The Archeology of Knowledge Michel Foucault contents part I introduction 3

4. RARITY, EXTERIORITY, ACCUMULATION

The enunci.ative analysis takes into consideration an element of rarity.

Generally speaking, the analysis of discourse operates between the twin poles of totality and plethora. One shows how the different texts with which one is dealing refer to one another, organize themselves into a single figure, converge with institutions and practices, and carry meanings that may be common to a whole period. Each element con-sidered is taken as the expression of the totality to which it belongs and whose limits it exceeds. And in this way one substitutes for the diver­sity of the things said a sort of great, uniform text, which has never before been articulated, and which reveals for the first time what men ^'really meant^' not only in their words and texts, their discourses and their writings, but also in the institutions, practices, techniques, and objects that they produced. In relation to this implicit, sovereign, communal ^'meaning^', statements appear in superabundant prolifer­ation, since it is to that meaning alone that they all refer and to it alone that they owe their truth: a plethora of signifying elements in relation to this single ^'signified^' (signifie). But this primary and ultimate mean-ing springs up through the manifest formulations, it hides beneath what appears, and secretly duplicates it, because each discourse

((134))

contains the power to say something other than what it actually says, and thus to embrace a plurality of meanings: a plethora of the ^'signi­fied^' in relation to a single ^'signifier^'. From this point of view, discourse is both plenitude and endless wealth.

The analysis of statements and discursive formations opens up a quite contrary direction: it wishes to determine the principle according to which only the ^'signifying^' groups that were enunciated could appear. It sets out to establish a law of rarity. This task involves several aspects:
—It is based on the principle that everything is never said; in relation to what might have been stated in a natural language (longue), in relation to the unlimited combination of linguistic elements, state­ments (however numerous they may be) are always in deficit; on the basis of the grammar and of the wealth of vocabulary available at a given period, there are, in total, relatively few things that are said. We must look therefore for the principle of rarification or at least of non-filling of the field of possible formulations as it is opened up by the language (longue). Discursive formation appears both as a principle of division in the entangled mass of discourses and as a principle of vacuity in the field of language (langage).
—We are studying statements at the limit that separates them from what is not said, in the occurrence that allows them to emerge to the exclusion of all others. Our task is not to give voice to the silence that surrounds them, nor to rediscover all that, in them and beside them, had remained silent or had been reduced to silence. Nor is it to study the obstacles that have prevented a particular discovery, held hack a particular formulation, repressed a particular form of enunciation, a particular unconscious meaning, or a particular rationality in the course of development; but to define a limited system of presences. The discursive formation is not therefore a developing totality, with its own dynamism or inertia, carrying with it, in an unformulated discourse, what it does not say, what it has not yet said, or what contradicts it at that moment; it is not a rich, difficult germination, it is a distribution of gaps, voids, absences, limits, divisions.

((135))

—However, we are not linking these ^'exclusions^' to a repression; we do not presuppose that beneath manifest statements something remains hidden and subjacent. We are analysing statements, not as being in the place of other statements that have fallen below the line of possible emergence, but as being always in their own place. They are put back into a space that is entirely deployed and involves no reduplication. There is no sub-text. And therefore no plethora. The enunciative domain is identical with its own surface. Each statement occupies in it a place that belongs to it alone. The description of a statement does not consist therefore in rediscovering the unsaid whose place it occupies; nor how one can reduce it to a silent, common text; but on the contrary in discovering what special place it occupies, what ramifications of the system of formations make it possible to map its localization, how it is isolated in the general dispersion of statements.
—This rarity of statements, the incomplete, fragmented form of the enunciative field, the fact that few things, in all, can be said, explain that statements are not, like the air we breathe, an infinite transparency; but things that are transmitted and preserved, that have value, and which one tries to appropriate; that are repeated, reproduced, and transformed; to which pre-established networks are adapted, and to which a status is given in the institution; things that are duplicated not only by copy or translation, but by exegesis, commentary, and the internal proliferation of meaning. Because statements are rare, they are collected in unifying totalities, and the meanings to be found in them are multiplied.
Unlike all those interpretations whose very existence is possible only through the actual rarity of statements, but which nevertheless ignore that rarity, and, on the contrary, take as their theme the compact rich­ness of what is said, the analysis of discursive formations turns back towards that rarity itself; it takes that rarity as its explicit object; it tries to determine its unique system; and, at the same time, it takes account of the fact that there could have been interpretation. To interpret is a way of reacting to enunciative poverty, and to compensate for it by a multiplication of meaning; a way of speaking on the basis of that poverty, and yet despite it. But to analyse a discursive formation is to

((136))

seek the law of that poverty, it is to weigh it up, and to determine its specific form. In one sense, therefore, it is to weigh the ^'value^' of statements. A value that is not defined by their truth, that is not gauged by the presence of a secret content; but which characterizes their place, their capacity for circulation and exchange, their possibility of trans-formation, not only in the economy of discourse, but, more generally, in the administration of scarce resources. In this sense, discourse ceases to be what it is for the exegetic attitude: an inexhaustible treasure from which one can always draw new, and always unpredictable riches; a providence that has always spoken in advance, and which enables one to hear, when one knows how to listen, retrospective oracles: it appears as an asset — finite, limited, desirable, useful — that has its own rules of appearance, but also its own conditions of appropriation and oper­ation; an asset that consequently, from the moment of its existence (and not only in its ^'practical applications^'), poses the question of power; an asset that is, by nature, the object of a struggle, a political struggle.

Another characteristic feature: the analysis of statements treats them in the systematic form of exteriority. Usually, the historical description of things said is shot through with the opposition of interior and exterior; and wholly directed by a desire to move from the exterior — which may be no more than contingency or mere material necessity, a visible body or uncertain translation — towards the essential nucleus of interiority. To undertake the history of what has been said is to re-do, in the opposite direction, the work of expression: to go back from statements preserved through time and dispersed in space, towards that interior secret that preceded them, left its mark in them, and (in every sense of the term) is betrayed by them. Thus the nucleus of the initiat­ing subjectivity is freed. A subjectivity that always lags behind manifest history; and which finds, beneath events, another, more serious, more secret, more fundamental history, closer to the origin, more firmly linked to its ultimate horizon (and consequently more in control of all its determinations). This other history, which runs beneath history, constantly anticipating it and endlessly recollecting the past, can be described — in a sociological or psychological way — as the evolution of mentalities; it can be given a philosophical status in the recollection of the Logos or the teleology of reason; lastly, it can be purified in

((137))

the problematic of a trace, which, prior to all speech, is the opening of inscription, the gap of deferred time, it is always the historico­transcendental theme that is reinvested.

A theme whose enunciative analysis tries to free itself. In order to restore statements to their pure dispersion. In order to analyse them in an exteriority that may be paradoxical since it refers to no adverse form of interiority. In order to consider them in their discontinuity, without having to relate them, by one of those shifts that disconnect them and render them inessential, to a more fundamental opening or difference. In order to seize their very irruption, at the place and at the moment at which it occurred. In order to rediscover their occurrence as an event. Perhaps we should speak of ^'neutrality^' rather than exteriority; but even this word implies rather too easily a suspension of belief, an effacement or a ^'placing in parentheses^' of all position of existence, whereas it is a question of rediscovering that outside in which, in their relative rarity, in their incomplete proximity, in their deployed space, enunciative events are distributed.
—This task presupposes that the field of statements is not described as a ^'translation^' of operations or processes that take place elsewhere (in men^'s thought, in their consciousness or unconscious, in the sphere of transcendental constitutions); but that it is accepted, in its empirical modesty, as the locus of particular events, regularities, relationships, modifications and systematic transformations; in short, that it is treated not as the result or trace of something else, but as a practical domain that is autonomous (although dependent) , and which can be described at its own level (although it must be articulated on something other than itself).
—It also presupposes that this enunciative domain refers neither to an individual subject, nor to some kind of collective consciousness, nor to a transcendental subjectivity; but that it is described as an anonym­ous field whose configuration defines the possible position of speaking subjects. Statements should no longer be situated in relation to a sover-eign subjectivity, but recognize in the different forms of the speaking subjectivity effects proper to the enunciative field.

((138))

As a result, it presupposes that, in its transformations, in its successive series, in its derivations, the field of statements does not obey the temporality of the consciousness as its necessary model. One must not hope — at least at this level and in this form of description — to be able to write a history of things said that is legitimately, in its form, in its regularity and in its nature, the history of an individual or anonymous consciousness, of a project, of a system of intentions, of a set of aims. The time of discourse is not the translation, in a visible chronology, of the obscure time of thought.
The analysis of statements operates therefore without reference to a cogito. It does not pose the question of the speaking subject, who reveals or who conceals himself in what he says, who, in speaking, exercises his sovereign freedom, or who, without realizing it, subjects himself to constraints of which he is only dimly aware. In fact, it is situated at the level of the ^'it is said^' — and we must not understand by this a sort of communal opinion, a collective representation that is imposed on every individual; we must not understand by it a great, anonymous voice that must, of necessity, speak through the discourses of everyone; but we must understand by it the totality of things said, the relations, the regularities, and the transformations that may be observed in them, the domain of which certain figures, certain inter-sections indicate the unique place of a speaking subject and may be given the name of author. ^`Anyone who speaks^', but what he says is not said from anywhere. It is necessarily caught up in the play of an exteriority.

The third feature of enunciative analysis: it is addressed to specific forms of accumulation that can be identified neither with an interior­ization in the form of memory nor with an undiscriminating totalization of documents. Usually, when one analyses already existing discourses, one regards them as having sprung from an essential iner­tia: they have survived vived through chance, or through the care with which men have treated them, and the illusions that they have enter­tained as to their value and the immortal dignity of their words; but now they are nothing more than written symbols piling up in dusty libraries, slumbering in a sleep towards which they have never ceased to glide since the day they were pronounced, since they were forgotten

((139))

and their visible effect lost in time. At most they may be lucky enough to be picked up and examined in some chance reading; at most they can discover that they bear the marks that refer back to the moment of their enunciation; at most, once these marks have been deciphered they can, by a sort of memory that moves across time, free meanings, thoughts, desires, buried fantasies. These four terms: reading — trace — decipherment — memory (however much importance one may accord to one or another of them, and whatever the metaphorical extent that one may accord it, and which enables it to embrace the other three) define the system that usually makes it possible to snatch past discourse from its inertia and, for a moment, to rediscover something of its lost vitality.

Now, the function of enunciative analysis is not to awaken texts from their present sleep, and, by reciting the marks still legible on their surface, to rediscover the flash of their birth; on the contrary, its func­tion is to follow them through their sleep, or rather to take up the related themes of sleep, oblivion, and lost origin, and to discover what mode of existence may characterize statements, independently of their enunciation, in the density of time in which they are preserved, in which they are reactivated, and used, in which they are also — but this was not their original destiny — forgotten, and possibly even destroyed.
—This analysis presupposes that statements are considered in the remanence (remanence) that is proper to them, and which is not that of an ever-realizable reference back to the past event of the formulation. To say that statements are residual (remanent) is not to say that they remain in the field of memory, or that it is possible to rediscover what they meant; but it means that they are preserved by virtue of a number of supports and material techniques (of which the book is, of course, only one example), in accordance with certain types of institutions (of which the library is one), and with certain statutory modalities (which are not the same in the case of a religious text, a law, or a scientific truth). This also means that they are invested in techniques that put them into operation, in practices that derive from them, in the social relations that they form, or, through those relations, modify. Lastly, it means that things do not have quite the same mode of existence, the

((140))

same system of relations with their environment, the same schemata of use, the same possibilities of transformation once they have been said. This survival in time is far from being the accidental or fortunate prolongation of an existence originally intended only for the moment; on the contrary, this remanence is of the nature of the statement; oblivion and destruction are in a sense only the zero degree of this remanence. And against the background that it constitutes, the oper­ations of memory can be deployed.
—This analysis also presupposes that statements are treated in the form of additivity that is specific to them. In fact, the types of grouping between successive statements are not always the same, and they never proceed by a simple piling-up or juxtaposition of successive elements. Mathematical statements are not added to one another in the same way as religious texts or laws (they each have their own way of merging together, annulling one another, excluding one another, complement-ing one another, forming groups that are in varying degrees indissoci­able and endowed with unique properties). Moreover, these forms of additivity are not given once and for all, and for a particular category of statements: medical case-history today forms a corpus of knowledge that does not obey the same laws of composition as medical case-history in the eighteenth century; modern mathematics does not accumulate its statements according to the same model as Euclidean geometry.
—Lastly, enunciative analysis presupposes that one takes phenom­ena of recurrence into account. Every statement involves a field of antecedent elements in relation to which it is situated, but which it is able to reorganize and redistribute according to new relations. It consti­tutes its own past, defines, in what precedes it, its own filiation, redefines what makes it possible or necessary, excludes what cannot be compatible with it. And it poses this enunciative past as an acquired truth, as an event that has occurred, as a form that can be modified, as material to be transformed, or as an object that can be spoken about, etc. In relation to all these possibilities of recurrence, memory and oblivion, the rediscovery of meaning or its repression, far from being fundamental, are merely unique figures.

((141))

The description of statements and discursive formations must there-fore free itself from the widespread and persistent image of return. It does not claim to go back, beyond a time that is no more than a falling off, a latency, an oblivion, a covering up or a wandering, towards that moment of foundation when speech was not yet caught up in any form of materiality, when it had no chances of survival, and when it was confined to the non-determined dimension of the opening. It does not try to constitute for the already said the paradoxical instant of the second birth; it does not invoke a dawn about to return. On the con­trary, it deals with statements in the density of the accumulation in which they are caught up and which nevertheless they never cease to modify, to disturb, to over-throw, and sometimes to destroy.

To describe a group of statements not as the closed, plethoric totality of a meaning, but as an incomplete, fragmented figure; to describe a group of statements not with reference to the interiority of an inten­tion, a thought, or a subject, but in accordance with the dispersion of an exteriority; to describe a group of statements, in order to rediscover not the moment or the trace of their origin, but the specific forms of an accumulation, is certainly not to uncover an interpretation, to discover a foundation, or to free constituent acts; nor is it to decide on a ration­ality, or to embrace a teleology. It is to establish what I am quite willing to call a positivity. To analyse a discursive formation therefore is to deal with a group of verbal performances at the level of the statements and of the form of positivity that characterizes them; or, more briefly, it is to define the type of positivity of a discourse. If, by substituting the analysis of rarity for the search for totalities, the description of relations of exteriority for the theme of the transcendental foundation, the analysis of accumulations for the quest of the origin, one is a positivist, then I am quite happy to be one. Similarly, I am not in the least unhappy about the fact that several times (though still in a rather blind way) I have used the term positivity to designate from afar the tangled mass that I was trying to unravel.

((142))