Skip to navigation – Site map
Symposia. Pragmatism and the Social Sciences: A Century of Influences and Interactions, vol.2
Section I. Classical Pragmatists and Contemporary Sociology

The Social Scientist, the Public, and the Pragmatist Gaze

Exploring the Critical Conditions of Sociological Inquiry
Philippe Gonzalez and Laurence Kaufmann


Although diverse and sometimes diverging, different approaches, from “pragmatic” to “pragmatist” to “praxeological,” have an important feature in common: the social order is said to be the practical accomplishment of ordinary agents who constitute and maintain in common the world they live in. After presenting the milestones of the main sociological version of pragmatism, that is, pragmatic sociology (sociologie pragmatique), initiated by Luc Boltanski and Laurent Thévenot, this paper will dwell on the complicated relationship between a pragmatic framework, centered on the insider point of view of agents and social critique, which apprehends the social world from the external point of view of the critical sociologist. To tackle such problematic relationship, we will dwell on two groundbreaking contributions to the “pragmatic turn” in the social sciences, that of Jeanne Favret-Saada’s work on contemporary witchcraft, and Michel de Certeau’s study on 17th Century possessions. They will allow us to show that the revival of the epistemological break is not a necessary step towards political awareness and that pragmatic sociology as such can be fully critical.

Top of page

Full text

The public compose a tribunal, which is more powerful than all the other tribunals together.
J. Bentham ([1791] 1843), An Essay on Political Tactics

1Within the last decades, a “pragmatic” – if not “pragmatist” – turn has taken place in the French social sciences, bringing about new ways of inquiry to empirical fields such as religious conversion, judicial trial, scientific controversy, or media coverage. Although diverse and sometimes diverging, those approaches share an important common feature, already present in the classical pragmatism initiated, among others, by Charles S. Peirce, William James, John Dewey, or George H. Mead. Instead of starting either from the individual or the social structure in order to subsequently establish how both could be related, they take action itself as their point of departure. The social order is thus said to be the practical accomplishment of ordinary agents who constitute and maintain in common the world they live in. But this common focus on action is very vague – vague enough, at least, to give rise to the various interpretations that the profusion of different action-centered schools of thought makes explicit, from “pragmatic” to “pragmatist” to “praxeological.” In fact, each of the main schools favors one specific aspect of action that is worth dwelling upon: (a) the central part played by action in the production of social order and the indeterminacy of human agency, mostly stressed by ethnomethodology and praxeological sociology (b) the performative power of language and actantial schemes to structure and institute social action and relationships, mainly focused on by linguistic pragmatics (c) the epistemological primacy of practices and the importance, for the making of a public, of a common, “bottom-up” inquiry into the indirect consequences of transactions, mostly emphasized by philosophical pragmatism.

  • 1 Among others: Marc Breviglieri, Damien de Blic, Nicolas Dodier, Cyril Lemieux, Dominique Linhardt, (...)
  • 2 See Bénatouïl (1999a: 2), and Lemieux (2007: 192).

2Interestingly, several aspects of action that we shortly introduced above are found in pragmatic sociology [sociologie pragmatique], the French fruitful research program initiated in the 1990’s by its leading figures, Luc Boltanski and Laurent Thévenot, which directly influenced a generation of young researchers.1 Pragmatic sociology will be our starting point not only because it is one of the most promising sociological schools which have emerged during the last decades, but also because it tries to reconcile a view that stresses the structural organization of the worlds of action with a view that emphasizes human agency and the formation of meaning in situation. Moreover, as will be seen, pragmatic sociology insists on the laypersons’ critical capacity to call into question the norms and values that are supposed to be carried out in a given course of action.2 In short, pragmatic sociology, which is also called “sociology of critique,” tries to leave aside the distant standpoint of critical sociology, such as that of Pierre Bourdieu, which entails the well-known epistemological break with commonsense and thereby the pre-dominance of the view of the sociologist over that of the social actor. Instead, pragmatic sociology tries to bring to light the “capacity to judge” and to criticize that ordinary actors themselves possess. Importantly, such capacity of critique is endogenous but also plural: whereas critical sociology aims at unveiling domination as if it were just one single, monolithic order, agents navigate plural orders of worth and hence have at their disposal different kinds of resources for criticism. So critique does not depend upon a unique operation of critical totalization, exerted from the external standpoint of the social scientist. Each order of worth is vulnerable to its own internal critique and to the constraints of justification that go with it.

  • 3 For notable exceptions, see Claverie 1994, 1998, and Thévenot 2011.
  • 4 The model of Thévenot goes in this direction, though.

3Obviously, the attention, in pragmatic sociology, to the capacity of ordinary people to resort to critical judgment and to be concerned with the justification for the common good that is at stake in a given world of action, is in phase with the insistence of philosophical pragmatism on the role of doubt, inquiry and experience in human activity and, above all, in modern democracy – even if effective references to this philosophy are very scarce.3 But despite this strong family resemblance, pragmatic sociology seems to remain at the threshold of philosophical pragmatism, mainly with regard to the issue of the making of a public and the status of public inquiry. Indeed, this latter emphasizes the moral and political necessity, for scientists, politicians and ordinary citizens, to take into account the indirect consequences of their own activities on others, and to allow those who are affected by these very consequences to launch a public inquiry. Philosophical pragmatism is thus underlain by a political model of the community as a whole whose normative dimension is central: only a demanding participative stance can allow individuals to turn their private trouble into a public problem and to enrich both the individual and the community. But pragmatic sociology, somewhat paradoxically for a model which greatly values justification, insists so much on the variety of the worlds of action and their internal worths and legitimacy that it tends to leave aside the very idea of a public sphere in which a political “meta-inquiry” into the validity of the orders of worth, their mutual relations and their hierarchy is carried out.4 It is maybe this absence of explicit normative stance and this lack of “meta” standpoint that explains why Luc Boltanski, in his recent book On Critique ([2009] 2011), partially repudiates pragmatic sociology, which he initiated himself, and returns to his original fascination for Bourdieu’s critical sociology.

4Our paper aims at showing that such renunciation is not necessary and that pragmatic sociology, if it really takes seriously the legacy of pragmatism, can fully reconcile a pragmatic and a pragmatist way of doing sociology. Under the auspices of philosophical pragmatism, indeed, critical inquiry can be held without reviving the external totalizing point of view of the critical sociologist. As we will see, philosophers such as Dewey provide the normative tools necessary to value the social situations triggering the formation of a public, that is, the formation of a “community of those indirectly affected” which succeeds in becoming a “community of investigators.” Such tools allow distinguishing “public-triggering” configurations from “publicidal” configurations that eclipse the “bottom-up” and publicizing movement that should enable ordinary people to make sense of the systemic interdependencies they are entangled in (Dayan 2001).

5To tackle the issue of critical inquiry, our investigation will make a detour via two French authors who exerted a great influence on the social scientists participating in the pragmatic turn. Thus Jeanne Favret-Saada’s work on contemporary witchcraft, and Michel de Certeau’s study on 17th Century possessions will prove astonishingly similar to, and compatible with, the perspectives advanced by pragmatist philosophers, especially John Dewey, about experience, action, and the constitution of the public. This detour will allow us to show that pragmatic sociology as such can be fully critical and that the revival of the epistemological break is not a necessary step towards political awareness. But before following the marvelous inquiries proposed by Favret-Saada and Certeau, we need to give a better account of pragmatic sociology.

I. Between Action, Pragmatics and Pragmatism

  • 5 Those social scientists came from different horizons, but converged on an editorial project, starti (...)

6By the late 1980’s, some French social scientists influenced by analytical philosophy, phenomenology, pragmatics, and ethnomethodology tried to rethink the relationship between the individual and the social structure in an altogether different manner.5 Instead of being an overhanging structure, the social order is said to be the practical accomplishment of ordinary agents. While being roughly along the same lines, the ground-breaking model of “economies of worth,” published in the 1990’s, which is the launching pad of pragmatic sociology, is particularly interesting. Not only does it draw from linguistic pragmatics but it also enters into resonance, in some respects, with the spirit of philosophical pragmatism.

7Indeed, linguistic pragmatics has emphasized the fact that language, even when it is supposedly used to describe a state of affairs or to merely exchange information, is always a way of doing things (Austin [1962] 1975). Language is not a transparent representation of the world: it aims at modifying the world, not at giving way to it. Although always context-dependent, the use of linguistic utterances is nevertheless governed by constitutive rules, which establish the order of words (e.g. we cannot say “I my vegetables eat”), a system of places (e.g. “to give” involves, by definition, three positions, the given object, the giver, and the receiver), and a public, impersonal set of rights and obligations (e.g. if I promise to come to your party, then I have the obligation to come to your party). In short, pragmatics articulates syntaxical structures and constitutive rules necessary for linguistic utterances to be intelligible with the context-dependence of enunciation that characterizes language-in-action.

  • 6 Latour himself defines his sociology as being “half Garfinkel and half Greimas” (2005: 54).

8In their theory of “economies of worth,” also called “theory of justification,” Boltanski and Thévenot do take inspiration from pragmatics: they aim indeed at laying out, in a grammatical form, the plural orders of worth (civic, commercial, and so on), the principles of evaluation and justification (equality, productivity, and so on), and the kinds of commitment (familiarity, justification, and so on) that allow the closure of what can be envisaged or, above all, argued in a given situation (Boltanski & Thévenot [1991] 2006; Boltanski 1990; Thévenot 1990). Just like linguistic grammar enables speakers to create and understand an unlimited number of utterances of their language, a “grammar-like” system of norms, modes of engagement and principles of actions enables agents to act appropriately in particular situations they have never encountered before. For instance, normally competent agents master the rules that a public denunciation of injustice must satisfy to be deemed relevant: such denunciation entails necessarily four actantial roles, that is, the ‘victim,’ the ‘prosecutor,’ the ‘denouncer’ and the ‘judge’ – an actorial structure that reveals the influence of the works of Algirdas Greimas and Bruno Latour on pragmatic sociologists (Boltanski 1993).6 Moreover, to accomplish a well-formed act of denunciation, the plaintiff must relate himself in a credible way to a collective susceptible to support his version of the facts and to share his indignation (e.g. a civil association) (Boltanski et al. 1984).

  • 7 See also Bénatouïl 1999b, and Boltanski (1999 [1993]).

9So pragmatic sociology clearly draws from pragmatics and semiotics to highlight the ordered set of rules, the actantial relationships, and the context-dependent spectrum of possibilities for acting that constitute and map out ordinary modes of action and justification. But it is also in phase, although more tenuously, with philosophical pragmatism when it insists on the dynamics and indetermination of action. Far from being caught in a unified, highly integrated cultural and social system, people navigate plural and distinctive action frames, made of situational constraints, material arrangements, and above all, collective norms of qualification. Pragmatic sociology then focuses on the situated way people agree over a frame of reference, take hold of their environment, material as well as symbolic, and adjust their mode of engagement within the situation. In this indeterminate and pluralist view of agency, action depends on the full range of competences that persons are inherently endowed with, including the cognitive and moral ones necessary for critique and distanciation. Far from being the ontological dopes, submitted to the overwhelming forces of the social order, that structuralist social sciences have portrayed them to be, social actors are thus seen as competent and critical subjects who are able to reach agreement about the “action-that-is-suitable” and the corresponding “grammar” of values, categories and beings that go with it (Thévenot 1990).7 In case of disagreement or dispute, the interactants are able to switch from a familiar, unquestioned regime of coordination and communication to a reflexive regime of conflict resolution and argumentation – the so-called “regime of justification” – which allows them to reach a new social agreement (Boltanski & Thévenot 1999).

  • 8 About this primacy, see Frega & Carreira da Silva 2011.
  • 9 To our knowledge, Putnam is not quoted by pragmatic sociologists such as Thévenot or Boltanski; but (...)
  • 10 On this resistance, inspired by the work of Mead, see Franks & Seeburger 1980.

10Last but not least, the epistemological primacy of practices and the emphasis on human agency that characterize pragmatism also led sociological pragmatism to call into question, at least implicitly, the boundaries between knowledge and action, fact and value.8 Since conceptual schemes give intelligibility to all our epistemic and practical relations with the external world, direct knowledge of “what there is” turns out to be impossible: facts are only knowable through a system of representations and practices that determines the relevant level of their individuation and description. Similarly to the second Wittgenstein, who systematized insights very similar to those of Dewey or Peirce, pragmatism argues that it is impossible to escape from the legislation of language and to go outside our forms of life, governed by pragmatic rules of action and discourse. For a neo-pragmatist such as Hilary Putnam (1981), even a basic factual inquiry into the number of objects that are on the table in front of me cannot lead to one single truth; indeed, to describe how many objects are on the table, we have to determine first what counts as an object: the book as a whole or the pages that compose it? My pen or its constitutive parts?9 In other words, pragmatism goes necessarily with a kind of “internal realism” that recognizes that the world can only be described “from within” a common system of representations and practices. It is this very system that enables us to determine what counts as a valuable candidate for truth-and-falsity, relevance-and-irrelevance, or usefulness-and-uselessness (Putnam 1981). This being so, such internal or pragmatic realism does not deny, contrary to some relativist or constructivist approaches, the reality of external facts; the real world does causally contribute to our perception and action – otherwise we could not distinguish between what is the case and what seems to be the case. Moreover, practical dealings with the world necessarily obey to a reality principle: practical reasoning involves by definition the functional adaptation to real circumstances and the anticipation of the consequences of one’s own actions (Anscombe 1957; Ricœur 1977). The “obdurate resistance” and partial unpredictability that the world offers to the ordinary investigations of our surroundings do serve as strong reality tests.10 Still, for internal realism, an external fact is not a “thing-in-itself”; as Putnam puts it, it is endowed with an “objectivity-for-us” and depends, as such, on the conceptual frameworks which indicate us how we should qualify it and what we should do with it.

Grammar, Phenomenology, and the Difficult Status of Critique

11This pragmatic framework has several interesting consequences for pragmatic sociology and, more generally, for social sciences. From a methodological point of view, only a fine-grained ethnographical approach can account for the concrete, practical adjustments, improvisations, micro-inquiries and critical disagreements that characterize the pluralist way persons deal with the world around them (Breviglieri & Stavo-Debauge 1999). Moreover, such a fine-grained approach reconciles a grammatical focus on the acceptable structures of action and discourse, which can be mapped out and modeled in a systematic way, with a phenomenological focus on experience. From this perspective, indeed, experience appears as a kind of two-sided entity. One of its sides is objective: it refers to the typical tests that pertain to such or such order of worth (e.g. art, industry, family) and that anyone has to experience and pass to be recognized as a normal, competent actor (e.g. creating an original work of art, being a good father, making a profit, and so on). Although typical, and “grammatically” expected in the different kinds of situations that people are deemed to encounter, such tests remain nevertheless pervious to the particularity of the course of action in which they occur and sensitive to the singularity of the persons that they are supposed to assess. The other side of experience is subjective: it refers to the plural ways people feel, experience, appraise, suffer, in short engage in, and are affected by, a given situation. Those “existential tests,” as Boltanski (2011 [2009]: 107) puts it, refer to what provokes suffering and to what affects, such as the experience of injustice or humiliation brought about by the contempt of those in position of power or the experience of emancipation created by rule transgression. In other words, on its objective side, experience is related to the fact of experimenting, of testing – and being tested by – a grammatical configuration, whereas, on its subjective side, experience is related to the phenomenological fact of experiencing “what it is like” to be in such or such grammatical configuration.

  • 11 On these two kinds of metaphysics, see Strawson 1992.

12This pragmatic framework has also interesting consequences for social sciences from an epistemological and political point of view. It means indeed that social sciences have to relinquish the external point of view of the critical sociologist, whose “revisionary metaphysics” tries to correct the ordinary world-view that would supposedly mislead agents, mainly by hiding the overall structure of domination that would remain, as such, out of their reach (Bourdieu). Instead of establishing a strong asymmetry between social scientists and those taking action, pragmatic sociology adopts a “descriptive metaphysics” that takes seriously the point of view “from within” of agents and follows the possibilities of critique they actualize in the disputing activity that arises when a joint action goes wrong.11

13Although very heuristic, this perspective raises two important issues. First, pragmatic sociology focuses on the critical moments, from domestic quarrels to long-term judicial litigations, in which agents call into question the implicit order of worth they were used to uphold (Boltanski & Thévenot 1999). In so doing, this approach tends to overemphasize the situations of dispute and justification in which participants transform a disagreement into a reflexive object of public inquiry and become thereby a “community of investigators,” as Dewey would put it. Boltanski recently made more explicit the similarities of such an approach with pragmatism: “We can therefore more or less link to the spirit of pragmatism the way in which the sociology of critique undertook to describe the social world as the scene of a trial, in the course of which actors in a situation of uncertainty proceed to investigations, record their interpretations of what happens in reports, establish qualifications and submit to tests” (2011 [2009]: 25). From this kind of “juridical” view of the social, agents and, by way of consequence, the sociologist who study them, are thus very busy producing the discursive, argumentative accounts that will re-establish a justified, legitimate agreement and will keep, then, violence away. And yet, grammatical constraints are not exclusive to modes of justification; they are also inherent to other forms of reciprocal actions, including those in which a kind of rule-governed, “legal” violence is “encapsulated.”

14The second issue raised by pragmatic sociology is its endogenous “challenge” and the underlying internal realism behind it: despite its attention to the internal resources for critique, such challenge might nevertheless render social critique difficult to sustain. Indeed pragmatic investigations avoid resorting to “beings” and values, such as class membership, social forces or symbolic violence, that the actors themselves do not explicitly bring into play. The problem is that, at least in certain situations that we will discuss later on, social scientists have to leave aside the insider point of view of the participant to adopt an external, reflective point of view that highlights and normatively assesses the grammatical structures of interaction in which agents are entangled without even knowing it.

15It is precisely this issue that the recent book of Boltanski (2011 [2009]) addresses forcefully. Surprisingly enough, at least for scholars familiar with his previous works, Boltanski proposes a “pragmatic sociology of critique” that supports a dual approach of the social world: “pragmatic sociology” of critique focuses on “society,” that is, the “regimes of action” and the power relations, diverse, partial, local or transitory, that are in an immediate relationship with the preoccupations of actors and their insider point of view. By contrast, pragmatic “sociology of critique” focuses on the “social order,” that is, the world apprehended from the external point of view of the critical sociologist, who must shed light on the overall, monolithic structure of domination that underlies the so-called “society” and remains invisible to the agent’s eyes. If we follow Boltanski, assuming an external standpoint that breaks with the “objectivity-for-us” of the world seen “from within” is imperative: only such standpoint can go beyond plural and superficial forms of power, which are readily observable, to unmask the profound, enduring asymmetries which, while assuming different forms in different contexts, are constantly duplicated to the point of colonizing reality as a whole. So whereas critique refers to the socially rooted, contextual forms of criticism to which ordinary agents and standard sociologists have access to, “metacritique” refers to the theoretical constructions that aim to unmask, in their most general, systematic dimensions, oppression, exploitation or domination – a domination which occurs in the semantic determination of “what there is” and in the normative qualifications and categorizations of beings.

  • 12 On this issue, see Kaufmann 2012, Stavo-Debauge 2011.

16As interesting as it might be, Boltanski’s framework and lexicon has a fatal weakness; it is strikingly dualist. Indeed, On Critique depicts a “double bottom” society: beyond the surface power relationships and insubstantial collectives that constitute the so-called “society,” there is the “social order” of domination, sort of big semantic Leviathan which structures and maintains, via a “top-down,” continuous process of totalization, the established order of beings.12 Unfortunately, such conceptual dualism between society and social order and thereby between pragmatics and semantics tends to bypass plural grammars of actions and self-qualifications of agents to better shed light on the deep, one-dimensional, transversal meta-grammar that pits dominant elite against dominated people. The “Précis of critique” that Boltanski proposes is thus as anti-pluralistic as politically radical: a true emancipation can be only reached through a revolutionary movement, necessary for rendering “the reality of the reality” in which agents are immersed unacceptable (Boltanski 2008). Revolution is indeed necessary to overthrow the overarching semantic institutions, including the Law and the Welfare State, that make the dominated unworthy and allows always the same privileged ones to win (Boltanski 2011 [2009]).

17In some respects, Boltanskian dualism is surprisingly similar, though more political, to the one advocated by one of the founders of French sociology, Émile Durkheim himself, in his critique of pragmatism. Indeed, in his Sorbonne lectures, given from 1913 to 1914, Durkheim severely criticizes pragmatism: “I can accept neither the statement of the idealists, that in the beginning there is thought, nor that of the pragmatists, that in the beginning there is action” (1983 [1955]: 67). For the so-called “father of sociology,” the problem with pragmatism lies in the fact that it conceives truth and reality as a matter of individual experience and fails thereby “to recognize the duality that exists between the mentality which results from individual experiences and that which results from collective experiences.” Instead, Durkheim advocates a dual perspective that distinguishes high-level, collective representations from low-level, individual representations, and posits that “what is social [including truth, morality, and reason] possesses a higher dignity than what is individual” (1983 [1955]: 68). According to Durkheim, such founding dualism has two important consequences. From an epistemological point of view, it allows the constitution of two autonomous scientific disciplines: whereas individual representations and instincts are the object of psychology, if not biology, the “laws of collective ideation” are the object of sociology. From an ontological point of view, society is not only a constraining whole which is more than the sum of its individual parts; society is also an ennobling factor which adds a necessary and universal dimension to the individual basic, primitive ways of thinking and acting. Since, within Durkheimian dualism, truth, morality, and reason are exclusively social emergent properties, they can be grasped only if individual-dependent thoughts and representations are left aside. For Durkheim, the break with individual perceptions is thus a double requirement that pragmatism does not satisfy (Durkheim 1982 [1895]).

18As seen above, such break with individual perceptions is also at the heart of the metacritical or metapragmatical project of Boltanski. While building on the project of carrying out a sociology that does not give up on critique, this paper does not take up such dual view of the social. We will propose another pathway to social critique by taking more seriously the pragmatic assumption according to which the action and its consequences – and neither the agent nor the historical, social, or economical context – must be the unit of inquiry. We will argue that it would be preferable, for sociologists, to give up on an overhanging standpoint and to focus on the rules, both constraining and enabling, that constitute and structure the plural, multi-layered, and collective architecture of human actions. By unfolding the plural grammars that ordinary agents navigate and enact in the course of their daily life, sociological inquiry can foster a reflexive attitude that potentially increases their power of action. That is at least what we are going to argue in the following pages.

Bewitched by Social Practices

  • 13 See de Certeau 1985.
  • 14 Of course, for an historian, the task is particularly challenging, if not impossible. For M. de Cer (...)

19To address the link between grammar, phenomenology and social critique, we are going to get back to the pioneering works of two very important scholars, the anthropologist Jeanne Favret-Saada and the historian Michel de Certeau, who have initiated in many respects the grammatical investigation that pragmatic sociology advocates. Each in their own way, their outstanding research on witchcraft and possession, published in the 1970’s, underlines the “dark side” of grammatical constraints that force agents into a logic of interaction which might be fatal. Their approaches have several features in common. Both struggle against the classical anthropology and history “from above,” which postulates the asymmetrical hold of the official authorities and talkative elite over ordinary people – an asymmetry that the condescending stance of social scientists towards “informants,” wrongfully transformed into the passive objects of their intellectual discourse, happened to step up. Both approaches thus introduced a symmetrical epistemology, systematized and extended later on by Bruno Latour, which acknowledges all “have-nots” as deserving a place in history and anthropology. Finally, Favret-Saada as well as Certeau favor a phenomenological standpoint that tries to do justice to the “thickness” of bodily experiences, emotions, and practical intelligence enabling anyone, including the uneducated persons, to make sense of the world around them. In so doing, both show that ordinary people do not blindly take up the public “institutions of belief”;13 they do not necessarily believe in what their culture induces them to say. So if public utterances and institutional rituals definitely delimit the official domain of the “believable,” they do not necessarily reveal the scope and intensity of the actual “believing.” To account for the beliefs that actually affect people, even when these beliefs look as unbelievable as witchcraft, the ethnographer or the historian must imperatively give up on the objectifying gap that distances the all-knowing “discourse on the other” from the mute, supposedly ignorant body that bears it.14

20But to be in a position to better compare the implications of the research of de Certeau and Favret-Saada, we need to lay out the main features of the grammatical logics at work in witchcraft and possession that they have so brilliantly described.

II. Immersed in the Bocage

  • 15 In 1970s, a Normand landowner is always a man for reasons related to the procedures of property tra (...)

21According to Favret-Saada (2009), the language of witchcraft is recruited to make sense of the extraordinary repetition of unexpected and inexplicable misfortunes that overburden a landowner and his possessions, whether they be human (wife, children), animal (livestock) or material (goods, production rate) (e.g., bankruptcy, child’s disease, wife’s miscarriage, heifer’s death, engine’s failure, etc.). Faced with all these misfortunes, the victim feels powerless, all the more as the official authorities, incarnated by the priest and the doctor, are of no help to him:15 whereas the unruffled doctor resorts to natural causes and bad coincidences for explaining his misfortunes, the priest just invokes faraway, immaterial beings on which he has no hold (Favret-Saada 1977). Above all, for the victim, neither the doctor nor the priest are able to explain why those puzzling events happen to him in particular.

22Why this repetition and above all, why “me” and why “now,” wonders the person stricken by misfortune. But even if the victim starts suspecting that his ordeals could have an unnatural cause, he cannot initiate by himself an inquiry into his possible bewitchment without being taken for a half-wit. Such an inquiry must be initiated by an ‘annunciator, either a friend or a neighbor, who, by dint of witnessing this unlikely series of misfortunes, asks the victim: “Should there not be, by chance, someone who wishes you ill?” This question has incontestably an incredible performative power, that is, it performs an action in saying what is said: it indeed modifies the status of its addressee, converting him from an unlucky person into a possible “bewitched” and converting his misfortunes into spell effects. This question, which also sounds as a diagnosis of mental sanity – you are not a lunatic or a misfit but a bewitched – gives rise to a quest, which will lead the victim and his wife to search for an “unwitcher. Even though there are no symbolic guarantees that the unwitcher will be able to cancel the spell, the mere fact of attributing those misfortunes to an intentional cause, namely that of the “witch, is the first step in a long process of recovery. With the help of the unwitcher, indeed, the bewitched will hopefully retrieve his life energy by fighting back against the enemy who allegedly wants his destruction.

23In contrast to other studies on witchcraft, which mostly emphasize the details of the rituals, the exact wording of the phrases or the kinds of objects that are used to sustain it, Favret-Saada’s ethnography insists on the relational and actional aspect of witchcraft. To understand how really witchcraft works, it is not possible to draw upon the second-hand knowledge that anthropologists, psychiatrists, folklorists, journalists and official authorities produce contemptuously about it. As Favret-Saada forcefully shows, only first-hand knowledge or rather first-hand “grasp” allows to see, from “within,” that witchcraft is a war – a war of “deadly words” and daily struggles but, more fundamentally, a war of forces in which the bewitched, the unwitcher and the witch are caught in, whether they like it or not. But if witchcraft is a warfare, it is a very well organized, rule-governed one that strongly constrains those who are involved in it. In fact, witchcraft necessarily involves an actancial system, composed of four interdependent places: the ‘annunciator,’ the ‘bewitched,’ the ‘unwitcher’ and the ‘witch.’ The ethnographer intensely experienced herself that it is not possible to escape this system of places and the expectations tied to it: uttering a single word on witchcraft is already getting involved in a power relationship in which one’s interlocutor is trying to determine one’s proper place, to evaluate one’s force or weakness, to assess one’s benevolence or malevolence, and so on.

  • 16 Usually, Favret-Saada describes women in the position of the unwitcher. Therefore, we will use a fe (...)

24In our own words, this grammar-like system of witchcraft is governed by constitutive rules that define the scope of what can be uttered and done by whom at which moment. Only the annunciator can “interpellate” the victim and turn him into a bewitched; accordingly, it is not possible to enter into witchcraft as a bewitched simply upon one’s own will: it is a place ascribed by others. But once caught in the place that others have chosen for “me,” there is no way out – besides leaving the area. Of course, the most striking hetero-ascription that forces some to get involved in witchcraft is that of which the unfortunate alleged sorcerer is subjected to. Indeed, the witch is not in a position from which it is possible to speak in the first person: he is reduced to a “third person,” that is, a “non-person” that those who are part of the witchcraft crisis “speak about” but “never speak to” (Benveniste 1966): he is spoken without being able to speak by himself. There is no need of confession from the witch: he is not an audible voice but a function in the system (Favret-Saada 1980 [1977]: 24). As for the unwitcher, it is a powerful but also difficult place to take up, because only someone who is “strong enough” and ready to “take it all” on oneself can endorse it. What makes an unwitcher, hence, is not her knowledge but her force16 – an enabling, “good,” “positive” force that she is nevertheless ready to use for rendering evil for evil in her fight against the witch.

25More generally, this preexisting system of places is polarized between those who possess “the force” (the witch, the unwitcher) and those who are deprived of it (the bewitched) (Favret-Saada 2011). Insofar as this force can be at work in the slightest utterance uttered by one of the warring factions, decoding the meaning of what is said is far less important than understanding who is speaking to whom. Just as the whole person of the witch embodies a negative force that turns his most trivial gestures into deadly attacks, it is the entire being of the unwitcher that enters into action by way of the words that she utters during the cure. In short, the contents of the words themselves count less than the very fact of enunciation and designation that allows the third-party intervention of the unwitcher to deviate the witch’s spells from his victim and to draw them towards an opponent worthy of himself.

Performing a New Social Contract?

26Favret-Saada’s research on witchcraft is fundamentally pragmatic, in the two senses we have unfolded above. First, it is pragmatic in the sense that it reveals the performative force of the language in action, which manifests itself in the inaugural act of instantiating the rule-governed system of witchcraft that the annunciator’s suggestion accomplishes (e.g., “Is there someone who wishes you ill?”). This inaugural act, which is a first step towards the normalization of the misfortune, starts the progressive restructuration and control of the unlimited, malevolent performative force of the sorcerer. In other words, two conflicting performative forces, of different kinds, are present in witchcraft. The performative force of the evil encapsulated into the deadly communicative acts of the witch is totally unbridled, immediate and unexpected, bursting from nowhere and without any reason. Put in our own words, such brute performative force constitutes a massive process of desinstitutionalization.

27Indeed, Favret-Saada marvelously shows in her recent book, Désorceler (more explanatory and social-centered than Deadly Words), that the alleged witch is a kind of free rider, who gets rich at the expense of others, monopolizes the goods of the others, and shows himself impervious to social and moral rules (Favret-Saada 2009, 2011). The witch shakes the common world that the bewitched shares with his fellow creatures and jeopardizes the social equilibrium by refusing the minimal symmetry of conditions necessary to hold society together (Favret-Saada 2011). The witch also jeopardizes social relationships at large since the very possibility of his existence creates a climate of mutual suspicion and mistrust: insofar as he is endowed with the power to cast a spell on anyone he wants to despoil or destroy, anything can happen at any time, anyone who speaks to you can prove fatal for you.

28From this perspective, witchcraft can be seen as the symbolic reinstitutionalization of the basics of the social contract, unduly jeopardized by the naked force of the witch but also by the vulnerability and weakness of the bewitched (Favret-Saada 2011). The unwitcher breaks off the direct and unequal confrontation between the culprit and his victim by imposing herself as the inescapable figure of the Third: her force is regulated and mediated, allowing the resocialization of the negative, unregulated power of the witch by forcing it into a rule-governed system of preestablished places. Because the unnamable is absolute and asocial, one of the most important acts of micro-institutionalization that witchcraft performs is naming. Indeed, the apparently boundless force possessed by a witch is a sort of black hole whose existence can be only inferred from its destructive effects. “Naming the witch is an attempt to enclose within a figure something which, in itself, escapes from figuration” (Favret-Saada 1980 [1977]: 74). The act of naming someone from the circle of acquaintances, carefully prepared during the unwitchment sessions, forces the witch into a system of names that turns him into a singular, recognizable individual, accountable, as such, for the harm he has done.

29So witchcraft progressively restores the social equilibrium by working as a system of communicating vessels that draws the force from the witch and awakes the force, if not the violence, of the victim. Indeed, the vulnerability of the bewitcher is due to his difficulty to take on the legal, institutional violence that his social function of landowner involves, from the dispossession of siblings to the exploitation of his wife to the competition with neighbors. Witchcraft therapy is a pedagogical undertaking that awakes his aptitude for violence while keeping the appearance of a right, justified and necessary fight between the principles of the Good against those of the Evil (Favret-Saada 2009). Thus what the unwitcher Mrs. Flora brings into play in her therapy sessions is not only the salvation of the singular person of the bewitched; it is more fundamentally the “ethical order of the world”: the bewitched is entrusted with the upholding of social and moral principles that transcend his singularity.

30If witchcraft is pragmatic in the sense of the performative force of enunciation, which has the power to do what it says, it is also pragmatic in a second sense: the partial indeterminacy of the course of action and the phenomenological intensity of the experience of those who are “caught in.” Actually, as will be seen below, individuals portrayed by Favret-Saada are not re-enacting some learned scripts or actions; they are affected by the way things appear to them in an unpredictable or uncontrollable fashion.

Believe It or Not: From Grammar to Phenomenology

31Favret-Saada (2009) describes the grammar-like system of witchcraft basically as an essential therapeutic device operating at the margins of public knowledge and institutional beliefs. By granting the victim a central place in an actancial system where what happens to him seems to make sense, and proposing him an intelligible narrative, witchcraft restores the unfortunate to subjecthood. To use our own words, witchcraft is a dialogic process of empowerment that opens up, for the victim, a hopeful, enabling but also constraining, field of action. He is no longer the patient of a series of untoward happenings but an actor who can and must fight back the obscure, evil forces of the alleged witch (Favret-Saada 2009). According to the anthropologist, the series of action that the unwitcher Mrs. Flora prescribes the bewitched to accomplish are like so many steps in a ritual whose real efficiency lies in its capacity of forcing the patient to leave his position of passive victim.

  • 17 One finds the same idea in the work of the anthropologist Albert Piette (2003).
  • 18 On the “conducts without beliefs,” see the well-known paper of Paul Veyne (1988).

32Interestingly, this process of resubjectivation does not require from the bewitched to believe in witchcraft, which remains rationally unbelievable for everyone, including those who are caught in it against their own will. Rather such process requires from the bewitched to suspend disbelief and, above all, to be ready to do anything that could bring his ills to an end. This requirement is not epistemic but pragmatic: what matters is to do something. As philosophical pragmatists have pointed out, belief, unlike knowledge, is fundamentally a propensity and a power to act. When one has “the death at one’s heels,” Favret-Saada says, one cannot afford to launch an epistemic inquiry, to make a cultural fuss or to search for symbolic guarantees: only the result counts, in this case the end of hardship (Favret-Saada 2011). From this perspective, belief in witchcraft, if any, is not a stable state of mind that one possesses and entertains in one’s inner world. On the one hand, indeed, people, including uneducated peasants, have the cognitive flexibility to believe and not to believe at the same time, to navigate a mobile and ambivalent world of shifting realities, in short to envisage, for the temporary sake of the situation, that bewitchment is not an impossible hypothesis. On the other hand, belief is not a proposition in the head: instead it is a way of being concerned with, or affected by, which remains vague enough to pass reality tests but involving enough to restore life energy.17 So as Favret-Saada powerfully suggests it, if the conducts adopted within the witchcraft world might possibly be “without beliefs,” they are incontestably with affects.18

33Drawing from the work of Favret-Saada, one can surmise that what one could call “pragmatic beliefs, if one can still speak of belief – which is far from obvious for this author – do not involve any epistemic stance: practical commitment suffices to lead temporarily the “affected” to leave behind the ordinary way of life to enter a world in which the insufferable can be turned into a series of words and actions. Inside this world of actions, present and future, the strange beings which are part of the witchcraft arsenal tend to be endorsed with an “objectivity-for-us” that escapes from scrutiny and ordinary reality tests. Importantly, the “objectivity-for-us” of witchcraft is neither the object of a “referential contract” that would place it in the empirical order of things that really exist, nor the object of a “fictional contract” that would defuse its empirical implications by specifying “it is only a story.” The “objectivity-for-us” of witchcraft is rather the object of a “deferential contract” that allows participants to “validate on credit” the strange creatures, such as witches, bewitched, evil, etc., which do not satisfy the usual demands of reference – validation that they accomplish by referring not to a state of affairs but to the belief of others (Kaufmann 2006). Of course, in the world of witchcraft that Favret-Saada describes, the actual “others” who guarantee the holding of pragmatic beliefs are more than scarce since they are mainly the annunciator and, in a second step, the unwitcher. But as Michel de Certeau (1981) points out in his comments on Favret-Saada’s research, the peasants of the Bocage do back the “indefinite plural” of those others who might believe in witchcraft: “there are people who believe in it,” the interviewees say, the alleged belief of indeterminate others vouching for the conceivable, if not the believable, in spite of its lack of institutional support and public maintenance.

34If witchcraft in the France of the 1970’s is deprived of institutional support and public maintenance, this is obviously not the case in the France of the 1630’s: the series of demonic possession that Michel de Certeau dwells on is characterized by the support of the centralized religious authority, at least at the beginning of the “case,” as well as by its public reach. As will be seen, comparing a crisis which lies under the seal of secrecy and a crisis which gives rise to a public dramaturgy raises differently the issue of the inquiry, social and scientific, that cultural practices can trigger. So after addressing Favret-Saada’s remarkable fieldwork on witchcraft among Normand farmers, let us consider Michel de Certeau’s magnificent study, written a few years before Deadly Words, about a crisis of possessions that took place, in the south of France, during a century plagued with wars of religion.

III. When Evil Becomes Public: The Torments of Possession

35The Possession at Loudun (2000 [1970]) narrates in a subtle way the contest that occurs between exorcists, physicians, and the king’s representatives as they try to characterize and put an end to the demonic deeds that bedevil an Ursuline convent from 1632 to 1638. In a chapter published subsequently, “Language altered: the sorcerer’s speech,” the author briefly recapitulates the perspective he adopted in his historical monograph: “Loudun is successively a metonymy and a metaphor allowing us to apprehend how a state policy [une “raison d’État”], a new rationality, replaces a religious reason” (Certeau 1988b [1975]: 246).

36To account for this historical shift in the relation between the sacred and the profane, religion and science, Certeau adopts two distinct but correlated and complementary perspectives. The first perspective tries to apprehend the diabolical spectacle as a social phenomenon. Indeed, the historian proceeds through a close examination of the social positions located in the prevailing fields – mainly the religious, the medical and the political ones – which work, sometimes agonistically, within the Loudun society during the possession crisis. Such careful examination allows Certeau to infer, from the alterations happening in the different logics at work, the shifts in the collective imagination of the time. The second perspective focuses on the discourse spoken by the possessed women – or, perhaps better, the discourse being spoken through them – with two complementary lenses, one grammatical, and the other phenomenological. The grammatical investigation aims at uncovering the actancial systems and the places of enunciation that constitute social activities such as exorcism. As for the phenomenological investigation, it digs up the experience of the possessed, deeply buried in, even under, the archives that document it. Reaching deeper into the pre-discursive strata of the subject’s experience, the analysis tries to account for an experience that shakes the system of social positions by playing with the assigned places of enunciation. In Certeau’s conceptuality, indeed, the subversion of discursive reason, whether theological, medical, or legal, takes place as the nuns’ enunciation [énonciation] plays with, and within, the interstices inherent to any social system of statements [énoncés].

37Our discussion of Certeau’s study will follow those two moments of analysis: the first moment focuses on the way the social fields advance diverging claims to adjudicate the crises of possessions whereas the second moment focuses on the grammatical and phenomenological aspects of those same possessions.

The Theater of the Possessed as a Social Phenomenon

  • 19 Cases are documented in Denmark, England, Germany, Netherlands, Switzerland, and, for France, mostl (...)
  • 20 See Certeau’s discussion of witchcraft (2005).

38According to Certeau, possession cases and witchcraft trials are both diabolical manifestations occurring contemporaneously in post-Reformation Europe, but differ in some important regards. Though waves of witchcraft trials spread across northern countries between 1570 and 1685, they remain relatively rare in the South.19 On the other hand, possessions are a southern phenomenon that stretches from 1559 to 1663, their typical form being the well-documented Gaufridy trial that took place in Aix-en-Provence (1609-11) and provided the plot for the events of Loudun. Furthermore, witchcraft happens in rural settings and has a distinctive binary structure that pits a sorcerer against urban judges. In that context, witchery discourse works as a way to frame and conceal a fight among protagonists of asymmetric social statuses and thus secret trials functions as an effective procedure for lettered elites to contain and crush popular unrest.20 Conversely, possession cases share a ternary structure, with the public attention focusing on the victim (the possessed), and not on the judge or the sorcerer. Contrary to witchcraft trials, those cases lead to an overt confrontation where central participants share a similar social status and urban setting.

39Since the Reformation and until 1632, Loudun has been a Protestant vanguard in a predominantly Catholic territory. The arrival in town of diverse Catholic religious orders protected by the king (Jesuits, Discalced Carmelites, etc.), since 1606, and the creation of convents indicate that the Counter-Reformation is on its way. Possessions have played a great part in this process, for the Catholic confrontation with the devils is a very effective way of confirming, in a supernatural manner, which of the contending persuasions is the true Christian faith. Yet, religious divisions are losing their power to define the line of confrontation between the parties. More and more, the dividing line will pass among defenders of, and opponents to, local privileges, threatened by the royal prerogatives of the centralized authority. Religious truth is losing ground, as reason [raison] and right are becoming an attribute of the State.

  • 21 “The discourse of possession turns on an absent figure whom it gradually renders more precise: the (...)

40The Loudun possessions start by the end of September 1632 within a context fraught with tension, as a plague episode that wiped out 3,700 of the 14,000 inhabitants is about to end—a tragic reminder of the epidemic which already afflicted the city in 1603. Interestingly, demonic afflictions are said to propagate just as the Black Death is thought then to disseminate, that is, by way of smell. Thus, the nuns present to their exorcists three thorns from a hawthorn and a bouquet of musk roses as a proof of the sorcerer’s enchanted misdeeds. The enchantment supposedly works through a scent that captured and obsessed the Ursulines in a supernatural manner. The enchanter – a sophisticated and handsome priest by the name of Urbain Grandier – is gradually designated, during the exorcisms, by the demons speaking through the possessed.21

41At Loudun, possessions are an utterly serious matter, for they will determine which reason – religious, scientific or political – will prevail against the devils that take hold and agitate the nuns. And in this confrontation, publicity plays a crucial role. As long as the demonic deeds remain enclosed within the convent, in the prioress’ room, the exorcists’ religious language provides an indisputable description of what is happening. But as soon as the phenomenon enters the public place, it becomes a spectacle that everybody duly attends according to his or her rank. The whole society, starting with aristocracy and bourgeoisie, come to see and to be seen, in an increasing publicity that has a corroding effect on the credibility of the exorcism, turning the combat with the demons into a disputable matter.

  • 22 Credibility is a key concept in Certeau’s thinking that has to do with “the machinery of representa (...)

42As possession cases go on, “His Majesty’s” superintendent is sent to settle the possession affair by lending the clergy a hand in order to secure the sturdiness of the social order. Of course, this order is sustained by a certain kind of public credibility, always entangled with credulities, which draws the boundaries of what is considered credible, trustworthy, and shareable in the whole society.22 As the old organization of certainties breaks apart, social critique steps into the breach. The king’s intendant tries to keep up appearances by ordering the magistrates sitting in court to regularly attend mass, so that they can adore the Holy Sacrament and listen to the exorcists’ homilies. Even so, rumours continue to spread. To silence the skeptical voices and put an end to the doubts taking over the population, the superintendent forbids, with a banning proclaimed and displayed on the streets, any malicious gossip against the afflicted nuns or their exorcists. But the placards will not last for long: their tearing up illustrates the kind of adherence they will meet among the public.

43Admittedly, the libertine priest accused of sorcery, Urbain Grandier, will be tried, tortured and burned at the stake, and the royal order of things, apparently still based upon religious foundations, restated. Still, the public, pluralistic turmoil that shakes Loudun bears witness to the decreasing legitimacy and credibility of an order on the point of vanishing: that of a religion that used to unify the experience of being in the world and provided a homogenous worldview to a unified political community. In spite of appearances, the Loudun crisis has thus resulted in a new distribution of powers and prerogatives: politics will grant its unity to a society more and more pluralistic; science will administer a natural truth; and religion will be left with the “spiritual,” the supernatural, that area, in a secular world, which stretches on the margins of human affairs (Certeau 1987 [1969]).

44Loudun is a theatre where the possessed bodies are publicly exposed and where a public competition, discursive and practical, takes place for getting them back in the grip of normality. On the Theater of the Possessed, divergent claims to administer the truth and to provide a foundation for the common ground on which society stands are made visible, discussed and contested. The confrontations about the naming, by men of power, of what is really happening to those possessed women, reflect what is happening at the level of collective representations: the troubles that bodily affect the nuns somatize the disturbances that run through the whole social body. Of course, the proliferation of dissonant discourses from the main actors of this theatrical drama, mainly the Catholic Church, the royal court, law courts, academies, medical schools, the Ursulines and the Jesuites, mostly affects the different fields, mainly religious, medical, and political, that they are supposed to represent and incarnate. Even if deviltries will be the transitional solutions to the erosion of certainties, the public spectacle of the overt dissent that divides the authorities acts in a transformative manner on the social structure.

45If the long-lasting crisis of possession in Loudun has definitely transformed the social positions and the links of interdependence between the different fields that sustained the ancient social order, it has also transformed the semiotic order that upheld it. If we follow Certeau, the relevant locus to investigate the transformative character of discourse is not only at the level of the social organization; it is also at the level of action and experience, whether they be individual or collective, as will be shown in the following pages.

Diabolical Enunciation: The Phenomenology Beneath the Grammars of Possession

  • 23 For a similar approach in sociology deeply influenced by ethnomethodology, see the work of Jean Wid (...)
  • 24 Certeau does not systematize nor make an important use of the concept of grammar. Nonetheless our u (...)

46Certeau’s pragmatic and phenomenological conceptions of discourse, especially in his paper “The Sorcerer’s Speech” (1988b [1975]), merge into his characteristic enunciative approach, which is influenced in particular by Lacanian psychoanalysis and Benveniste’s linguistics of enunciation. In Certalian conceptuality, discourse is a public action that depends and acts in a transformative manner on the social structure.23 Importantly, discourse is always enunciated from a place [lieu], which in turn coordinates two dimensions closely related to one another, even though they obey in part their own logic: the first dimension is social, referring to a location in society; the other is semiotic, referring to a position in discourse, that is, an enunciative position. Though Certeau’s concepts are extremely efficient in grasping the intricate links that relate, impact and transform both local interactions and social structures by way of discourse, they remain relatively vague. While drawing inspiration from this author’s insights, we will use more specific conceptual tools, mainly the concept of grammar, in order to approach the power of discourse to establish and to transform enunciative positions and, more generally, the social order.24

  • 25 As Certeau (1981) reflects on trust and credibility, he will resort on the etymological properties (...)

47Indeed, the concept of grammar allows to better specify the transformative power of discourse over the social order. As it can be seen, the selection of the relevant grammar, either religious, medical, or political, supposed to define the crisis of possession, has serious consequences for the organization of the social order and, potentially, for society as a whole. In particular, the competition between the religious and the medical grammars, which try both to make sense of the experience of the possessed, is a major social and political issue. For exorcism is one of the privileged grammars of the religious field, even if it starts to be called into question by some prominent priests, whereas medical diagnosis is the central grammar of the growing reason-oriented field of science. Of course, the victory of one of these grammars over its rivals positively affects the social positions of those who are particularly concerned with its enactment. Inversely, the lack of currency – the loss of credibility25 – of a given grammar has downgrading effects on the positions closely related to it, as shown by the erosion of the religious language during the Loudun events and the closing supremacy of the political reason over its competitors. But to understand the transforming impact, at the social level, of the selection of a given grammar from among the available grammatical repertoire of the time, we need to dwell more precisely on the normative and semiotic work of grammar.

48If we extrapolate from Certeau’s analysis, the order of action and experience has, at least in part, a grammatical status: it constitutes both a system of rule-governed participation and a way of symbolization or, stated differently, a way to represent and make sense of what is happening in a given context and, further, in the social body. A grammar operates as an actancial system, which defines a system of places (e.g. exorcist-possessed, doctor-patient, and so on), opens up a field of action and delimits the enunciative positions that those who are involved in it can endorse. For instance, medical diagnosis, which is particularly threatening for the authority of the priests in charge of curing the possessed, entails two complementary, structural places of enunciation, that is, the doctor and the patient’s body. By contrast, exorcism entails four places of enunciation, namely the exorcist, the sorcerer, the victim, and the evil spirit. Thus, the exorcism that a priest performs on an afflicted person leads to count this latter as a possessed, whereas the same person treated by a doctor will count as a patient suffering from a disease. Of course, those different systems of places render possible different kinds of activities. When the exorcist names and addresses the evil spirit speaking through the possessed, he is awaiting from the demon to confirm that the suggested name was a correct guess – confirmation that would grant the priest a binding authority over the demonic entity. The words uttered by the afflicted nun, as a possessed body, are central to the unfolding of the exorcism. On the other hand, the nun’s speech, even demonized, is unnecessary, if not unwanted, during a medical encounter with a 17th century physician: the symptoms affecting the patient’s corporal surface (e.g., sweat, pulse, etc.) are readable as a proof not of the presence of a supernatural entity but of a natural illness.

49We arrive then at the heart of possession and are able now to elucidate how that phenomenon weaves together the social, grammatical, but also phenomenological dimensions. As mentioned briefly, a central aspect of exorcism has to do with naming: “exorcism is essentially an enterprise of denomination intended to reclassify a protean uncanniness within an established language” (Certeau 1988b [1975]: 255-6). The nun’s identity is altered by a subterranean affliction that needs to be expressed in a shared language to find its etiology and cure. This process of symbolizing, deeply inspired by Certeau’s psychoanalytical conceptuality, is a way of stating what is evil and reinstating the afflicted person within an official grammar and, more generally, within the social order – a social order that is thereby reestablished: “naming simultaneously posits a linkage and a place. It functions at once as participation in a system and access to the symbolic” (Certeau 1988b [1975]: 262). Interestingly, both exorcists and physicians, in spite of the competing, exclusive explanations that they advocate, rely on the same naming procedure to tame the (d)evil that lurks within the afflicted woman. As Certeau puts it, doctors and exorcists agree enough to “eliminate an extra-territoriality of language” and to ascribe the possessed nuns to a determinate place, simultaneously a place of enunciation and a position within a social order. “What they are fighting through acts of naming is a text-off, a writing of otherness, where the possessed woman is located when she presents herself as the statement of something that is fundamentally other” (Certeau 1988b [1975]: 247).

  • 26 Certeau draws those postulates from the relationship between the modern psychiatrist and the insane (...)

50This “verbal imperialism” leaves very few opportunities of resisting to the possessed, who have locked themselves into the tautological, “magical circle” of possession. Still, the hetero-ascription of an enunciative place does not tell in advance how an actual person will feel, experience, nor fulfill or experiment the social role she is subjected to. Here, then, grammar intersects with phenomenology, that is, with concrete experience. Such intersection leads Certeau to make two claims.26 The first claim is that there is a gap between the lettered, systematic discourses on possession and the unarticulated, tentative experience of the possessed. The second claim is even more radical: the possessed cannot articulate a discourse before encountering the symbolic systems proper to the grammars of exorcism or disease; there is just an altering disturbance.

51For Certeau, such inarticulate disturbance leads us to the cause of the trauma, which has to do with the afflicted person’s incapacity to state, and account for, her own identity. A rift is opened between the speaking subject and a definite proper name, a rift expressed by the paradoxical utterance that Certeau quotes from Rimbaud, “Je est un autre, or “I is another” (1988b [1975]: 255). As the possessed speaks, her “I” is always unstable, changing. Therefore the naming performed by the exorcist (or the doctor) “aims at restoring the postulate of all language, that is, a stable relation between the interlocutor, ‘I,’ and a social signifier, the proper name” (1988b [1975]: 256). Exorcism tries to solve this enunciative aberration by giving the possessed woman a proper name taken from a definite and cultural list of demons. Thus, as she recognizes the action of a particular devil within her, the name standing for a character and a set of specific attributes (Asmodeus, Leviathan, etc.), the nun reoccupies a place, though an intermediary one before full recovery, within discourse and social organization. The process works then as a kind of re-calibration of her social coordinates. The ascription of a stable proper name, even a demonic one, permits her rehabilitation among society by ascribing to her determined and reliable properties.

52Though powerful, the naming procedures calling the possessed to order can still be twisted, at least momentarily. To take up Certeau’s latter terminology (2011 [1980]), even if the strategic definition of the prevailing grammar does belong to the subjects of will and power, in this case the authority representatives, seemingly powerless agents such as nuns can use tactics to resist this definition. In fact, the possessed women can be said to use two different tactics to escape from the strategies of symbolization and nomenclature confinement that cultural authorities impose upon them. A first tactic is to refuse to enter the grammar of exorcism. By becoming mute or begging to be left alone, the possessed nuns display a glimpse of the first-person authority that is supposed to disappear from the diabolical grammar of possession. In so doing, the nuns transgress the constitutive rules of demonological experience and draw the investigators on uncertain ground, forcing them to attend a more obedient possessed. A second tactic is to play with the grammar of the exorcism, the Ursulines navigating through the predefined places offered to them by the catalogue of demonic proper names. Instead of occupying a definite place of enunciation, they would constantly wander from Astaroth to Balam to Behemoth to Isacaron, and so on, in an infinite diabolical dance. Such a move, which seems at first to corroborate the effectiveness of the exorcism, ends perverting and undermining it as the exorcist keeps repeating the same naming operation, like a desperate parent tries in vain to impose one’s authority on a mischievous child by repeating incessantly the same ultimatum.

  • 27 In Certeau’s framework, thus, experience is definitely the two-sided entity that we have outlined i (...)

53This reveals how the religious grammar was unsettled from within. The refusal to cooperate that some nuns oppose to the public eye, whether medical, clerical or ordinary, does not take the form of an articulated discourse. Instead it takes the form of a fragmentary, stealthy “art of opportunity” that momentarily twists the force-relationships that are imposed upon them. The fact that even the oppressed can find refuge in the “makeshift creativity” proper to tactics shows that the structural, strategic power of the grammatical places of enunciation is never wholly determining; it is always negotiated and potentially resisted by the subjective, particular way in which agents hold, and are affected by, the place of enunciation they are supposed to take up.27

54As the exorcist tries to capture the devil and prove the truth of the Catholic faith, the nuns’ enunciative moves turn the horrific confrontation with the supernatural into an edifying show staging the appearance of civilized demons and – as Certeau summarizes it bluntly – where snacks are served to the spectators that come to fill the churches (2000 [1970]: 3). The crumbling of the religious empire upon consciences, starting with the possessed, necessitates an urgent action from another power, that of the State. “Since enclosure within the religious onomastic checkerboard does not work, it will be replaced by another grid, that of the police. Thus will end the story of Loudun. Laubardemont, Richelieu’s clerk, will assign places to possessed women – no longer in onomastic squares, but now in the confinement of cells. State policy now classifies by means of walls – another problem” (Certeau 1988b [1975]: 260).

55Thus, the circle comes to an end. Certeau’s analysis brilliantly demonstrates how grammar and phenomenology, social order and the individuals experience are intricately interwoven. The same affection that shakes the possessed unsettles the whole city. The possession at Loudun is then an “existential test” [épreuve] for the nuns and their interlocutors, but also at test for the society as a whole, the principal orders of credibility on which society rests being fully investigated. The nuns’ bodies become the locus where the troubles of Loudun are somatized and exposed. Simultaneously, those same bodies are the place where available grammars are explored and put to the test in order to re-form the social body around other means of symbolization.

IV. At the Heart of Public Critique

56As we have tried to show, the works of Jeanne Favret-Saada and Michel de Certeau share common features, even if they address empirical fields situated in very different epochs and cultures. Both approaches focus on how, at a social, grammatical and phenomenological level, witchcraft and demonology enhance and constrict the scope of possible actions that actors can appropriately undertake. After emphasizing the main meeting points of those approaches, we will discuss on which aspects they diverge. Those divergences, we will see, are mainly due to the different nature of their fieldworks, which are situated in various spaces and times and raise differently the issue of the social and scientific inquiry.

The “Grammatical Correction” of Unspeakable Troubles

57Favret-Saada reconstitutes the actantial scheme of witchcraft by experiencing it in the first person. Indeed, her brilliant ethnography was made possible because – and only because – she was “caught up” in it herself. For solely those directly affected by actual witchcraft situations can grasp the logic of being bewitched and unwitched. By dint of investigating her own experience, the anthropologist succeeds eventually in unearthing and reconstituting the grammar which shapes it in depth: the grammar of an actantial system which unites both the bewitched and the unwitcher against a designated common enemy, in this case the sorcerer. By definition, this system only allows two enunciative positions, that of the bewitched and the unwitcher, thereby silencing the unfortunate sorcerer whose point of view, we will get back to it, is made literally impossible to hear.

58As it summons magical forces, witchcraft therapy provides a sense of coherence to the disruptive events that have affected the bewitched, and endows him with the power to overcome the evil tricks plotted by the sorcerer. But while treating the alleged victim at a supernatural level, the same therapy simultaneously distorts ordinary social relationships and usual communication between neighbors within the community – let us recall, indeed, that the suspected sorcerer must be a landowner from the neighborhood. At the social level, witchcraft thus appears as a survival fight where the capacity of the bewitched to become a landowner is put to the test: he must become able to exert the legal and social violence that is involved in the monopolizing of his familial inheritance and thereby in the despoliation of the share of his siblings. Thus, Favret-Saada proceeds in an ascending way: beginning from her phenomenological experience, she moves to the grammar of witchcraft and finally unveils the social stake that the inheritance of a familial patrimony represents in the rural context of the Bocage.

  • 28 See Sacks 1974; J. Widmer 2001; and P. Gonzalez 2006, 2010 about the normative character of descrip (...)

59Also dwelling on the interplay between social, grammatical and phenomenological dimensions, Certeau’s research covers an analytical spectrum similar to that of Favret-Saada but starts from the opposite pole. Indeed, his analysis begins with the three contending social fields, religion, science, and politics, that the possessions of Loudun unsettle, leading to a redistribution of the influence that those fields exert in society. As seen above, this shift is triggered by the grammars invoked to make sense of the disruption caused by the possessions. Thus, the religious grammar of “exorcism” configures the respective roles of the protagonists (‘exorcists,’ ‘possessed,’ etc.) and their possible actions in a way different from that of the medical grammar of ‘disease.’ Whereas the scientific grammar does not need an external help, its explanations being founded on the natural order of things, the religious etiology will hold out only thanks to the political raison d’État, which goes to its rescue in order to secure the (Catholic) religious ideology that provides the divine law of the French king with its symbolical foundations. But Certeau does not stop his inquiry at the grammatical level; he also reveals how the experience of the nuns operates with and within the actantial system of exorcism, undoing it from the inside and thereby participating to the deconstruction of the homogenous religious order. Importantly, even though the social, grammatical and phenomenological dimensions keep a relative autonomy, they are none the less interwoven, not only descriptively but also ontologically speaking: social structures and individual experiences are closely interrelated by normative grammars of description and action.28

60Strikingly, both Favret-Saada and Certeau thematize the central role played by the act of naming in witchcraft, exorcism and diagnosis. Naming contributes to defining the situation and ascribes definite roles to the participants. Here, the pragmatic dimension of discourse is inseparable from its meaningful counterpart: a grammar provides concrete possibilities for action but also a reason, an order, to what is happening. Extending a psychoanalytical approach to social issues, both studies show how naming participates in the symbolizing of a disruptive trouble. By resorting to the symbolic features of a shared language, symbolization reintegrates the troublesome event or person within the bounds of the community. In other words, symbolizing has a socializing effect: it reunites society around a common etiology and attributes to each participant a role in the grammar selected to make sense of the situation.

Reviving Public Inquiry

61Beyond their striking convergence, the works of Favret-Saada and Certeau have also important dissimilarities that deserve to be addressed. Those differences revolve around the conceptions of publicity and the role of public inquiry, in their practical and normative dimensions. Indeed, the phenomena studied by those authors have a different relationship to publicity and publicisation and are differently open to third parties.

62Thus, witchcraft seems to be unspeakable for two reasons. First, it cannot be the object of declarative discourse or propositional knowledge from those that are involved or have been involved: within witchcraft, speech, beliefs, and experience have no “aboutness”; not only are they pure acts, but also potentially deadly ones. Secondly, witchcraft must be kept secret because it is publicly despised and held up to ridicule, by the medical, political and clerical authorities as well as by the ordinary inhabitants of the Bocage themselves. By contrast, the cases of possession that Certeau dwells on are characterized by their public reach. For the troubles afflicting the nuns quickly give rise to a public inquiry where different reasons, mainly religious, scientific, political, are tested. Whereas witchcraft is deprived of any endogenous publicity and condemned to secrecy, the exorcism is, right from the start, conceived as a public “spectacle” or “theatre,” to take up Certeau’s words.

  • 29 See Favret-Saada 1971.

63From an epistemological and normative perspective, comparing the way witchcraft and possession deal with publicity is very informative. Since the events of Loudun have an endogenous propensity towards publicity and publicisation, it suffices, for the analyst, to unfold the public disputes and to follow the actors in their exchanges and critiques. So if Certeau can adopt a descriptive stance, this is because the phenomenon that he investigates, which is a public, pluralistic inquiry into the critical transformation of religious values and practices, is so to speak doing the “normative job” in his place.29 Even if this public inquiry progressively turns into a collective, fatal dramaturgy, which leads to sentencing to death the alleged sorcerer, the doubt about what or who to believe has been cast, the religious criteria have been dislocated, and a more pluralistic and open order has replaced the monistic “closed” one.

  • 30 See Arendt’s phenomenological account about how objectivity and reality are closely tied to publici (...)

64By contrast, the system of witchcraft described “from within” by Favret-Saada is neither public, nor pluralistic: it is a private interlocution, an individual therapy performed in camera to restore the strength of the bewitched, supposedly drawn away by the witch’s spell. The problem, here, is that the ethnographer, as pragmatic as she may be, cannot count on “folk” resources for critique, which are strikingly absent from her fieldwork. In the Bocage, indeed, a public, critical inquiry into sorcery cannot possibly exist because witchcraft is not an official resource or a public theory of misfortune that would allow natives to attest to their status of competent members of community. On the contrary, witchcraft needs secrecy to survive – a secrecy that the peasants have no interest in disclosing given the symbolic benefits that witchcraft is likely to provide. In the Bocage, witchcraft is sustained by the vicious circle of pragmatic beliefs which are incorrigible because, as we saw, they are shielded from reality tests, but also because they escape from the critical plurality of points of view, a plurality that is a constitutive feature of social and scientific inquiry.30

65Now, when the actors themselves do not perform critique and normative distanciation, we can wonder whether the “endogenous challenge” of pragmatic approaches can be held to the end, even if it means leaving aside the tragic consequences of magical therapy for the unfortunate witch. Indeed, the micro-politics of witchcraft, if it looks like a valuable process of resubjectivation when it is seen from inside, allowing the bewitched to shift position from the status of patient to that of agent, looks very different when seen from outside. Even without endorsing the condescending view of the official authorities or the rational-like, distant stance of the peasants when they are prompted to speak theoretically about it, witchcraft-from-the-outside appears not very commendable. In fact, it is a secret, non-public inquiry which responds to a kind of “schmittian logic,” in the rather negative sense of Carl Schmitt: the recovery of the “bewitched” relies on the old trick of the “enemy within.” The deadly opposition between “me” and “him” enables the landowner and his family to act again as a collective body. In short, witchcraft is governed by an exclusion principle and a process of boundary making whose price is very high: the sacrifice of the scapegoated witch, who bears the brunt of the whole cure.

66Paradoxically, after powerfully criticizing the objectifying discourse stated by external “authorities” (State, science, church) about natives, the endogenous ethnography of Favret-Saada cannot avoid taking up the violence exerted against the alleged witch whose tragic experience of ostracism remains desperately unspoken of. “No need to listen to him,” the bewitched, the unwitcher, if not the ethnographer, say, “his death speaks for him.” Reduced to the status of a “third person,” the alleged witch is excluded from the space of the interlocution, he is never an “I” or a “you,” including for the ethnographer.Now, as a lot of commentators inspired by Benveniste (1966) have emphasized, the moral and political significance of a system can be measured by its capacity to extend the number of people who can say “I” and then refer to themselves in a self-actualizing manner.

  • 31 About “experience-near,” see Geertz l975.

67It goes without saying that, from a normative point of view, this “schmittian-like” logic of witchcraft is at the opposite of the public inquiry which, for pragmatic philosophers such as Dewey, allows people to distance themselves from institutional systems and to recover the individual and collective power of determining the orientations of the common life. But is this normative appreciation of the moral and political implications of witchcraft compatible with the symmetrical anthropology pioneered by Favret-Saada? This question, very close to Boltanski’s reflections on critique where we started from, raises a fundamental issue: the comprehensive description of a social phenomenon, based upon the “experience-near” stance which is essential to understand what the “affected” go through, cannot take into account its moral and political implications.31 As fine-grained and demanding as they are, descriptive accounts seem then to be only the first step of social and scientific inquiry. Sooner or later, they should be followed by a second step, that of the normative assessment of the moral and political implications of the phenomenon under scrutiny.

68If we follow John Dewey (1991 [1927]), such second-step, normative stance is fully necessary in a social and scientific inquiry, which is ideally governed by what he calls “the method of democracy” – a method aiming at bringing conflicts, interests, and experiences out into the open where they can be publicly discussed, judged and improved. Here again, this method is particularly well – and unintentionally, of course – illustrated by the case of Loudun. Indeed, the “great public trial” which turns the relation between the sacred and the profane into an object of collective inquiry continually expands, up to and including the Princes of the Church, of the State and inquisitive laypeople. As the voice of the Devil makes its own’s way through the social circles, more and more wide-ranging, it changes status: from an impromptu, destructive, violent force, it slowly becomes a civilized figure of speech. “With the possessed, the Devil speaks, he writes. If I dare say so, he publishes” (Certeau 2000 [1970]: 8). The overcirculation of the words of the demon takes them away from their authoritative sources and diminishes their value: as an unsteady currency, they become more uncertain and, above all, more human. Torn apart by divergent intellectual systems, indeed, they no longer belong to the supernatural language, but are downgraded to human language and disputations. The progressive weakening of the initial force of the Devil illustrates fantastically well the potential emancipatory effect of publicity: it erodes the power of totalization that enclosed phenomena tend to entail. Once open to public scrutiny, the force of totalization can hardly endure, and this is why public inquiry, either social or scientific, is the mainstay of democracy in the normative, ideal sense of the term.

69As seen above, witchcraft in the Bocage appears as an anti-democratic problem solving, an “anti-public” grammar that is not held accountable for its moral and political consequences by the actors themselves. So, in this case, some sort of “external” standpoint of the social scientist is needed: only such standpoint can show that witchcraft is not solely the site of validation of the social fate of the alleged victim, but also the site of exclusion of the ostracized witch. How is it possible, then, to critically reveal the ill-formed moral and political constitution of witchcraft without giving up on the endogenous stance of the ethnographer? How to reach, in this context, the “broadened way of thinking” or “enlarged mentality,” based upon the possible or actual judgment of others, which ensures a publicly-minded, universalizing form of political judgment (Arendt 1982)?

Complex Exteriority or an Eye for Publicity?

The latter Boltanski (2011 [2009]) distinguishes two kinds of critical processes. The first process, called “simple exteriority, is internal to the activities of the social actors and able, at best, to address power issues; to produce it, the social scientist simply follows and describes what the actors are doing and how they assess their doings. The second process, called “complex exteriority, is external to the point of view of the people involved in social action; as such, it is able to unveil domination, not by assessing some local activities, but by producing a theoretical critique of the social order. For, unlike power, domination is invisible to the social actors and can only be revealed by the technical and theoretical skills of the (critical) social scientist. Boltanski defines then this second operation as metacritical in that it is able to produce a totalizing – as opposed to a partial – point of view on reality (which is his concept for “social order”).

  • 32 “The spectator, because he is not involved, can perceive this design of providence or nature, which (...)
  • 33 “We take then our point of departure from the objective fact that human acts have consequences upon (...)

We reach here the heart of the disagreement between Boltanski’s metapragmatical approach to critique vis-à-vis the more pragmatist one we tried to lay out by commenting on Favret-Saada’s and especially Certeau’s work. Rather than opposing a simple exteriority to a complex one, we would like to stress, along the lines of Dewey’s pragmatism and Arendt’s phenomenology, the continuum between the internal points of view about an action or event, and more external ones. Such a continuum is a feature of the phenomenon itself, which is always open to both internal and external gazes and, by way of consequence, to a plurality of perceptions – a plurality that is thus intrinsic to the whole phenomenon. Interestingly, in Arendt’s phenomenological language, the distinction between the internal and external gaze matches up with the distinction between the stance of the actor and that of the spectator: “only the spectator occupies a position that enables him to see the whole; the actor, because he is part of the play, must enact his part – he is partial by definition” (2006 [1961]: 55). Of course, the position of the spectator varies and can go from that of the impartial judge attending an event to the judgment of history but, in any case, it is the spectator’s view that “carrie[s] the ultimate meaning of the event.”32 Such understanding is very similar to Dewey’s experimental conception of the public: the public arises when individuals, indirectly affected by the consequences of others’ actions, perceive those effects and gather together in order to secure or avoid them.33 Even though Arendt’s spectator seems less politically active than Dewey’s public, both authors posit that the nature of a phenomenon exceeds the internal point of view of the actors, and that the third party perspective is consubstantial with it.

  • 34 Dewey provides a point of method about the use of theory: “Political theories have shared in the ab (...)

It is precisely this “openness to the third” that allows the inquirer to remain faithful to the nature of the phenomenon without automatically endorsing the actors’ commitments. Since the critical point of view is already built in the phenomenon, normative critique does not require radical exteriority, contrary to what Boltanski’s approach problematically suggests. Indeed, the dichotomy that Boltanski posits between simple and external exteriority breaks the unity of the phenomenon and separates the direct elements of the phenomenon (the actors’ points of view) from their indirect counterparts (the public’s points of view). This leads to a second difficulty: since direct and indirect elements have been disconnected, the third point of view linked to the indirect consequences of the phenomenon has been obliterated, forcing Boltanksi to reintroduce it under the form of an alien critical point of view: that of the metapragmatical critique. The problem is that, as Dewey put it, such critical move, far from being emancipating, is alienating, for the ideological – hence, dogmatic – critique that it advocates is disconnected from the real consequences of the phenomenon.34 Paradoxically, such disconnection severs the link between the phenomenon and social action, and prevents the actors or the public from acting upon its effective consequences. Rather than empowering people, such critical stance deprives them of their capacity to act in an appropriate manner, and replaces social inquiry with an ideological construct.

  • 35 Again, see Stavo-Debauge 2011, and Kaufmann 2012.

But there is a third – and somehow more disturbing – criticism that can be addressed to Boltanski’s position. According to the author, a metacritical point of view, which provides a totalizing perspective on reality, is necessary to deconstruct the reigning social order. However, it is far from certain that such a totalizing perspective is neither needed nor desirable, for it might rapidly degenerate into a totalitarian point of view, hostile as such to a critical pluralism.35 A totalizing perspective – especially one that severs the link between the direct and indirect consequences of a phenomenon – risks turning into a unique, arbitrary claim that bypasses the pluralistic composition of society and the multiplicity of opinions.

70Our criticism to Boltanski’s new sociology allows us to better specify the more pragmatist approach we advocate here. The puzzle produced by the articulation of the internal description of a phenomenon with its normative assessment, which make it accountable vis-à-vis the rest of society, can indeed be solved through pragmatist means. In concrete terms, the solution, ethnographically and normatively correct, is to follow the direct and indirect consequences of a phenomenon, to lift the ban of enunciation and to map out as many points of views as possible – including, in our witchcraft cases, the point of view of the alleged sorcerer (or witch) and that of the official authorities. This is the only way to increase the range of the possible views of the phenomenon and, thereby, to render the scope of the thought as general and pluralist as possible. Such a way to proceed will preserve the unity of the phenomenon: while acknowledging the diverse – and critical – perspectives that it offers to scrutiny, from the actors to the public to the social scientist, it nevertheless embraces them with an overall pragmatist account.

Top of page


Anscombe E., (1957), Intention, Oxford, Oxford University Press.

Arendt H., (1998 [1958]), The Human Condition, Chicago, University of Chicago Press.

Arendt H., (2006 [1961]), Truth and Politics, in Between Past and Future. Eight Exercises in Political Thought, New York, Penguin Books, 227-64.

Austin J. L., (1975 [1962]), How To Do Things With Words, Cambridge, Harvard University Press.

Bénatouïl T., (1999a), “Critique et pragmatique en sociologie. Quelques principes de lecture,” Annales. Histoire, Sciences Sociales, 54 (2), 281-317.

Bénatouïl T., (1999b), “A Tale of Two Sociologies. The Critical and the Pragmatic Stance in Contemporary French Sociology,” European Journal of Social Theory, 2 (3), 379-96.

Benveniste É., (1966), Problèmes de linguistique générale, 1, Paris, Gallimard.

Boltanski L., (1990), L’amour et la justice comme compétences. Trois essais de sociologie de l’action, Paris, Métaillé.

Boltanski L., (1999 [1993]), Distant Suffering. Morality, Media, and Politics, Cambridge, Cambridge University Press.

Boltanski L., (2008), Rendre la réalité inacceptable, Paris, Démopolis.

Boltanski L., (2011 [2009]), On Critique. A Sociology of Emancipation, Cambridge UK, Polity.

Boltanski L. & L  Thévenot, (2006 [1991]), On Justification. Economies of Worth, Princeton, Princeton University Press.

Boltanski L. & L  Thévenot, (1999), “The Sociology of Critical Capacity,” European Journal of Social Theory, 2 (3), 359-77.

Boltanski L., Darré Y. & M.-A. Schiltz, (1984), “La dénonciation,” Actes de la recherche en sciences sociales, 51 (1), 3-40.

Breviglieri M. & J. Stavo-Debauge, (1999), “Le geste pragmatique de la sociologie française. Autour des travaux de Luc Boltanski et Laurent Thévenot,” Antropolítica 7, 7-22.

Certeau M. de, (1987 [1969]), Autorités chrétiennes et structures sociales, in La faiblesse de croire, Paris, Seuil, 77-128.

Certeau M. de, (2005 [1969]), Les magistrats devant les sorciers du xviie siècle, in Le lieu de l’autre. Histoire religieuse et mystique, Paris, Gallimard/Seuil, 301-22.

Certeau M. de, (2000 [1970]), The Possession at Loudun, Chicago, University of Chicago Press.

Certeau M. de, (1988a [1975]), The Writing of History, New York, Columbia University Press.

Certeau M. de, (1988b [1975]), Language Altered: The Sorcerer’s Speech, in The Writing of History, New York, Columbia University Press, p. 244-68.

Certeau M. de, (2011 [1980]), The Practice of Everyday Life, Berkeley, University of California Press.

Certeau M. de, (1981), Une pratique sociale de la différence: croire, in Faire croire: modalités de la diffusion et de la réception des messages religieux du xiie au xve siècles, Rome, École française de Rome, 363-83.

Certeau M. de, (1985), “Le croyable, ou l’institution du croire,” Semiotica, 54 (1-2), 251-66.

Claverie É., (1994), “Procès, affaire, cause. Voltaire et l’innovation critique,” Politix, 7 (26), 76-85.

Claverie É., (1998), “La naissance d’une forme politique: l’affaire du chevalier de La Barre,” in Cheyronnaud J. & Claverie É. et al., (eds), Critique et affaires de blasphème à l’époque des Lumières, Paris, Honoré Champion, 185-260.

Dayan D., (2001), “The Peculiar Public of Television,” Media, Culture & Society, 23 (6), 743-65.

Dewey J., (1991 [1927]), The Public and Its Problems, Athens, Swallow Press.

Durkheim É., (1982 [1895]), The Rules of Sociological Method, New York, Free Press.

Durkheim É., (1983 [1955]), Pragmatism and Sociology, Cambridge, Cambridge University Press.

Favret-Saada J., (1980 [1977]), Deadly Words. Witchcraft in the Bocage, Cambridge, Cambridge University Press.

Favret-Saada J., (1971), “Review of Michel de Certeau, La Possession de Loudun,” L’Homme, 11 (2), 121.

Favret-Saada J., (2009), Désorceler, Paris, Éditions de l’Olivier.

Favret-Saada J., (2011), “La mort aux trousses,” Penser/Rêver 20, 207-17.

Frega R. & F. Carreira da Silva, (2011), “Editor’s Introduction to The Symposia ‘Pragmatism and the Social Sciences: A Century of Influences and Interactions’,” European Journal of Pragmatism and American Philosophy, 3 (2), 2-6

Franks D. & F. F. Seeburger, (1980), “The Person Behind the Word: Mead’s Theory of Universals and a Shift of Focus in Symbolic Interactionism,” Symbolic Interaction 3, 41-58.

Geertz C., (1974), “From the Native’s Point of View: On the Nature of Anthropological Understanding,” Bulletin of the American Academy of Arts and Sciences, 28 (1), 26-45.

Gonzalez P., (2006), “De l’économie morale du discours sociologique: autour d’une ‘classe’ et d’un propos ‘déplacé’,” Carnets de bord 11, 25-39.

Gonzalez P., (2010), “(D)écrire : catégoriser, prendre des notes, écrire,” in Cefaï D. (ed.), L’engagement ethnographique, Paris, Éditions de l’EHESS, 107-28.

Kaufmann L., (2006), “Les voies de la déférence. De la nature des concepts sociopolitiques,” Langage & Société 117, 89-116.

Kaufmann L., (2012), “Agir en règle. Le pari grammatical de la sociologie pragmatique à l’épreuve de la critique,” Raison publique, 16, forth.

Latour B., (2005), Reassembling the Social. An Introduction to Actor-Network-Theory, Oxford, Oxford University Press.

Lemieux C., (2007), “À quoi sert l’analyse des controverses?,” Mille neuf cent, 25 (1), 191-212.

Piette A., (2003), “Forum: Religious Behaviour and Ordinary Activity,” Social Compass, 50 (3), 361-5.

Putnam H., (1981), Reason, Truth, and History, Cambridge, Cambridge University Press.

Ricœur P., (1977), La sémantique de l’action, Paris, CNRS.

Sacks H., (1974 [1972]),On the Analysability of Stories by Children,” in Turner R. (ed.), Ethnomethodology. Selected Readings, Harmondsworth, Penguin Books, p. 216-32.

Stavo-Debauge J., (2011), “De la critique: une critique. Sur le geste ‘radical’ de Luc Boltanski,” EspacesTemps, [] & [].

Strawson P. F., (1992), Analysis and Metaphysics: An Introduction to Philosophy, Oxford, Oxford University Press.

Thévenot L., (1990), “L’action qui convient,” Raisons pratiques 1, 205-42.

Thévenot L., (2011), “Power and Oppression from the Perspective of the Sociology of Engagements: A Comparison with Bourdieu’s and Dewey’s Critical Approaches to Practical Activities,” Irish Journal of Sociology, 19 (1), 35-67.

Veyne P., (1988), “Conduites sans croyance et œuvres d’art sans spectateurs,” Diogène 143, 3-22.

Widmer J., (2001), “Catégorisations, tours de parole et sociologie,” in Fornel M., Ogien A. & Quéré L. (eds), L’ethnométhodologie. Une sociologie radicale, Paris, La Découverte, 207-38.

Widmer J., (2010), Discours et cognition sociale. Une approche sociologique, Paris, Éditions des archives contemporaines.

Top of page


1 Among others: Marc Breviglieri, Damien de Blic, Nicolas Dodier, Cyril Lemieux, Dominique Linhardt, Jean-Philippe Heurtin, Joan Stavo-Debauge, Danny Trom.

2 See Bénatouïl (1999a: 2), and Lemieux (2007: 192).

3 For notable exceptions, see Claverie 1994, 1998, and Thévenot 2011.

4 The model of Thévenot goes in this direction, though.

5 Those social scientists came from different horizons, but converged on an editorial project, starting in 1990, a journal entitled Raisons Pratiques [practical reasons], published by the École des hautes études en sciences sociales [EHESS] of Paris. Daniel Cefaï, Bernard Conein, Élisabeth Claverie, Bruno Latour, Louis Quéré, Laurent Thévenot, Jean Widmer and others, have played a very important role in this enterprise.

6 Latour himself defines his sociology as being “half Garfinkel and half Greimas” (2005: 54).

7 See also Bénatouïl 1999b, and Boltanski (1999 [1993]).

8 About this primacy, see Frega & Carreira da Silva 2011.

9 To our knowledge, Putnam is not quoted by pragmatic sociologists such as Thévenot or Boltanski; but it seems to us that his “internal realism” fits very well the ontological premises of their framework.

10 On this resistance, inspired by the work of Mead, see Franks & Seeburger 1980.

11 On these two kinds of metaphysics, see Strawson 1992.

12 On this issue, see Kaufmann 2012, Stavo-Debauge 2011.

13 See de Certeau 1985.

14 Of course, for an historian, the task is particularly challenging, if not impossible. For M. de Certeau (1988a [1975]), the writing of history has to face the tragic loss of those practices, experiences and affects that, yesterday, were alive, and are dead today, a loss that the act of writing tries to ward off, converting what is lost into a text. Thus the text of archives is a first trace of the past that a second text, the text written by the historian, reinstates in a meaningful relationship with the present by imposing the – sometimes wrong – unity and coherence of discourse upon the heterogeneity of life as it was actually experienced in the past. Paradoxically, hence, past life can survive only if it is captured in a system of representation and knowledge that has, thereby, the indispensable and painful “beauty of the dead.”

15 In 1970s, a Normand landowner is always a man for reasons related to the procedures of property transmission. Therefore, when evoking him, we will use the masculine.

16 Usually, Favret-Saada describes women in the position of the unwitcher. Therefore, we will use a feminine pronoun to designate that role.

17 One finds the same idea in the work of the anthropologist Albert Piette (2003).

18 On the “conducts without beliefs,” see the well-known paper of Paul Veyne (1988).

19 Cases are documented in Denmark, England, Germany, Netherlands, Switzerland, and, for France, mostly in its northern regions. Spain and Italy will be spared (Certeau 2000 [1970]: 5).

20 See Certeau’s discussion of witchcraft (2005).

21 “The discourse of possession turns on an absent figure whom it gradually renders more precise: the sorcerer. Contrary to what one might suppose, the theatre at Loudun is not provoked by that formidable or fantastic figure. It is not determined by his approach or his visibility. It needs him in order to function. Thus, as it organizes itself for itself, developing and refining its procedures, it defines the silhouette, the name, the misdeeds of the ‘possessor,’ upon whom possession depends” (Certeau 2000 [1970]: 52).

22 Credibility is a key concept in Certeau’s thinking that has to do with “the machinery of representation”. In The Practice of everyday, the author states: “The credibility of a discourse is [both] what moves believers and leads them up the garden path” (our translation) – and not, as the English translation puts it: “what first makes believers act in accord with it [the discourse]” (Certeau 2011 [1980]: 148). There is always an ambiguous element in credibility, especially when that credibility works as a foundation for the social order.

23 For a similar approach in sociology deeply influenced by ethnomethodology, see the work of Jean Widmer (2010).

24 Certeau does not systematize nor make an important use of the concept of grammar. Nonetheless our utilization seems consistent with his way of understanding it: “What makes the discourse of possession possible […] is that the nun must not remember what happened, that no personal element be permitted to compromise the automatic functioning of the diabolical grammar” (Certeau 2000 [1970]: 40).

25 As Certeau (1981) reflects on trust and credibility, he will resort on the etymological properties and the semantic field of “credit,” with its economic overtones. See also Chapter XIII of The Practice of Everyday Life (2011 [1980]) originally entitled “Political Credibilities” [Crédibilités politiques], “Believing and Making People Believe” in English.

26 Certeau draws those postulates from the relationship between the modern psychiatrist and the insane, a parallel that the author applies to the cases of possessions.

27 In Certeau’s framework, thus, experience is definitely the two-sided entity that we have outlined in our introductive part: the structuring, objective power of grammar, both enabling and constraining, does have a phenomenological, subjective counterpart, mostly revealed in singular “ways of doing.”

28 See Sacks 1974; J. Widmer 2001; and P. Gonzalez 2006, 2010 about the normative character of descriptive devices.

29 See Favret-Saada 1971.

30 See Arendt’s phenomenological account about how objectivity and reality are closely tied to publicity: “[public] means, first, that everything that appears in public can be seen and heard by everybody and has the widest possible publicity. For us, appearance – something that is being seen and heard by others as well as ourselves – constitutes reality” (Arendt 1998 [1958]: 50). Arendt clearly distinguishes between the political plurality of opinions and the multiple tests that a scientific statement or fact must pass in order to be established as true: “Truth in the sciences is dependent on the experiment that can be repeated by others; it requires general validity” (1982: 40). See also her important essay on “Truth and Politics” (2006 [1961]).

31 About “experience-near,” see Geertz l975.

32 “The spectator, because he is not involved, can perceive this design of providence or nature, which is hidden from the actor. So we have the spectacle and the spectator on one side, the actors and all the single events and contingent, haphazard happenings on the other. In the context of the French Revolution, it seemed to Kant that the spectator’s view carried the ultimate meaning of the event, although this view yielded no maximum for acting” (Arendt 2006 [1961]: 52).

33 “We take then our point of departure from the objective fact that human acts have consequences upon others, that some of these consequences are perceived, and that their perception leads to subsequent effort to control action so as to secure some consequences and avoid others. Following this clew, we are led to remark that the consequences are of two kinds, those which affect the persons directly engaged in a transaction, and those which affect others beyond those immediately concerned. In this distinction we find the germ of the distinction between the private and the public. When indirect consequences are recognized and there is effort to regulate them, something having the traits of a state comes into existence” (Dewey 1991 [1927]: 12).

34 Dewey provides a point of method about the use of theory: “Political theories have shared in the absolutistic character of philosophy generally. By this is meant something much more than philosophies of the Absolute. Even professedly empirical philosophies have assumed a certain finality and foreverness in their theories which may be expressed by saying that they have been non-historical in character. They have isolated their subject-matter from its connections, and any isolated subject-matter becomes unqualified in the degree of its disconnection” (1991 [1927]: 194-5).

35 Again, see Stavo-Debauge 2011, and Kaufmann 2012.

Top of page


Electronic reference

Philippe Gonzalez and Laurence Kaufmann, « The Social Scientist, the Public, and the Pragmatist Gaze », European Journal of Pragmatism and American Philosophy [Online], IV - 1 | 2012, Online since 23 July 2012, connection on 23 September 2017. URL : ; DOI : 10.4000/ejpap.766

Top of page

About the authors

Philippe Gonzalez

University of Lausanne.

Laurence Kaufmann

University of Lausanne.

Top of page


Licence Creative Commons
Author retains copyright and grants the European Journal of Pragmatism and American Philosophy right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Top of page
  • Logo Associazone culturale Pragma
  • Logo DOAJ – Directory of Open Access Journal
  • Logo École des Hautes Études en Sciences Sociales – EHESS