A Method for Studying Social Actors

So little agreement exists on what constitutes sociology that it seems impossible to de?ne its speci?c methods. We can however proceed through a series of eliminations. Light has been shed on many types of social behavior as economic studies have taken more interest in issues of social strati?cation and mobility or in consumer behavior, and as they increasingly incorporate elaborate quantitative analysis into this type of data. Correlations between social statuses and social behavior tell us about the logic of the system, yet not about that of the actors. Hence, we must imagine other methods in order to reach the actor as an autonomous being, as an agent of transformation of his environment and of his own situation, as a creator of imaginary worlds, as capable of referring to absolute values or of being involved in love relations.

This double dissociation leads us to a classical defi nition of sociology.Of course, sociology is the study of social relations, to the extent that such relations are established between two or more actors and that such relations are meaningful and take place within a social framework defi ned either as a social order, as a process of social change, or as both at once.All social relations are limited on two sides: one is never free to do absolutely anything; there exists no society without limits, norms and institutions.There exist no purely creative social relations.But, at the same time, behavior is not determined.The intentions and the values upon which social actions are based can never be entirely reduced to norms.If such norms were sovereign, there would be no deviance, no invention, no claims, no change.Thus we can broadly and initially defi ne the specifi c domain of sociology (without undermining in any way the importance of socio-economic and socio-historical studies) as the study of the relations between social determinism and freedom, between the application of social norms and the reference to human values, between institutional order and practices that voluntarily deviate from such norms.We must therefore stress that, within the present constellation of the social sciences, sociology is concerned with the study of actors and communication, while economy and history analyze the mechanisms and organization of a great variety of different systems.Such a defi nition is suffi cient for us to ask ourselves what is the specifi c method of sociological research.The answer to that question will help us to understand better the object of sociology.
the actor outside of the system Notwithstanding those schools of thought that reduce social action to rational choices of an economic nature, or to the manifestation of cultural patterns or social institutions considered to be determinants of individual and collective action, we constantly face the question: how does the distancing from established norms lead to creative freedom, rejection of old rules or non socially regulated emotions and fi nally to the creation of new norms?This problem is central because all forms of social communication produce change.For communication exists only to the extent that the original message is altered through the interaction between the speakers.Yet change is made up of two complementary aspects: the destruction of established norms and the construction of new ones.Sociology can accept neither extreme interactionism, doing away with any reference to social norms in the course of the exchange, nor can it accept an extreme sociological determinism, which reduces change to the confrontation with an enemy, to a cultural transformation imposed from without.The major diffi culty for sociology is to combine order and change, creativity and conformity.
This explains why sociology has more and more distanced itself from defi nitions of social actors which refer to their statuses, to the roles they play or are expected to play.This change of perspective is characterized by the passage from a system-oriented sociology to an actor-oriented one, by the substitution as a central approach of "actor' s projects" for "social situations."These projects combine in the most diverse manner a desire for integration and the defense of one' s own identities.The actor tries to attribute a social meaning in harmony with his own self-image.Sociology has thus come very close to social psychology, to the point of incorporating it within its realm of study.Let us take an example.We speak of immigrants (even though they may be born in the country), of the banlieues or inner cities (even though such defi nitions are often rejected by those concerned), or of marginalization (which deprives the actor of his autonomy).Moreover, in all countries, interesting research analyzes the behavior of those who are considered as foreigners and of those who react to their arrival, often through xenophobia.But the two dominant approaches are to consider immigrants solely as victims of inequality, injustice and prejudice, or to study the various phases of their integration and assimilation.However, studies have shown both the diversity of responses to a situation and the frequency of initiatives that lead these victims to become actors by constantly redefi ning their horizon, their possibility for action and the nature of the obstacles facing them.Actors, in this case as in many others, are not defi ned by their conformity to rules and norms, but by a relation to themselves, by their capacity to constitute themselves as actors, capable of changing their environment and of reinforcing their autonomy.
To recognize the absence of correspondence between situation and actor leads us to an even more profound change in sociological analysis.We have been accustomed to defi ne social behavior in terms of oppositions, which, in one way or another, corresponds to a general opposition between tradition and modernity.Such an approach implies that in order to understand an actor one must be able to place him or her on one side or the other of the axis that separates tradition from modernity.This means that actors are defi ned in the same terms which characterize the society to which they belong.
If, on the contrary, we separate the categories that are applied to actors from those which describe situations, we are led to defi ne actors' attitudes vis-a-vis a social organization or a norm by the idea of ambivalence.The notion, fi rst used by Simmel, is once again gaining importance after having been somewhat ignored.For ambivalence implies the impossible identifi cation of the actor with a situation defi ned in historical, economic or social terms.The actors, led by categories pertaining to themselves and not to a situation, are in a state of ambivalence, both accepting and rejecting any situation in which they are placed.We both demand and reject progress, modernization, order and all other categories that defi ne a situation in normative and "objective" terms, at the same time as reconstructing the actor' s ambivalence with regard to all aspects of his situation and the type of reference to their interests, interpersonal relations, or values.
This explains why qualitative research is once again gaining ground after a period during which quantitative studies were predominant.Such studies are of course indispensable, although for economic rather than for sociological analysis, since they separate various types of behavior rather than undertake the necessary recomposition of the actor' s "project."

the meaning of action
There is in all forms of social action, and therefore in all processes of social communication, an element of non-conformity, an innovative claim or the assertion of a non-social principle and a claim for the limitation of society' s norms or institutions.This formulation is extreme, maybe excessive, and might even appear to draw us away from the fi eld of sociology and enclose us within the realm of ideology and of the actors' subjectivity.There is, however, no other basic characteristic of what we call social action.Therefore, sociological research must fi rst of all uncover this reference to a nonsocial principle, whatever it may be.Is this non-social principle individual interest or principles deemed to be superior to social rules?Are they both able to displace or transform social norms?
The answer must be negative.Action can seldom be explained in pure terms of individual interest, especially when it is collective.Those who seem to defend "interests" may go sacrifi cing their own lives, when confronted with the resistance of their adversaries.On the other side, the degree of conviction is linked to the context in which it is expressed, making this more or less radical, or, on the contrary, open to debate, so that both interpretations of the non-social principle of social action are unsatisfactory.What we observe is different: the more an actor feels his or her self-esteem threatened, the more norms and institutions appear to him or her unfair or illegitimate.He or she becomes convinced that social norms have not in fact been created for the common good, but rather that they are the expression of a power which endangers freedom, responsibility and dignity of people.In its most extreme form, this confl ict can lead to a holy war, or to a confl ict based on the opposition of friends and enemies, as defi ned by Carl Schmitt.
This brief analysis brings us directly to the elaboration of one of sociology' s specifi c methods.It is the aim of the sociologist to uncover claims, confl icts or debates, which are often overshadowed by the authority of social norms and by repression which is imposed for the sake of institutions or of those who hold positions of power.If we admit that the reference to a non social principle can be neither entirely unconscious (this would make all analysis arbitrary), nor completely conscious (this would reduce sociological enquiry to the study of opinions), the sociologist' s aim must be to unveil or to reinforce behavior and practices that are latent or reduced to biased expressions.
We are often tempted to defi ne the role of the sociologist as passive: it is he who must create the conditions under which might best be expressed the "deep" feelings of the actors.Yet this tends to give free range to the infl uence of institutional norms.Studies that are conducted in a "cold" situation tend to produce results showing the importance of conformist behavior, not withstanding that a wave of protest or strikes may disrupt this apparent state of tranquility in the space of a few days.Must one intervene in the heat of the action then?The opposite criticism might then be formulated, since when engaged in confl ict or when affi rming his beliefs, the actor has a militant representation of the opposition between friends and foes.After the death of Robespierre the Revolution soon followed a different course and different attitudes were expressed.
Faced with these contradictory objections, the sociologist must create an adequate context, not only by choosing the time and space for the observa-tion, but by directly intervening.This means that he will study the behavior he has himself induced, not by taking the place of real actors, but rather by offering an interpretation of their own conduct.The researcher must bring actors to "discover" the highest possible meaning of their own action.If he proceeded in the opposite manner he would only create resistance to his intervention.By seeking, on the contrary, to uncover a positive meaning of an action, he might risk seducing the group.However, such a risk is limited, for authentic reactions soon reappear and reject the effects of seduction as overly optimistic hypotheses on the part of the researcher lead to rejection by disappointed or demoralized actors.
Such an approach is frequent among historians.While they always show how a historical fi gure or a social practice is representative of its time period, they are even more concerned, as is Jacques Le Goff studying Saint Louis, with uncovering the most personal, and at the same time the most innovative, aspects of a personal behavior.Since the sociologist studies the present, and can therefore create situations and documents rather than be entirely dependent upon pre-existing documents, it is his responsibility to be more "conscious" than those he is studying.However, he must, in the end, confront the actors themselves with his analysis and must fi nally determine whether it re-enforces or weakens the actor' s capacity for action and analysis.
Such is the defi nition of sociological intervention, whether socio-psychologically oriented on the inner life of a group, or truly sociological in that it introduces actors who in some way question institutionalized norms by claiming rights which must command social behavior.This is true even when interpersonal relations, such as family life, are concerned.Over the past twenty years, many such studies-known as sociological interventions-have been carried out.They have taken the following form: Over a long period of time two researchers study a group of individuals having participated in the same collective action, yet who do not form an actual unit, such as a strike committee or the directors of an existing association.In order to prevent the group from resorting to a self-indulgent self-consciousness one does not question it on its attitudes or opinions, but rather the group is directly confronted with partners or enemies chosen by the members themselves and whose presence serves to place the actors in a similar situation to that which has already been experienced.At the same time, the presence of researchers constitutes a direct reminder that actors are in a situation of analysis.While one of the researchers helps the group to make its positions more explicit, the other formulates a hypothesis concerning the highest possible meaning of the action in question.When this researcher communicates his hypothesis to the group, a favorable reaction is generally obtained, since the group feels fl attered by a judgment that gives such importance to its action.One then lets such an easy acceptance settle, and observes the level of implication at which the group is able to become stabilized.This requires a great deal of time and the confrontation of several groups as well as prolonged analysis of whether or not the hypothesis helps the group to interpret its own action effi ciently.
This intervention goes much farther than simple interviews, even prolonged ones, since it recreates a social situation.An entire group, and not simply an individual, participates in the intervention.Rather than simply being confronted with questions, the group is forced to interact with actual partners who hold positive or negative positions in relation to itself.Only after having reconstructed its actual experience under the sociologists' gaze does the group, with the help of one of them, engage in an interpretation of itself that will lead the other sociologist to formulate a hypothesis.When this hypothesis is solemnly presented to the group, signifi cant reactions are forthcoming whose meaning appears only after a certain period of time during which the group will have had the opportunity to react to a new situation.One may then observe whether the intervention has increased or diminished the group' s capacity to understand its situation.
This brief description of a method that, despite its length and complexity, has already been applied a number of times, leads us to formulate a general proposition: the best test of the accuracy of a sociological interpretation is its ability to increase actors' awareness and their capacity for action, since the very defi nition of an actor is his ability to construct and to justify behavior that cannot be reduced to the simple application of rules and norms.The historian has a better capacity than the sociologist to make use of the dimension of time, to verify a hypothesis formulated at a certain time by applying it to other time periods.But the sociologist, even though his perspective may not have the same temporal breadth, has the advantage of being able to modify his hypotheses according to the actors' responses.
This method, which is characteristic of an actor-oriented sociology, is far removed from more conventional methods.Yet, this is not so true if we include in economic analysis a large part of what is referred to as sociology.
In fact an increasing portion of writings which are considered sociological are concerned with social actors, and many among them deal with collective action and even with social movements carried out in the name of dominant or dominated actors.This explains why, for some time now, one of the main sociological debates has posed the question of whether dominated categories possess a critical awareness of their own subordination or whether they are alienated, manipulated, deprived of the knowledge of their own situation and of their possibility for action?It has often been said that, for instance, Jews who faced extermination, the unemployed, colonized people or prisoners had or have a limited awareness of their own situation, given the extent of their victimization.This has led many sociologists to consider that a power structure can be strong enough so as to destroy any potential for resistance among its victims.An opposite view, which I am presently advocating, says that no group, whether in a dominant or subordinate position, is entirely lacking in the awareness of hierarchical social relations and of the forms of domination in which it takes part.Sociologists, like historians, often help us to hear the voices of those whom many believe to have defi nitely lost the possibility for expression.In particular, we discovered that the "silence" of Jews in ghettos and camps did not exist.Thus, that which at fi rst sight appears to be an event, part of a series or of a larger ensemble, is discovered to be either the will of an elite for destruction or for domination, or the result of confl icting social claims, or of what might be called a political process of decision making.
As sociological analysis advances, what was previously called a social situation is shattered and the impersonal becomes intentional, the "objective" reality becomes simply one of the many possible outcomes.Are we not following such a movement when, behind the urban organization, we see the effects of urban policies, social unrest, speculation and political compromise?It is, for the same reason, possible to discover, as did Michel Crozier, that behind the organizing rules of an company are hidden strategies, confl icts and innovation.The study of social movements constitutes the heart of a sociology of the actor.Yet such an approach is not limited to the direct study of opinions, because these are infl uenced by institutionalized expressions of social life and mass communication processes.

reaching the subject
We must proceed even farther than this double movement of decomposition and re-composition of social norms and reach the really central principle of such an analysis: the idea that any form of action, to the extent that it is not limited to the conventional application of norms and rules, implies the recourse to a non-social principle.We must fi rst set aside the notion of rational choice.Not because such choices are unimportant both on the collective and the individual level, but because this domain pertains most notably to economic analysis.Liberals and Marxists converge toward the economically oriented analysis of social behavior, which cannot be but marginal to the sociologist.
Very far from this "rationalism," many forms of social conduct refer to a principle of equality that cannot be reduced to the ideas of general good and people' s sovereignty.In the 1789 French Declaration of Human Rights two principles are juxtaposed: "all men are born and remain free and equal in rights" (article 1), and the principle according to which "sovereignty resides essentially within the Nation" (article3).This article, inspired by the Rousseauian perspective, leads to the idea that good and evil are determined according to the interest of society, an idea expounded by Proudhon and then Durkheim.The fi rst article, on the other hand, refers to Locke' s idea of natural law, and thus poses limits to social power.The example of Human Rights, on which are founded both the French and the American republican systems may be generalized.Religious claims and sentiments cannot be reduced to a reference to the common good, in spite of the interpretations of so many sociologists.If religion expresses a certain sacralization of the social domain, it also refers to a divine power to limit or fi ght any social and political power.In other contexts, such as the workers' movement, the liberation movements of colonized peoples, or even women' s liberation, the main principle is that of equality more than the idea of justice, for justice embodies the equitable relation between what each and everyone contributes and what they receive.
Sociology, however, cannot directly call upon transcendental principle.It cannot assume the existence of God or of a perfect society.The only way to avoid such a theological perspective is by interpreting all forms of resistance against society as the search for a self-created human being who is also a creator of his social environment to the extent that he is involved in confl icts, negotiations and the struggle for freedom in the name of these rights.Far from discovering within society itself the defi ning principle of good and evil, we are referring here to what I have called the Subject, or in other words, to individual and collective claims to the right to become free actors.
The need to develop an actor and a subject-oriented society results from the fact that the idea of a norm producing social system has become weak and is shaken as capitalism and industrialization have progressively deprived legal and political institutions of their control over social life.While sociology of systems and sociology of actors cannot be entirely mutually exclusive, the latter has a superior capacity for discovery in present day society.
Yet how can the researcher grasp, beyond his social roles, the human being' s call for "dignity" without giving way to ideological claims, all the more noble and sincere since they are distinct from actual behavior?The answer is that the Subject has no positive content, for such a content would inevitably be overpowered by social determinants.As I have consistently argued, the Subject is a recourse, a refusal, a liberation, a direct or indirect call to the Subject' s creative freedom against social statuses and social roles.
Such a revelation of the Subject is all the more necessary at a time when one increasingly perceives a tendency for political and economic domination, manipulation and cultural alienation.Are we not living in a world of power, resistance to power and confl icting social movements rather than in a universe of rules, institutions and socialization processes in which so many sociologists imprison themselves?What the researcher can reveal is not the effi ciency of social control but the resistance against the established order, through testimonies of marginalized categories, through the resistance of dissidents, of victims, of theocracies, of participants in labor movements or campaigns of decolonization.Isn't it true that such critical and liberating action quickly becomes mixed up with ideological discourses; but the strength of resistance, of protest as its own end survive all efforts of destruction and manipulation?
All forms of such a refusal, of such a call to Human Rights constitute the core of social movements, of political programs and of movements which are limited at the public opinion level.Supposedly self-regulating social life is increasingly invaded on the one hand by the rise of the Subject and on the other by the logic of war and power.The aim of sociological intervention is to bring forth demands for reforms in the midst of socially determined behavior, of negations and denials on which is based behavior that refers primarily to principles but weakens as it transforms itself into strategies and political management.Such a case lately appeared in France.With almost no former organization, high-school students arose and for several days one could hear the strength of their appeal, expressed as the desire to be put at the center of the school system, to be actors of its transformation, to gain greater autonomy and to have better communication with their teachers.But rapidly the need to elaborate a strategy and to communicate with the media led them to formulate increasingly quantitative demands.They demanded more teachers, a lighter course load, and the improvement of the material conditions of their high schools.From this point on the social movement rapidly fell apart.
On a larger scale, and in a former period, working-class consciousness expressed the defense of their "craft" against managerial organization of their work.This working-class consciousness was generally hidden either behind legal demands or behind the dream of a proletarian and revolutionary society.Such a transformation of principles into attempts to reconstruct social norms is present everywhere.The highest form of struggle is not that which is most ideologically charged.On the contrary, it directly questions social organization, but for basic, non-social reasons (like the defense of justice or dignity) which always remain beyond the domain of negotiation.All social movements start by referring to the unacceptable, the intolerable, before moving on to the phase of possible negotiations and in the middle of these, one can always spot the presence of the non negotiable, that which is opposed to social norms.Even the most moderate actors appeal to a nonnegotiable principle, to the recognition of a moral principle opposed to the norms which impose and legitimate social organization.

sociology and its object-subject
What we have said about sociological intervention clearly indicates that the relation of the sociologist to his object of research is different from that which can be found in most other fi elds of research, which do not need to consider the consciousness of those under scrutiny, either because such a consciousness does not exist, or because one is dealing with "natural sciences of man," such as structuralist linguistics or anthropology.Sociology does not seek to do away with subjectivity.It gains the necessary distance from subjectivity by delving deeply enough within it to reach what constitutes the individual, or the group, as an actor.We are not stretching the limits of our vocabulary when we propose to go beyond subjectivity to discover subjectivation, that is the self-representation of an individual or a group as an actor, trying to impose their own ends to their environment.So that the sociologist' s relation to his object-subject is more diffi cult to defi ne than the relationship of other human sciences to their object of study.
What proof does the sociologist who follows our defi nition have of the validity of his interpretation, and how does he protect himself against his own subjectivity, his opinions or beliefs?One fi rst and elementary answer is that the study of a social actor or movement requires the participation of several research teams who analyze similar groups or documents.As we have already mentioned, such research is based on the principle that the main hypothesis may be rejected or validated by responses to situations that arise after the central moment of the research.As far as sociological intervention is concerned, I have known of several cases in which the hypothesis presented to the actors turned out to be false or required readjustment.Our own publications, and those of other researchers, testify to this fact.
But this answer is not suffi cient.How can the researcher be certain that he did not create a system of interpretations that fooled both researchers and actors?This can happen in particular if the dissolution or transformation of the actors cancelled the possibility of the hypotheses.Notwithstanding the fact that any research team must critically evaluate itself, the only possible answer is that the social actor is never defi ned on his own, by his assertions, his representations or even his practices, but is defi ned primarily through his relationships to other actors, whether different or similar, yet to whom this actor is connected by a specifi c relationship, in the fi eld of action which is studied.For example, to study union members, we confront them with employers or with non-unionized workers or members of a different union.This constitutes the second phase of sociological intervention: the reconstruction of the set of social relations, of confl icts, compromises or patterns in which the actor builds himself as an actor.Once again, the aim is to discover the actor as actor, in other words as a participant in "the production of society."This is done most directly through the study of the mechanisms that shape political decisions and public or private policies.
The reconstruction of the processes and meaning of decision-making is one of the best means for validating hypotheses concerning the meaning of the action in question.After having attempted to reach the Subject through and beyond his intentional actions, we must proceed to the study of organized activity and even of institutions and texts.Much like the historian, the sociologist seeks, even in what appears to be the most "institutionalized" value orientations, signs of actions and social relations of those who are agents of the "institutionalization" of values and cultural orientations.Sociological analysis loses some of its strength if it separates the two following moments: the upward oriented movement from the system to the actor, we might even say to the Subject, and the descent toward an explanation of organized or even institutionalized forms of social life.There is a constant movement between detachment from established social norms and the reformulation of different norms.This makes social change possible.
Under what conditions can a sociologist accompany or even precede this double movement of getting out of social organization, and-through a transformation of orientations and expectations-the descent toward modifi ed forms of social organization?The fi rst condition has nothing to do with the sociologist himself.He must observe serious disruptions, confl icts, failed negotiations, etc. From this moment on, the following question is posed: where will this lead to?Which level of social existence does it reject?Is it the organizational one, in other words the defi nition of the relative position of each actor and his rights and obligations?Or is it the institutional level, defi ned as the decision-making system which transforms the way in which an organization functions?Or, does the refusal or confl ict reach the level of the cultural orientations of a society and of its main power relations?The sociologist, taking into account available documents, and most of all what he has heard during numerous group reunions, especially those in which group members exchange views with their friends or foes, must intervene and defi ne what appears to him to be the highest possible level of action.
In fact, sociological intervention aims at uncovering the existence of a social movement, in other words, of opposition to the general orientations and to the system of power relations.But it is also possible to demonstrate that an action, while carried out under the guise of a social movement, is in fact a more limited demand or a political claim whose aim is only to modify the decision-making system.Once the researcher is convinced of the validity of a hypothesis, he seeks to convince the group.But quickly after this success, he must observe whether the group is capable of reinterpreting its action and of analyzing its potential in the proposed terms or not.This formula does not only apply to a particular research technique.Whether one is studying policy formation, work relations, family ties, the treatment of minorities, or institutions of an educational, religious or legal nature, the main test of the validity of one' s hypothesis is its ability to increase the capacity for action of those being studied, the prerequisite being a better understanding of the meaning of their own activity.The main danger is that the researcher is so active that he seeks to convince and therefore to seduce; that is why he must be under the surveillance of a second researcher whose role is to reinforce the group' s resistance and help it to voice its doubts and criticisms.

fields of research
The researcher who engages in this type of research is motivated fi rst of all by the rejection of positivist concepts of social organization, and by the belief that such organization is constantly created and undone through relations between actors who are never defi ned entirely according to the place they occupy within an organization.Is this really too much to ask of members of an ever changing society, full of technological innovation, social confl icts, and cultural transformations?The real diffi culty is how to distinguish between various processes and levels of change, from those who appeal to personal interests (and who therefore seek to accomplish rational choices), to those who oppose the ways in which the main material and cultural resources of society are used and withheld by such and such a group, class or social elite.Initially, there is a wide difference between those who seek to prove the existence of rational choices, those who want to understand strategies for change, those who wish to uncover truly political confl icts and those who look for social movements per se.Each of these positions, of these hypotheses, corresponds to a very specifi c sociological outlook.But all these approaches are part of an actor-oriented sociology, that I believe to be on its way to replacing the traditional system-oriented sociology.
Each analytical orientation refers to a particular method.They differ essentially according to the representation they present of social life.The study of rational choice supposes an individual actor, even though he must seek to fulfi ll his personal interest through collective environment.The study of strategies supposes an organizational domain characterized by constraints.At a higher level of social integration, the study of political realities involves actors who are already connected to one another by a relatively clear defi nition of their confl icts of interest.Finally, the study of movements requires that they be defi ned by social confl icts, while at the same time sharing with their enemies an explicit reference to the same cultural and social stakes.The higher up we move in the analysis, the more the relations between actors become defi ned in general terms.We could even reason the other way around, by proclaiming as the highest levels of analysis those who place the actor in a struggle for the appropriation of the cultural orientations and resources of a society.
This immense extension of the fi eld, and of the general method of a sociology of actors, does however have its limits.There exists a large domain of sociological research which has conquered much territory, especially within French sociology, and that refuses the concept of social actors and rejects even more the idea of the Subject.For these sociologists, the domination of power is complete.It imposes its own logic on all actors and turns them into agents of domination or into bearers of false consciousness.For these sociologists, social change can only arise from the demise of the social system, weakened by its own contradictions, or through the rise of new forms of production.The more industrial society has grown beyond mass production to include mass consumption and communication, the weaker institutional safeguards and roles expectations have become.
Let us mention the role of teachers as defi ned by these sociologists.Often convinced of their own responsibility for transmitting norms which serve a certain form of domination, they nonetheless resist the pressure of those in power while at the same contributing to their domination.Socialization is none other than the interiorization of norms which conform to the interest of those in power.The most infl uential works of Michel Foucault showed how the weakening of absolute power and authority acting from without gave way to the interiorization of norms and categories so consistent with the interests of those in power that they need less and less to punish, since they already manipulate minds and attitudes through mass culture, and even through social sciences.This is a brilliant argument that often speaks the truth, but which we have no reason to adopt completely, for it has never been proven that internalized social control is all powerful, that actors can be reduced to mere mechanisms of norm-oriented production and consumption.
Change exists and is inseparable from the agents of change, led forth by values, hope, anger, analyses or strategies.Yet neither is there any reason to wholeheartedly accept a sociology of actors as free and creative beings.It is impossible to specify the extent of the territory occupied by such an actor-oriented sociology, yet we can assert that it is growing.In any case, purely critical sociologists name only very minor groups when they designate the agents who are supposed to bring about the collapse of the system, weakened by its own contradictions, even if these sociologists are occasionally carried away by a wave of popular, yet not purely radical, discontent.
Although it is always risky to explain the state of the social sciences in reference to historical circumstances, one may suspect that over the last fi fty years the important infl uence of economic facts and analyses has naturally given rise to a reaction of purely critical sociology.The fragility of former social movements and main actors does not favor the development of an actor-and social-movement oriented sociology and its weakness enhances the infl uence of sociologists who try to demonstrate the absolute domination of power-holders.One can imagine that the decline of liberal illusions and the growing need to develop more voluntaristic economic policies will lead, as they have already done, to a renewal of actor-oriented sociology against which it is no longer possible to refer to the supposed vacuity of the social and political scene.But a period of high tension and of great danger can also bring movements to depend on outdated modes of analysis and to do away with the renewal of vocabulary and strategy.As Marx observed, this tendency to fall back upon the past has often been observed in French social history, but it soon exhausts the possibility for action and analysis.

historical verification
It is tempting to assert from the start that there are visible traces of an action, and then to consider the "internal logic of a situation" as an obstacle limiting the action' s effects.Such an approach can be useful in the case of very specifi c events whose cause and effects are closely linked, such as when a confl ict in a work situation leads to a collective agreement.In such a case one can judge the nature of the confl ict through the effects produced.But in more global contexts, one needs a more elaborate analysis of the situation in order to assert the effects of an action, and thus its nature.The more limited the defi nition of the situation, the easier it is to determine what might alter it.The most favorable solution is that in which the defi nition of the situation makes use of categories that can also be applied to the action one is seeking to explain.An economic emergency provokes relatively predictable social reactions.If these reactions are much more extreme than expected, then the correct explanation does not lie only within the economic situation itself.This feeling of estrangement from classical explanations is our best guide to research.When there exists no obvious cause for an action, then we are forced to consider, fi rst of all, the nature of the action under consideration.The vision of history as a continuous process of change is so weak, so undemanding, that we are immediately forced to analyze the intrinsic nature of the action, before we seek to discover its effects, too easily characterized in such a superfi cially and arbitrarily defi ned situation.
Thus we easily go from the apparently accurate idea that the cause and effects of an action are revealed by historical events, to the opposite idea that the fl ow of events can be understood only by isolating various kinds of action.The sociologist must analyze a system of agents and the confl ict or non-confl ict relations between actors who, within a shared fi eld of cultural resources, all seek to infl uence their social use.The more elaborate such an analysis becomes, the easier it will be to fi nd its equivalent in the analysis of events.If a certain type of action leaves no trace at any of the levels of social life (organizational, institutional, historical, and class-based) we must prudently conclude that we are dealing with representations and ideology rather than with action.This happens frequently.I defended the idea that the main interpretation given in the heat of the moment during the massive French strike in the winter of 1995 was a false one.The absence of continuity, the failure to prolong the action, are important, if still partial, arguments to reinforce my interpretation.Such an interpretation was based on the defi nition of the action, of its goals, its rallying calls and its own discourse.It is obvious that the more we construct political, economic and social schemes of analysis, the richer will become historical analysis, since, rather than describing change, such analysis must observe the complex relations between diverse types of social action.

on the use of comparison
Nothing appears farther removed from the orientations we have exposed than comparitivism.Such an approach seems to combine the defects of both evolutionism and of culturalism.Either it defi nes very large units, such as religions or empires, in order to show what distinguishes them, or, on the contrary, it places such entities on an evolutionary path which generally leads to the observer' s own culture or society as an "end of society."How numerous are the French, British, German or American books that conclude to the superiority of their own civilization, never entirely equaled by others!Nonetheless comparisons are indispensable and ever-present because they make it possible to separate the analysis of actors and situations from the understanding of processes of change, two approaches that are too often confused.To speak of industrial society is to defi ne the actors' social relations and the ultimate meaning of their interactions.To speak, on the contrary, of capitalist, socialist, or dependent societies, is to analyze and compare processes of social transformation, such as industrialization.This latter type of study can proceed only through comparison, since it implies the existence of several paths to modernization or to political change.On the contrary, a societal type is defi ned less through its contrast with other types of society, than in terms of its internal dynamic, and thus of the formation and behavior of its actors.
These two types of study are complementary, since industrial society cannot be entirely disconnected from processes of modernization, even though the West likes to believe that it represents modernity itself and not simply one form of modernization.The comparative method is the only security against the failure to differentiate between the way a society functions and its processes of change.This explains the importance of a sociology of development.It has been rejected because it was too often dominated by an ideology of modernization which considered that all roads lead to London, New York or Moscow, and that "under-developed" countries would overcome their condition only if they were to follow in the path toward progress set by developed countries.It is by rejecting such a sociocentric vision of modernization that development studies may be useful for understanding how the unity of each societal type is combined with the diversity of historical processes that give rise to each type.The most we can say is that historians are more easily attracted by the comparative method, while sociologists center their studies on the construction of systems of actors.
Sociology was able to formulate a defi nition of itself when it was functionalist, in other word, when good and evil were defi ned in reference to social utility.After Talcott Parsons, the last great representative of this functionalist approach, sociology has accepted too often weak defi nitions of itself as the study of contemporary societies, as the reconstruction of social evolution, and sometimes simply as the analysis of national societies.A limited defi nition of the sociological approach, as I have presented it here, makes it more diffi cult to analyze its relationship to other disciplines, since they overlap the social domain from all sides.It is more useful to describe sociology through its method, rather than through its object.Some may consider such an attempt as too ambitious and will accept the disappearance of sociology as an analytical fi eld.Such an idea was largely accepted at the time when post-structuralism was triumphant, and there seemed to remain only, on the one hand structural approaches such as linguistics and anthropology, and on the other the historical approach which emphasizes the uniqueness of each situation.This epoch and its representations now seem very far away.Consequently, it is not an exaggeration to predict that the specifi city of sociology as the study of social actors and their relations will be recognized.Sociology will succeed in making a place for itself only if it is willing to engage in a debate over its methods.Not in order to give way to self-criticism and self-destruction, but in order to defi ne itself through the search for autonomous actors, between economic determinism, the logics of power and religious, political or class-oriented theories.