The Order of Discourse
Notes on Michel Foucault’s The Order of Discourse, an inaugural lecture at the Collège de France, given on Dec. 2, 1970. (L’Ordre du Discours, Paris: Gallimard, 1970). (Here is an outline in English, and here a summary in French). (My comments are added in parentheses and italics).
Foucault begins by commenting on the context of his own discourse—the fear of speaking, the institutional context. He opposes the automatisms which would make us let discourse speak through us (as in Beckett’s The Unnamable). Desire wants a transparent discourse, one which carries us along, not an order full of hazard and risk. On the other hand, institutions try to control discourse through the very gesture of giving it a place. But discourse (this discourse of Foucault’s too, by implication—JAGL) has a subversive power, beyond the play of desire and bureaucratic institutions.
Therefore, he assumes "that in every society the production of discourse is at once controlled, selected, organised and redistributed by a certain number of procedures whose role is to ward off its powers and dangers, to gain mastery over its chance events, to evade its ponderous, formidable materiality" (52).
A) Procedures of exclusion
1) Prohibition. The subject matter of discourse may be forbidden; so may the speaker, or the occasion. E.g. sexuality and politics are often forbidden. But they proliferate thorugh discourse, instead of becoming extinguished. Discourse is also an object of desire, a power to get hold of, not just a medium.
2) Division of discourses, or rejection. Such is the opposition between madness and reason. The discourse of madness is rejected, taken as noise, but it has a strange circulation: the "wise madman" was given a say only on the stage. Today the same division is played otherwise, filtered through the discourse of psychoanalysis, etc.
3) The opposition between truth and falsity. The will to know is governed by a system of exclusions. The ancient discourse of truth in Greece had a performative nature, it was linked to power: but in the fifth century BC, the discourse of truth became linked to its meaning, its reference, the value of the utterance. Truth becomes semantic: it is displaced from the act to the utterance. It is the origin of philosophy, of the will to truth, born in opposition to the sophists. There is an evolution in this will to truth, not always due to a discovery. Around 1600 (Foucault is probably thinking of Francis Bacon) there is a new regime of classification and mensuration being born in England. It is the origin of technicism/positivism. The will to truth has a history of its own, which is not the history of constraining truths: a history of the delimitation of methods, of the objects of knowledge... It is sustained by institutional practices, among them the use of knowledge. This "truth" exerts pressure on the other discourses, which converge towards the discourse of truth. (E.g. realist or sincere literature). Thus, the modern penal code is no longer founded on power or spectacle, but on the scientific discourse of psychology, psychiatry, etc. (See on this point Foucault’s work Discipline and Punish). That is, procedures of exclusion (1) and (2) become subordinated to (3): they become more fragile, while (3) grows all the time. But (3) is masked: in the sense that it necessarily ignores its links with power and desire. In Nietzsche, Artaud, Bataille... we appreciate a will to truth turned towards a critique of the notion of truth itself; they are the models for Foucault’s analysis.
B) Internal procedures
Other procedures for the control of discourse are internal—principles of order within the discourses themselves.
4) Commentary. That is, the division between canonical texts and their commentaries. Some texts are privileged (the canon, in religion, law, literature or science); others are commentaries of these major texts. This hierarchy is always active, this opposition between the "truly original" text and the commentary, in spite of diverse blurrings. For instance, Borges and others suppress one of the terms (e.g. writing commentaries of imaginary texts), but not the relationship itself.
5) The author (as a principle for the grouping of discourses, a principle of unity and origin of their signification, as a focus of coherence) is another "principle of rarefaction" in discourse. The name of the author has a different use and value in scientific discourse and in literary discourse; these two types of discourse have exchanged their faith in the author since the Middle Ages. Scientific authors are no longer treated as authorities, but in literature "The author is what gives the disturbing language of fiction its unities, its nodes of coherence, its insertion of in the real" (58). The individual in question receives his modes of behaviour from the author-funciton as it is defined in his age, "or as he modifies it in his turn" (59).
(On the author-funcion as a principle of constraint and production, Foucault’s observations here must of course be complemented with his paper "What Is an Author?").
6) Disciplinarity. Disciplines constitute an anonymous system, in contrast to (4) and (5). There is a constant need for new formulations, for new propositions within the discipline. Disciplines define the kind of discourse on their object which will become a part of the discipline (not just any kind of discourse). For instance, from the eighteenth century the discourse of botany no longer includes the symbolic values of plants, which also used to fall within its purview, for instance in the sixteenth century. Disciplines have theoretical horizons. (This kind of analysis is undertaken in Foucault’s works The Archaeology of Knowledge and The Order of Things: An Archaeology of the Human Sciences. Foucault’s analysis at this point is to be related with T. S. Kuhn’s notions of paradigm and scientific revolution, and with Paul Feyerabend’s critique of disciplinary methodology). In order to be recognized as such by a discipline, in order to be true or false, a discourse must be "in the true", it must in a sense play the game of the discipline, accept its discourse. Disciplines are also a principle of control of the production of discourse; they fix limits to what can be said within the discipline. They are at once an element of constraint and an element of creation and proliferation. But let us examine another set of principles linked to constraint:
C) Conditions of access to discourse, for instance
7) The qualification of the speaking subjet to enter the order of discourse. Rituals define this qualification, as well as the signs which must accompany the discourse.
8) Societies of discourse, which preserve discourses and make them circulate within a closed space. Today they are loose, but the very act of writing and publishing in modern societies is a society of discourse in the wider sense. Literatury conventions as well as technical or scientific secrets, set their own constraints. (Curiously enough, Foucault does not draw attention to the role of academic societies, universities, etc. as societies of discourse in the most literal sense. Whereas it could be argued that the University is itself a materialized society of discourse, an official society of disciplinary discourse, physically and administratively organized in the shape of buildings, departments, and groups of teachers. The role of societies of discourse in shaping attention should be studied within the wider anthropological context of attention strategies in human groups. See e.g. Brian Boyd’s analysis in The Origin of Stories, and my paper "Atención a la atención"—JAGL).
9) Doctrines— They belong to a group, but they tend to become disseminated, and to require the allegiance of individuals. "Heresy and orthodoxy do not derive from a fanatical exaggerantion of the doctrinal mechanisms, but rather belong fundamentally to them". (64) Doctrine is a way of binding individuals to certain types of enunciation. But it is also an enunciation which functions as a sign which binds the individuals in a group (a double binding).
10) Appropriation—i.e. the social appropriation of discourse. Discursive value being an object of desire. (Here one thinks of such works as Pierre Bourdieu’s studies of symbolic value in Language and Symbolic Power, or the essays collected in Shakespeare and Appropriation, ed. Christy Desmet and Robert Sawyer—JAGL). Social appropriation takes place, for instance, through educational systems.
One should note, Foucault adds, that all these types for the subjection and ordering of discourse are found mixed up or combined, their separation is an act of abstraction. (And one should stress the double action of these mechanisms: the existence of doctrines or disciplines, the qualification of a subject to speak, etc., are not only a means for the "rarefaction" or constraining of discourse, but also a principle of proliferation of discourse along certain lines or in certain contexts—a stimulus for the production of certain kinds of discourse, e.g. research papers, sermons, poems... They may constitute the very conditions of possibility for those discourses to exist—JAGL).
A number of notions promoted by philosophers develop in keeping with these activities of limitation and exclusion of discourse. For instance: the notion of ideal truth as the law of discourse, of immanent rationality as a principle of unfolding for discourse. The desire for truth itself, or the ability to think it. Philosophical discours in the West presents itself as a simple putting in words of a preexisting thought—or conversely (and this would be the case for structuralist views on the matter) thought would be the mere effect of preexisting linguistic structures. (One thinks of Derrida’s critique of "logocentrism"—according to him, the philosophical tradition of the West tries to present the ideality of the pure presence of thought or meaning, evading the recognition of the materiality of signs, texts, discourses, etc. which articulate or shape it). There exists, therefore, an elision of the reality of discourse, through a variety of means.
a) Through the notion of the founding subject, the subject manifesting himself without a mediation through discourse.
b) Or else through the notion of the originating experience. Which assumes the existence of pre-discursive meanings in the world, of which the discourse would be a modest reading. (One may intuit here Foucault’s critique of Heidegger’s notion of signification. However, and quite from another tack, the notion of discursive meaning resting on pre-discursive meaning is criticized by contemporary evolutionary and cognitivist approaches to signification. For a strong statement against constructivism, see Joseph Carroll’s papers, for instance "Literary Study and Evolutionary Psychology". To put it in a nutshell: if "man gave names to all the animals" is the foundational act of discourse, there must have been a pre-discursive notion of which animals to name, before the names came along).
c) Or through the idea of universal mediation. At first glance this would seem to favour discourse and concept everywhere, but actually this notion makes discourse rest on self-consciousness, "discourse is little more than the gleaming of a truth in the process of being born to its own gaze" (66).
(a)= writing; (b)=reading; (c)=exchange. All three never put anything at stake except signs; "discourse is annulled in its reality and put at the disposal of the signifier" (66).
In spite of appearances, the logophilia which characterises our culture is due to a fear of the uncontrollable nature of discourse; it is actually logophobia. This deserves analysis: "we must call into question our will to truth, restore to discourse its character as an event, and finally throw off the sovereignty of the signifier" (66) (—a position on the part of Foucault which acknowledges the power of structures without reducing all events to a predetermined play of pre-existing structures). The method to carry out this analysis would be as follows.
Strategies for future work on the analysis of discourses:
- A reversal of the analytical tradition. What is currently valued as a source of creativity, fecondity, ideals... Foucault is suspicious of.
- Discontinuity: Nonetheless, marginal or suppressed discourses should not be idealised. Discourses do not form coherent wholes; they are discontinuous practices which crisscross or ignore each other.
- Specificity: Against the myth which consists in believing that one is decyphering a pre-established signification. Discourse must be conceived as an active principle, a form of violence exerted on things. The world is not discursive in nature.
- Exteriority: The external conditions of the possibility of discourses must be analyzed.
"Term for term we find the notion of event opposed to that of creation, series opposed to unity, regularity opposed to originality, and condition of possibility opposed to signification" (67).
Foucault argues against the prevalent notions of origin and essence. Historical studies today do not abandon the study of specific events: rather, they enlarge the field of study of specific events, and conceive new groups of events. The conditions of possibility of events are circumscribed, but this notion is not yet a structural one: it is grounded on positivist assumptions. An event takes place in the material realm, as a relationship between material elements or a selection of material elements. Foucault notes the need to develop a materialism of the incorporeal, a study of discursive events. Series of discursive events are heterogenous because discourses dissolve the subject and the instant into a plurality of positions and of possible functions. Randomness is a necessary category, to be opposed to the assumption of a mechanical causality. Chance, discontinuity, and materiality are to be found at the roots of thought; there are attempts to exorcise them "by narrating the continous unravelling of an ideal necessity" (69). (One finds here a critique of narrativisation, of the retroactive construction of "well-made" stories. See my critique of retroactive dynamics and hindsight bias in the field of criticism and theory—JAGL).
Foucault’s proposal is to link history to historiography— to the actual writing of that history by historians (Which is to some extent the project of cultural materialists in the Anglo-Saxon countries: see for instance the essays collected in Dollimore and Sinfield’s Materialist Shakespeare).
Kinds of analysis
Two kinds of analysis are suggested by Foucault: (a) critical analysis (b) genealogical analysis.
(a) Critical analysis, in the philosophical tradition, against sophistry: it would be concerned with
(1) An analysis of strategies of exclusion (e.g. the opposition between madness and reason, a study of sexuality and its evolution). (One thinks here of Foucault’s own analyses, both past, in Madness and Civilisation, and future, The History of Sexuality. As we see, The Order of Discourse also organizes Foucault’s own discourse, mapping a programme of critical analysis which encompasses the whole of his work).
(2) An analysis of the birth of sciences of observation. (Ditto for The History of Sexuality).
(3) A study of the great foundational acts of modern science in the 19th century, of positivist ideologies.These are "three stages of our philistinism".
Critical analysis will address the study of scientific expertise as a way of grounding the penal code. Following this tack, the procedures of limitation of discourse (the author, the commentary, the discipline) must be studied as they apply for instance to the history of medicine. Also, the development of the ideology of authorship and of the Work in literature must be studied as they unfold, "using, modifying, and displacing the procedures of religious exegesis, biblical criticism, hagiography, historical or legendary ’lives’, autobiography, and memoirs" (71). E.g. a study of the role of the figure of Freud in psychoanalytic knowledge. (See on this point the notion of founders of disciplinary practices as discussed in "What Is an Author?").
(b) Genealogical analysis, not really separable from critical analysis, since "any critical task, putting in question the instances of control, must at the same time analyse the discursive regularities through which they are formed" (72). Nonetheless, there is a difference in perspective. Each discursive series has its own kind of regularity, and the constitution of a new system out of the previous ones is not a smooth transition. (At this point one must think of Foucault’s notion of episteme as expounded in The Order of Things, —or again, see Thomas S. Kuhn’s quite related notion of scientific revolutions).
The critical approach studies the mechanisms for the control of discourse; the genealogical approach addresses the formation of domains of objects by means of discourse, the genesis of the possibility of truth within discourse. Discourse is both stimulated and rarefied by these formations of discourse; but we do not find in them the universal rule of one sense, there is no reign of the signifier (i.e. discourses are not automatically generated by a signifying system) and in this sense Foucault is not a structuralist.
VIII. Models and forerunners
Models and forerunners of this kind of analysis are to be found in the work of Dumézil—in his overcoming of traditional exegesis and of linguistic formalism through the use of comparison, in his transformation of critical discourse and of its relationship to the institution. Also in Canguilhem, in his studies on the history of science. Canguilhem sees the history of science as a succession of models and conceptual instruments, not as a chronicle of events and discoveries.
In the work of Jean Hyppolite—a Hegelian who allows us to escape Hegel by acknowledging our debt to him. Is a non-Hegelian philosophy possible? Hyppolite sees philosophy not as system but as process, a perpetual interrogation. It is in moving contact with non-philosophy, with the other disciplines of knowledge (psychoanalysis, mathematics...). In Hyppolite one finds a study of the foundations of philosophical thought and of its formal structure: the essential problem of philosophy is faced, that is: the problem of a discourse which aspires to universal value coming nonetheless from an individual and historically located subject.
It is with a moving homage to his master Hyppolite that Foucault closes his discourse on discourse—one which is also universal and historically situated.
Page references and quotations come from the English translation by Ian McLeod, published in Untying the Text: A Poststructuralist Reader, ed. Robert Young (London: Routledge, 1981). Some notes on Foucault’s essay from Young’s introduction:
We find here a self-critical work, in which Foucault records his own progression from "archaeology" to "cartography"—a more directly political mapping of discourse practices. Discursive practices delimit fields of objects and perspectives of the agent of knowledge; they fix norms for concepts and theories. It is impossible to think outside them (it is the space of madness), and they are linked to power, control, domination. Discourse is defined as a play between desire and the institutions (comparable to the play between id and ego). Analysis separates the two. Procedures for mastery include exclusion and prohibition, or restrictive principles: rarefaction, disciplinarity, the imposition of roles. Foucault is against the notion of the founding, originating subject or experience in hermeneutics and in phenomenology. Against neo-Hegelianism, understood as the reduction of discursive practices to textual traces (against Derrida too!). He advocates a reversal of concepts: madness undermines philosophy and thought. He advocates the treatment of discourses as discontinuous practices, and is against the assumption of unities and regularities in works, epochs... We find at this stage of his thought the implication of a powerful determinism at work; later he will emphasize the possibility of resistance. As he writes in The History of Sexuality, "Discourse transmits and produces power; it reinforces it, but also undermines and exposes it, renders it fragile and makes it possible to thwart it."