilizes emotions as well (Kemper 1978, 1981, 1990;
Hochschild 1979, 1983; Scheff 1990). Passions and feelings, love and
hate, faith and fear are all part of a body acting collectively, particularly
in those areas of social life that are less institutionalized, such as
the social movements. To understand this part of collective action as
'irrational', as opposed to the parts that are 'rational' (a euphemism for
'good'), is simply nonsensical. There is no cognition without feeling
and no meaning without emotion.
Let us try now to understand more closely this interactive and communicative
construction, which is both cognitively and emotionally framed
through active relationships.
Process and form
The term 'identity' is most commonly used to refer to the permanence over
time of a subject of action unaffected by environmental changes falling
below a certain threshold; it implies the notion of unity, which establishes
the limits of a subject and distinguishes it from all others, and a relation
between two actors which allows their (mutual) recognition. The notion of
identity always refers to these three features: namely, the continuity of a
subject over and beyond variations in time and its adaptations to the
environment; the delimitation of this subject with respect to others; the
ability to recognize and to be recognized.
The notion of a certain stability and permanence over time seems to contrast
with the dynamic idea of a process. At any given moment social actors
72 Theory of collective action
no doubt try to delimit and stabilize a definition of themselves; so do the
observers. But the concept of collective identity as defined above can precisely
help to understand that what appears as a given reality, something
more or less permanent, is always the result, at least to a certain extent, of
an active process which is not immediately visible.
Such a process involves continual investments and as it approaches the
more institutionalized levels of social action it may increasingly crystallize
into organizational forms, systems of rules, and leadership relationships.
The tendency and need to stabilize one's identity and to give it a permanent
form create a tension between the results of the process, which are
crystallized in more or less permanent structures, in more or less stable
definitions of identity, and the process itself which is concealed behind
those forms.
The concept of collective identity can be of help addressing the interactive
and sometimes contradictory processes lying behind what appears
as a stable and coherent definition of a given collective actor. We should,
however, take notice of the fact that the term 'identity' remains semantically
inseparable from the idea of permanence and may, perhaps for this
very reason, be ill suited for the processual analysis for which I am
arguing. Nevertheless, I have retained 'identity' as a constitutive part of
the concept of 'collective identity' for the simple reason that for the
present, no better linguistic solution seems available. Because, as I will
argue, such collective identity is as much an analytical tool as an object to
be studied, it represents by definition a temporary solution to a conceptual
problem, and should be replaced if and when other concepts prove
themselves more adequate. In the meantime, my work continues situated
within the limits of the available language, confident that the inevitable
shift towards new concepts will not amount to a mere matter of different
terminology but to an overall emergence of a new paradigm. The way out
of the legacy of modernity will be a difficult process, and our time will
notice its completion only at the end, when, as in any major scientific shift,
we find ourselves already in a new conceptual universe. Meanwhile, for the
sake of communication, we cannot but continue using old words to
address new problems.
One way to overcome the apparent contradiction between the static and
the dynamic dimensions implied by collective identity is to think of it in
terms of action. Collective identity enables social actors to act as unified
and delimited subjects and to retain control over their own action; conversely,
however, they can act as collective bodies because they have completed,
to some extent, the constructive process of collective identity. In
terms of the observed action, one may thus speak of collective identity as
The process of collective identity 73
the ability of a collective actor to recognize the effects of its actions and to
attribute these effects to itself. Thus defined, collective identity presupposes,
first, a self-reflective ability of social actors. Collective action is
not simply a reaction to social and environmental constraints; it produces
symbolic orientations and meanings which actors are able to recognize.
Secondly, it entails that they have a notion of causality and belonging; they
are, that is, able to attribute the effects of their actions to themselves. This
recognition underpins their ability to appropriate the outcomes of their
actions, to exchange them with others, and to decide how they should be
allocated. Thirdly, identity entails an ability to perceive duration, an ability
which enables actors to establish a relationship between past and future and
to tie action to its effects.
The relational dimension of collective identity
Collective identity thus defines the capacity for autonomous action, a
differentiation of the actor from others within the continuity of that identity.
However, autoidentification must also gain social recognition if it is
to provide the basis for identity. The ability of a collective actor to distinguish
itself from others must be recognized by these 'others'. It would be
impossible to speak of collective identity without referring to its relational
dimension.
Social movements develop collective identity in a circular relationship
with a system of opportunities/constraints. Collective actors are able to
identify themselves when they have learned to distinguish between themselves
and the environment. Actor and system reciprocally constitute themselves,
and a movement only becomes self-aware through a relation with its
external environment, which offers to social action afieldof opportunities
and constraints, that in turn are recognized and defined as such by the
actor.
In this way, the unity of collective action, which is produced and maintained
by autoidentification, rests on the ability of a collective actor to
locate itself within a system of relations. A collective actor cannot construct
its identity independently of its recognition (which can also mean denial or
opposition) by other social and political actors. In order to act, any collective
actor makes the basic assumption that its distinction from other actors
is constantly acknowledged by them, if only in the extreme form of denial.
There must be at least a minimal degree of reciprocity in social recognition
between the actors (movement, authorities, other movements, third parties)
even if it takes the form of a denial, a challenge, or an opposition (Gamson,
Fireman and Rytina 1982). When this minimal basis for recognition is
74 Theory of collective action
lacking there can only be pure repression, an emptiness of meaning nullifying
the social field in which collective identity can be produced.
The autonomous ability to produce and to recognize the collective reality
as a 'we' is then a paradoxical situation: in affirming its difference from the
rest of the society, a movement also states its belonging to the shared
culture of a society and its need to be recognized as a social actor. The
paradox of identity always consists of the fact that difference, in order to
be affirmed and lived as such, presupposes a certain equality and a degree
of reciprocity.
Identity and conflict
Collective identity as a process can be analytically divided and seen from
the internal and external point of view. This separation of two sides is obviously
a way of describing what should be seen as a basically unified process.
Collective identity contains an unresolved and unresolvable tension
between the definition a movement gives of itself and the recognition
granted to it by the rest of the society.
Conflict is the extreme example of this discrepancy and of the tension it
provokes. In social conflicts reciprocity becomes impossible and the struggle
for scarce resources begins. Both subjects involved deny each other their
respective identities, refusing to grant their adversary that which they
demand for themselves. The conflict severs the reciprocity of the interaction;
the adversaries clash over something which is common to both of
them but which each refuses to confer to the other. Beyond the concrete or
symbolic objects at stake in a conflict, what people fight for is always the
possibility to recognize themselves and be recognized as subjects of their
action. Every conflict which transgresses a system of shared rules, whether
it concerns material or symbolic resources, is a conflict of identity. Social
actors enter a conflict to affirm the identity that their opponent has denied
them, to reappropriate something which belongs to them because they are
able to recognize it as their own.
During a conflict the internal solidarity of the group reinforces identity
and guarantees it. People feel a bond with others not because they share the
same interests, but because they need that bond in order to make sense of
what they are doing. The solidarity that ties individuals to each other
enables them to affirm themselves as subjects of their action and to withstand
the breakdown of social relations induced by conflict. Moreover, they
learn how to gather and focus their resources in order to reappropriate that
which they recognize as theirs. Participation in forms of collective
mobilization or in social movements, involvement in forms of cultural
The process of collective identity 75
innovation, voluntary action inspired by altruism - all these are grounded
in the need for identity and help to satisfy it.
Collective identity over time
Collective identity is a learning process which leads to the formation and
maintenance of a unified empirical actor that we can call a 'social movement'.
As that process passes through various stages, the collective actor
develops a capability to resolve problems posed by the environment and
becomes increasingly independent and autonomous in its capacity for
action within the network of relationships in which it is situated. The
process of collective identity is thus also the ability to produce new definitions
by integrating the past and the emerging elements of the present into
the unity and continuity of a collective actor.
It is above all in situations of crisis or intense conflict that the identity of
a collective actor is put to challenge, when it is subjected to contradictory
pressures which set a severe test for the ability of the collective actor to
define its unity. It can respond by restructuring its action according to new
orientations, or it can compartmentalize its spheres of action, so as to be
still able to preserve a certain amount of coherence - at least internally to
each of these spheres. The most serious cases provoke a breakdown or fragmentation
of the collective actor or a breach of its external confines. This
can lead to the incapacity to produce and maintain a definition of the
'movement' that could exhibit a certain stability or, vice versa, to the compulsive
assumption of a rigid identity from which it is impossible to escape,
as in sects or terrorist groups.
Collective identity ensures the continuity and permanence of the movement
over time, it establishes the limits of the actor with respect to its social
environment. It regulates the membership of individuals, it defines the
requisites for joining the 'movement', and the criteria by which its
members recognize themselves and are recognized. The content of this
identity and its temporal duration vary according to the type of group
concerned.
When we consider organizational structures, leadership patterns,
membership requisites, we deal with levels of collective action which presuppose
the notion of collective identity: they incorporate and enact the
ways a collective actor defines ends, means and field of his action. One
should consider those levels as empirical indicators of a possible collective
identity and, conversely, should use this concept as an analytical tool to dismantle
the 'reified' appearance of those empirical dimensions of a social
movement and to attain the constructive process behind them.
76 Theory of collective action
De-reification of collective identity
One cannot treat collective identity as a 'thing', as the monolithic unity of
the subject; it must, instead, be conceived as a system of relations and
representations. Collective identity takes the form of a field containing a
system of vectors in tension. These vectors constantly seek to establish an
equilibrium between the various axes of collective action, and between
identification declared by the actor and the identification given by the rest
of the society (adversaries, allies, third parties).
Collective identity in its concrete form depends on how this set of relations
is held together: this system is never a definite datum; it is instead a laborious
process where unity and equilibrium are reestablished over and over
again in reaction to shifts and changes in the elements internal and external
to thefield.Collective identity therefore patterns itself according to the presence
and relative intensity of its dimensions. Some vectors may be weaker or
stronger than others, and some may be entirely absent. One may imagine it
as a field which expands and contracts and whose borders alter with the
varying intensity and direction of the various forces that constitute it.
At any given moment both actors and observers can give an account of
thisfieldthrough a unified, delimited, and static definition of the 'we'. This
tendency for 'reification' is always part of a collective actor's need for
continuity and permanence. But today this unavoidable necessity has to
confront important changes in the ways identification takes place.
Identification processes have been gradually transferred from the outside
of society to its interior. From entities that are transcendent and metaphysical,
from metasocial foundations such as myths, gods, ancestors, but also
from the more recent avatars of God such as History or the Invisible Hand
of the market, identification processes shift to associative human action, to
culture, communication, and social relations. As identity is progressively
recognized as socially produced, it becomes obvious that notions like
coherence, limit maintenance, and recognition only describe it in static
terms; in its dynamic connotation, however, collective identity increasingly
becomes a process of construction and autonomization.
For recent social movements, particularly those centred around cultural
issues, collective identity is then becoming ever more conspicuously the
product of conscious action and the outcome of self-reflection, and,
correspondingly, loses its status based on a set of given or 'structural'
characteristics. The collective actor tends to construct its coherence and
recognize itself within the limits set by the environment and social relations.
Collective identity tends to coincide with conscious processes of 'organization'
and it is experienced not so much as a situation as it is an action.
The process of collective identity 77
To express this increasingly self-reflexive and constructed manner in
which contemporary collective actors tend to define themselves, I suggest
using the term identization. Within the boundaries of our language, it is a
rough and provocative acknowledgement of a qualitative leap in the
present forms of collective action, and also a call for an equivalent leap in
our cognitive tools.
The lens of collective identity: what one can see through it
Collective identity is a concept, an analytical tool and not a datum or an
essence, a 'thing' with a 'real' existence. As far as concerns concepts, one
should never forget that we are addressing not 'reality', but rather instruments
or lenses through which we read reality. The concept of collective
identity can function as a tool only if it helps to analyse phenomena, or
dimensions of them, that cannot be explained through other concepts or
models and if it contributes to the formation of new knowledge and to the
understanding of these same phenomena.
As was stated in the opening section, the concept of collective identity
was devised to overcome the shortcomings of the dualistic legacy still
present in the study of collective action, and the difficulties of the current
approaches in explaining some dimensions of contemporary social movements,
particularly the central role of culture and symbolic production in
these recent forms of action. It also addresses the naive epistemological
assumptions, often only implicitly present in the many approaches to the
study of social movements. It is then a concept that is intended to introduce
changes in our conceptualization of social movements and for this very
reason should contribute to a different understanding of the changing significance
of social movements in contemporary society.
These two levels, changes in conceptualization and changes in our understanding
of the practical-political significance of collective phenomena, are
connected by a circular relation. The circle is not a vicious one if concepts
help us to see more of the phenomena to which they apply, to see them
differently. Reversely, if these empirical phenomena are filtered and interpreted
through the conceptual lenses, they may help us to refine and
improve the quality of the lenses themselves.
1 The notion of collective identity is relevant to sociological literature
because it brings along with it afieldperspective on collective action and a
dynamic view of its definition. It implies the inclusion of the socialfieldas
part of the movement construction and it means that beyond the formal
definitions (public speeches, documents, opinions of participants) there is
always an active negotiation, an interactive work among individuals,
78 Theory of collective action
groups or parts of the movement. This again shifts the attention from the
top to the bottom of collective action and does not consider only the most
visible forms of action or the leaders' discourse. It looks to the more invisible
or hidden forms and tries to listen to the more silent voices.
Processes of mobilization, organizational forms, models of leadership,
ideologies and forms of communication - these are all meaningful levels of
analysis for the reconstruction from within of the system of action that
constitutes a collective actor. The whole of part IV will be devoted to a
consideration of these aspects. But also relationships with the outside, with
competitors, allies, adversaries, and especially the reaction of the political
system and the apparatus of social control, must be taken into account to
understand how the collective actor takes shape, perpetuates itself or
changes. The importance of this dimension has been stressed for example
by authors like Gamson (Gamson, Fireman and Rytina 1982), Tarrow
(1989, 1994) McCarthy (Me Carthy et al. 1991; McCarthy 1994) and will
be developed in particular in part III.
2 The concept of collective identity can also contribute to a better
understanding of the nature and meaning of the emerging forms of collective
action in highly differentiated systems. As the quantity and quality of
work in the area has increased and improved our understanding of recent
phenomena (Rucht 1988, 1990, 1991; Scherer-Warren and Krischke 1987;
Dalton and Kuechler 1990; Pakulski 1991; Koopmans et al. 1992; Kriesi
1993; Kriesi et al. 1995; Giugni 1995; Ahlemeyer 1995; Proietto 1995), we
know that contemporary 'movements' increasingly address cultural issues
and tend to differentiate themselves from the model of political action. The
concept of collective identity aids in the making of distinctions that mark
off this cultural level from all others, and particularly from dimensions that
are political in the proper sense. Such dimensions do not disappear from
the scene, but come to play different roles that can be captured only if one
relies on conceptual tools that allow the recognition of the complexity of
present collective actors, tools that do not take for granted 'social movement'
as a unified and homogeneous reality.
3 Collective actors are neither historical heroes nor villains. By identifying
specific levels that enter the construction of collective identity, movements
can be seen as action systems. They are not 'subjects' that act with
the unity of purposes that leaders, ideologues, or opponents attribute to
them. They are always plural, ambivalent, often contradictory.
4 The concept of collective identity has important consequences in
clearing up the misunderstanding on the so-called new social movements.
The notion of 'newness' that I introduced as a temporary and relative qualification
of emergent collective action (Melucci 1980) has been criticized by
The process of collective identity 79
comparing different historical cases (see as examples of critics of the
'newness' Kivisto 1986; D'Anieri et al 1990; Scott 1990a; Calhoun 1993),
or it has been attributed to recent movements as a whole as their intrinsic
quality (Offe 1985a; Dalton and Kuechler 1990). Contemporary movements
are not 'new' or 'old' in themselves, but rather comprise different
orientations with their components belonging to different historical layers
of a given society. The notion of collective identity can help to describe and
explain this connection between the apparent unity, which is always our
empirical starting point, and the underlying multiplicity, which can be
detected only by an appropriate analytical tool.
5 Another important consequence of the concept of collective identity
has to do with the theory of domination and conflict. The notion of collective
identity can prevent sociological analysis from ridding itself too
quickly of the theoretical question of whether there are dimensions of contemporary
collective action which express new systemic conflicts and challenge
new forms of social domination in complex societies. To dismiss this
question by reducing recent movements only to their political dimension is
to hide or deny the new location of power.
I have suggested that collective action of many recent social movements
constitutes a communicative act which is performed through the form of
action itself, making visible new powers and the possibilities of challenging
them. Action still pursues political goals or instrumental advantages, but
within a limited scope and with a degree of interchangeability. But in so
doing it also throws light on hidden issues that are not accounted for by the
rationality of dominant apparatuses.
6 However, this antagonist dimension cannot explain everything, and
the concept of collective identity is a permanent warning about the necessity
of recognizing a plurality of levels in collective action. This is perhaps
the most important contribution that the concept of collective identity can
bring to the field of social movements studies.
7 Finally, collective identity has some radical methodological implications.
Sociological analysis is not free from the risk of reducing collective
action to just one of its levels - which in fact is often the 'official' definition
of a movement - and of considering it as a unified empirical object. When
sociology still rests on an essentialist idea of social movements as characters
acting on the stage of history, it may contribute, even unwillingly, to
the practical denial of difference, to a factual and political ignorance of
that complex articulation of meanings that contemporary movements
carry in themselves. Putting into question the unity usually taken for
granted by ideologists, sociology may help to reveal those dimensions of
collective action that are not visible at first sight. To understand how a
80 Theory of collective action
'social movement' succeeds or fails in becoming a collective actor is therefore
a fundamental task of the sociologist.
Actors, of course, act under the practical necessity of having to hypostatize
their action-in-the-making in order to be able to speak about it. So
do the opponents and the observers, including the researcher.
'Objectification' is a basic feature of the operation of human cognition and
also a means of cognitive economy employed in speaking about the world.
Yet this does not release us, as researchers, to taking this reification for
granted, as if the relational texture of social phenomena would thereby disappear.
The task of analysis is precisely that of deconstructing this apparent
reality and letting the plurality of relations and meanings appear.
This analytical task allows some distinction between collective identity
as a constructive process and its objectified results: collective actors have
always a 'public identity' (Johnston et al. 1994), they always act as 'historical
actors' (Mueller 1994), in the same sense in which individuals present
their self according to Goffman. But the concept of collective identity can
help us precisely to question the surface and to reach the deep relational
texture of the collective actor. How means and ends are interpreted by
different groups of the movement? How are resources and constraints held
together in the movement discourse? What kind of relation with the
environment shapes the movement and how do its different groups interpret
it? What kind of conflicts, tensions, and negotiations can be observed
during the process of construction and maintenance of a movement as a
unified empirical actor? These are some of the questions that can be derived
from the concept of collective identity and that bring us closer to the beginnings
of a different research practice. Chapter 20 will deal with the
methodological consequences of this theoretical stance.
Identity and collective emotional experience
In conclusion to the preceding discussion, I want to briefly engage a level
of analysis whose importance has often been underevaluated by sociologists.
Smelser (1962) has already stressed the need for an analysis of collective
behaviour which moves simultaneously on various levels and which
combines, yet without confusing them, both psychological and sociological
variables. Attempting to go beyond the simple juxtaposition of these two
different points of view, I will address an analytical level which one might
call collective experience, which involves people's feeling and emotions, and
which should be neither confused with other dimensions of collective
action nor generalized into an overall explanatory model. I believe it
important for the understanding of the formation of collective action to
The process of collective identity 81
complete an analysis of the meaning it assumes at the level of emotional
dynamics. This is a complementary dimension which must be kept distinct
from 'structural' analysis, but which nevertheless forms a constituent part
of any analysis which takes seriously the task to understand 'action', not
merely behaviour. Indeed, many misunderstandings and much pointless
debate has been provoked by the confusion of these two levels. The analyst
is caught between, on the one hand, the risk of an explanation using the
categories of dynamic psychology, generalized at the level of structural
explanation, and, on the other, the endeavour to use structural analysis also
to explain processes which involve the motivations of individuals, the emotional
meanings of action.
I shall attempt to make a step in the direction of overcoming this simple
duality by indicating a possible level of analysis in terms of emotional experiential
dynamics. Smelser has shown that collective action expresses both
of the terms of the ambivalence (love/aggression) which characterizes relationships
from an emotional point of view. This observation is echoed in
Alberoni's theory of statu nascenti (Alberoni 1984), in which he formulates
'three principles of the dynamics' which regulate the relationship with the
love-object: (i) ambivalence (love/hate); (ii) reciprocity of the energetic
investments (love to those who love, hate to those who hate); (iii) tendency
to reduce ambivalence.
Developing such a perspective, we can assume that, in terms of emotional
investments, the objects of love are always affected by the love/hate
ambivalence. It is difficult to endure the ambivalence and to accept its
contradictory emotional content. If we hate those who we love we feel
guilty. When the emotional charge of ambivalence reaches a threshold
where it becomes unbearable for where it exceeds the actors' available energetic
capacity, an attack on the enemy could constitute the defensive reaction
reflecting the guilt of hating the loved one. Blaming the enemy projects
onto him/her that part of aggressive tendencies towards the loved one
which one cannot tolerate. By separating the terms of the ambivalence this
enables both of them to be covered through investing the love-object only
with eros and discharging aggression onto the enemy.
This model can be applied to what I call antagonist social conflicts. At
the level of experiential dynamics, conflict takes the form of a reduction of
ambivalence. Humankind's relationship with its own production is a relationship
with objects imbued with an energetic positive charge (loveobjects),
for we identify ourselves with our products and always invest them
with emotional and not just instrumental meaning. But love-objects are
also those which enslave us and from which we become dependent. Hence
the ambivalence towards our own production. In social conflicts, the
82 Theory of collective action
struggle among adversaries for control over social production is a way of
reducing this ambivalence. Each party reduces the ambivalence by onesidedly
projecting one of the poles onto the other; each one attributes to
the adversary in a projective way the aggressiveness s/he feels towards
her/his own love-object. Opposition towards the adversary who appropriates
or threatens the love-object redirects aggressiveness onto him/her and
resolves the ambivalence towards the loved one, who thus retains only positive
features.
This explanation applies to ongoing conflict but leaves its genesis unexplained.
It does not discuss, that is, how the situation arises in which the
same object (social production) is contested by adversaries and in which
each perceives a threat from the other. Reflection on the process whereby
control over production is delegated in the process of division of labour
and of social differentiation (the sociological reasons I have already
addressed in chapter 2) may shed light on the matter. The distribution of
control over the allocation of social production and the creation of imbalances
of power among social positions involves a transfer of ambivalence.
Those delegating power and control to others reassign onto the function
delegated the ambivalence which characterized their relationship with the
love-object. To delegate power is not only a relieving and facilitating experience;
it also implies a sense of loss and incapacity. Those delegating power
invest the ones in power with aggressive emotions, in order to free their
love-object (their own action) from any negative feeling. The recipients in
the delegation experience the same situation of ambivalence vis-a-vis their
mandators. To be invested by other people's mandate and expectations is a
gratification and a honor but also a burden. The aggressive feelings are
diverted towards the mandators, so that the action of those in power can
be kept uncontaminated by negative emotions.
For both of the actors, the relationship marks out a situation of
loss/conservation of the love-object. Those delegating power lose the direct
control over their action, but through aggressiveness directed at the adversary
can at least partially repair this sense of loss and preserve the positiveness
of their action. Those in power lose the unburdened freedom of not
having to exert control and take care of others, but they can assign the
blame to the mandators and channel towards them their aggressive feelings.
When there exists a situation of direct control over the delegation relationship,
a certain reciprocity of recognition prevails between the partners,
for each of them accepts of receiving from the other something that is considered
equivalent to one's own investment. One accepts the burden of
exerting the function delegated to her/him; in exchange, however, recognition,
deference, and gratification is received. One accepts to delegate power,
The process of collective identity 83
but only to receive in exchange service, help, support, security. The situation
can be interpreted as a type of relationship which permits reduction of
ambivalence through a limited acceptance of the pain connected to partial
loss, which is perceived as tolerable. Aggressiveness towards the other takes
the form of subjugation by the delegator and that of service by her/him who
exercises the control function. Reciprocal direct control and frequent faceto-face
confrontations, as in small-group situations, allow continuous monitoring
of the relationship, reduce the risk of an aggressive clash, and
favour the possibility of making sense of the situation as a limited and
acceptable loss, which is reciprocal (submission in exchange for service and
responsibility, and vice versa).
But when, for the structural reasons we have already examined
(differentiation of the system, increasing distance among social positions),
the possibility of direct control over the delegation disappears, so too does
the reciprocity of recognition break down. We may presume that the sense
of an acceptable loss no longer predominates and aggressive feelings gain
in prominence. Uncontrolled ambivalence emerges once again and triggers
the mechanism whereby aggression is redirected onto the adversary. Thus
is created the situation of conflict described above.
This brief outline provides one possible reading, in terms of collective
emotional experience, of the 'structural' theory of conflict that I presented
earlier. The opposition of the adversaries and the desire to appropriate
social production between its different producers could thus be explained
also in terms of the emotional investments that characterize social production
and collective action.
Collective identity, the construction of a 'we', is then a necessity also for
the emotional balance of social actors involved in conflicts. The possibility
of referring to a love-object ('Us' against Them') is a strong and preliminary
condition for collective action, as it continuously reduces ambivalence
and fuels action with positive energies. Collective actors constantly
need to draw on this emotional background in order to feed their action, to
make sense of it, to calculate its costs and benefits. When facing changes, the
necessity of renewing and possibly renegotiating the bond that ties individuals
and groups together originates from this deep emotional commitment
to a 'We' which must maintain its integrity in order to motivate action.
Collective identity in historical context
One could argue that the concepts proposed by my theoretical framework
are historically related to a very specific wave of collective action, that of
the movements which started to appear in the Western countries in the
84 Theory of collective action
1960s. It is indeed difficult to separate the analytical level from its historical
sources which, in my case, were the movements which I started to
analyse in the early 1970s. The concepts on which my work relies are certainly
influenced by contemporary social movements; yet I have always seen
it important to draw the distinction between the conceptual level and the
empirical analysis of concrete social movements. The concept of collective
identity is important for my work as it provides a way of addressing the
question of how a collective becomes a collective, which is usually taken for
granted. In other fields of sociology the situation may differ, but in the case
of collective action the question itself is apparently raised by no one. We
usually take the collective actors for granted and quasi-spontaneously
attribute a kind of essentialist existence to them. The theoretical problem
for us today is this unity, the creation of a collective subject of action as a
process which needs to be subjected to explanation.
Of course, this question could probably not be raised for the movements
of the past, not because these processes were inactive in earlier movements
but because they were less important and less visible. Collective actors of
the past were more deeply rooted in a specific social condition in which they
were embedded, so that the question of the collective was already answered
from the beginning through that social condition that accounted as such for
the existence of a collective actor. A working-class movement is first of all
the expression of a working class social background; it is already defined
by the social conditions of that particular group. For the working-class militants
it was extremely important that they belonged to a specific culture
which was organized in structures of everyday life, in forms of solidarity
which shaped the identity and grounded it in the material and cultural
conditions of the everyday (Calhoun 1982; Fantasia 1998). This continuity
between the structural location of the actor and the material and cultural
world of its experience is what I mean by the class condition.
Today, as we are increasingly more dealing with movements which
cannot be referred to any specific social condition, the question of how a
collective becomesjust that has become more prominent. In my work, I was
able to raise this question as it was focusing precisely on such particular
realities, but nevertheless the question in itself seems a reasonable scientific
question which can be addressed also to other historical movements.
Theoretically speaking, this question also could not be raised within
another cultural and intellectual context, as there simply existed no social
and conceptual space in which to advance thought in these terms. We are
thus always proceeding within the circular relationship between concepts
and objects, but if concepts prove heuristically effective once they are in
place they can nevertheless find application beyond the historical context
The process of collective identity #5
in which they were produced. They can help us to see differently objects in
other historical contexts.
A final note could be added concerning the use of the notion of collective
identity. As already stated, the term 'identity' is conceptually unsatisfactory:
it conveys too strongly the idea of the permanence of a subject. At
this moment, however, no other designation seems in possession of the
capacity to replace it in its purpose. Thus, for the time being we must continue
being trapped in the usage of the term in the near-contradictory situation
where in order to bring to light the processual dimension of collective
identity as an interactive construction, we inadvertently stress the reality
and the permanence of the actor. What I am trying to do with this concept,
however, is to bring it to its limits. Scientific enterprises proceed in this
manner, through an increasing effort to 'use up' the relevant concepts until
they reach their internal limits, to allow then the situation itself to affect a
change in concepts when the older tools have proven useless and outdated
themselves. To be sure, there exist even now notions that better stress the
dynamic side of identity, but they seem partial successes as well. The term
'project', for instance, points out one dimension of identity: the capability
of relating to the future, starting from the present. However, one has to be
located somewhere in order to think of the future, to be rooted in an already
established definition in order to have projects; otherwise one is left in the
world of fantasies and images. A 'project' is possible only when a location
somewhere in the present is established. It is the here and now, and this
point of consistency provides the only possible starting point to think of
the future. But the relation to the past is equal in importance to the projectual
dimension in the definition of an identity (Halbwachs 1975; Namer
1987; Middleton and Edwards 1990). The relation with the past is necessary
the same way as the capacity to make projects, and this has become
particularly concrete and obvious against the backdrop of the dominant
trends in our present culture. Ours is a culture which is making of speed
and change its central values, and through them it creates new forms of
power. There is a rhetoric of change and speed as values that reveals its
belongingness to the dominant logic of big organizations, of production
and circulation of goods, of the world media system; its consequence is the
erasure of past and permanence.
More than in any previous culture identity today is in need of a relationship
to the past. Such a relationship is created by the necessity of retaining
something while changing, of maintaining roots, of reconstructing our
history without which there is no possibility of progress. The whole notion
of time is redefined with the new relationship to the past and the future: the
present becomes a crucial dimension, not as a point-like, instantaneous
86 Theory of collective action
dimension but rather as the possibility of forging in the here and now the
connection between the past and the future, between memories and pro-
jects.
At the same time, however, the use of the notion of identity addresses
and points out a contradictory situation which is important for collective
actors. Every actor is faced with a two-sided problem. On the one hand, the
actor must maintain a permanence which, on the other, must be produced
continuously. This tension is always present, and probably the currency
acquired by 'identity', the apparent paradox (speaking of the 'process of
identity') it contains, captures something of it by signalling the contradictory
necessity of permanence in the continuous constructive process. At
any given time, when requested its identity, a collective actor (excluding the
extreme, completely schizophrenic situation) is able to provide an answer
through its many mouths in a definite way. Any such stable definition,
however, is at the same time the outcome of constructive processes.
Ultimately, identity becomes a matter of the question that is asked and
of the position taken by the observer. If one is interested in defining who
the actor is at a given moment, identity provides a useful concept. Should
one, however, be more interested in the constructive process behind its
formation, then probably other concepts should be created that are more
appropriate than identity for addressing this particular point of view and
for account for the tension between maintenance and production of the
definition that a collective actor gives of itself.
Having outlined our theoretical framework, analysis of 'movements' in
complex societies may now begin.