Michel Foucault - The Order of Things (1)

Preface

PART 1

CHAPTER I

Las Meninas

CHAPTER 2

The Prose of the World

I THE FOUR SIMILITUDES

II  SIGNATURES

III  THE LIMITS OF THE WORLD

IV THE WRITING OF THINGS

V THE BEING OF LANGUAGE

CHAPTER 3

Representing

I DON QUIXOTE

II  ORDER

Ill THE REPRESENTATION OF THE SIGN

IV DUPLICATED REPRESENTATION

V THE IMAGINATION OF RESEMBLANCE

VI MATHESIS AND ' TAXINOMIA'

CHAPTER 4

Speaking

I CRITICISM AND COMMENTARY

II  GENERAL GRAMMAR

III THE THEORY OF THE VERB

IV ARTICULATION

VI DERIVATION

VII THE QUADRILATERAL OF LANGUAGE

CHAPTER 5

Classifying

I WHAT THE HISTORIANS SAY

II NATURAL HISTORY

III  Structure

IV CHARACTER

V CONTINUITY AND CATASTROPHE

VI MONSTERS AND FOSSILS

CHAPTER 6

Exchanging

I THE ANALYSIS OF WEALTH

II MONEY AND PRICES

III MERCANTILISM

IV THE PLEDGE AND THE PRICE

V THE CREATION OF VALUE

VI UTILITY

VII GENERAL TABLE

VIII DESIRE AND REPRESENTATION

PART 2

CHAPTER 7

The Limits of Representation

I THE AGE OF HISTORY

II THE MEASURE OF LABOUR

III THE ORGANIC STRUCTURE OF BEINGS

IV WORD INFLECTION

V IDEOLOGY AND CRITICISM

VI OBJECTIVE SYNTHESES

CHAPTER 8

Labour, Life, Language

I THE NEW EMPIRICITIES

II  RICARDO

III CUVIER

IV BOPP

V LANGUAGE BECOME OBJECT

CHAPTER 9

Man and his Doubles

I     THE RETURN OF LANGUAGE

II THE PLACE OF THE KING

III THE ANALYTIC OF FINITUDE

IV THE EMPIRICAL AND THE TRANSCENDENTAL

V  THE 'COGITO' AND THE UNTHOUGHT

VI THE RETREAT AND RETURN OF THE ORIGIN

VII DISCOURSE AND MAN'S BEING

VIII THE ANTHROPOLOGICAL SLEEP

CHAPTER 10

The Human Sciences

I THE THREE FACES OF KNOWLEDGE

II THE FORM OF THE HUMAN SCIENCES

III THE THREE MODELS

IV HISTORY

V  PSYCHOANALYSE AND ETHNOLOGY

VI  IN CONCLUSION



VINTAGE BOOKS EDITION, APRIL 1994 Copyright © 1970 by Random House, Inc.

All rights reserved under International and Pan-American Copyright Conventions. Published in the United States by Vintage Books, a division of Random House, Inc., New York. Originally published in the United States by Pantheon Books, a division of Random House, Inc., in 1971, and in France as Les Mots et les chases by Editions Gallimard. Copyright © 1966 by Editions Gallimard, and in Great Britain by Tavistock Publications Limited.

Library of Congress Cataloging in Publication Data Foucault, Michel. The order of things. Translation of Les mots et les choses.

Reprint of the 1971 ed. published by Pantheon Books, New York, in series: World of man. Includes bibliographical references. i. Learning and scholarship. I. Title. [AZroi.F69l3 1973]    901.9    73-5618 ISBN 0-679-75335-4

Frontispiece photo © 1900 by Archivi Alinari

Manufactured in the United States of America 579B864



Contents of paperbook

Foreword to the English edition         ix

Preface                               xv

PART I

1 Las Meninas  3

2 The prose of the world               17

I The four similitudes     17

II Signatures     25

III The limits of the world             30

IV The writing of things 34

V The being of language 42

3 Representing 46

I Don Quixote  46

II Order             50

III The representation of the sign 58

IV Duplicated representation       63

V The imagination of resemblance             67

VI Mathesis and 'taxinomia’         71

4 Speaking        78

I Criticism and commentary         78

II General grammar        81

III The theory of the verb              92

IV Articulation 96

V Designation  104

VI Derivation   110

VII The quadrilateral of language 115

CONTENTS

5 Classifying 125

I What the historians say 125

II Natural history 128

III Structure 132

IV Character 138

V Continuity and catastrophe 145

VI Monsters and fossils 150

VII The discourse of nature 157



6 Exchanging 166

I The analysis of wealth 166

II Money and prices : 168

III Mercantilism 174

IV The pledge and the price 180

V The creation of value 189

VI Utility 196

VII General table 200

VIII Desire and representation 208

PART II

7  The limits of representation 217

I  The age of history 217

II The measure of labour 221

III The organic structure of beings 226

IV Word inflection 232

V Ideology and criticism 236

VI Objective syntheses 243



8 Labour, life, language 250

I The new empiricities 250

II Ricardo 253

III Cuvier 263

IV Bopp 280

V Language become object            294
vi

CONTENTS

9 Man and his doubles         303



I The return of language      303

II The place of the king       307

III The analytic of finitude   312

IV The empirical and the transcendental            318

V The 'cogito' and the unthought         322

VI The retreat and return of the origin               328

VII Discourse and man's being           335

VIII The anthropological sleep            340



10 The human sciences       344



I The three faces of knowledge          344

II The form of the human sciences     348

III The three models           355

IV History           367

V Psychoanalysis and ethnology         373

VI In conclusion  386

vii

Publisher's Note

A literal translation of the title of the French edition of this work (Les Mots et les choses) would have given rise to confusion with two other books that have already appeared under the title Words and things. The publisher therefore agreed with the author on the alternative title The order of things, which was, in fact, M. Foucault's original preference.

In view of the range of literature referred to in the text, it has not proved feasible in every case to undertake the bibliographical task of tracing English translations of works originating in other languages and locating the passages quoted by M. Foucault. The publisher has accordingly retained the author's references to French works and to French transla­tions of Latin and German works, for example, but has, as far as possible, cited English editions of works originally written in that language.

viii

Foreword to the English edition

This foreword should perhaps be headed 'Directions for Use'. Not be­cause I feel that the reader cannot be trusted - he is, of course, free to make what he will of the book he has been kind enough to read. What right have I, then, to suggest that it should be used in one way rather than another? When I was writing it there were many things that were not clear to me: some of these seemed too obvious, others too obscure. So I said to myself: this is how my ideal reader would have approached my book, if my intentions had been clearer and my project more ready to take form.

1. He would recognize that it was a study of a relatively neglected field. In France at least, the history of science and thought gives pride of place to mathematics, cosmology, and physics - noble sciences, rigorous sciences, sciences of the necessary, all close to philosophy: one can observe in their history the almost uninterrupted emergence of truth and pure reason. The other disciplines, however - those, for example, that concern living beings, languages, or economic facts-are considered too tinged with empirical thought, too exposed to the vagaries of chance or imagery, to age-old traditions and external events, for it to be supposed that their history could be anything other than irregular. At most, they are expected to provide evidence of a state of mind, an intellectual fashion, a mixture of archaism and bold conjecture, of intuition and blindness. But what if empirical knowledge, at a given time and in a given culture, did possess a well-defined regularity? If the very possibility of recording facts, of allow­ing oneself to be convinced by them, of distorting them in traditions or of making purely speculative use of them, if even this was not at the mercy of chance? If errors (and truths), the practice of old beliefs, including not only genuine discoveries, but also the most naive notions, obeyed, at a given moment, the laws of a certain code of knowledge? If, in

ix

FOREWORD TO THE ENGLISH EDITION

short, the history of non-formal knowledge had itself a system? That was my initial hypothesis - the first risk I took.

2. This book must be read as a comparative, and not a symptomatological, study. It was not my intention, on the basis of a particular type of knowledge or body of ideas, to draw up a picture of a period, or to reconstitute the spirit of a century. What I wished to do was to present, side by side, a definite number of elements: the knowledge of living beings, the knowledge of the laws of language, and the knowledge of economic facts, and to relate them to the philosophical discourse that was contemporary with them during a period extending from the seventeenth to the nineteenth century. It was to be not an analysis of Classicism in general, nor a search for a Weltanschauung, but a strictly 'regional' study.'

But, among other things, this comparative method produces results that are often strikingly different from those to be found in single-discipline studies. (So the reader must not expect to find here a history of biology juxtaposed with a history of linguistics, a history of political economy, and a history of philosophy.) There are shifts of emphasis: the calendar of saints and heroes is somewhat altered (Linnaeus is given more space than Buffon, Destutt de Tracy than Rousseau; the Physiocrats are opposed single-handed by Cantillon). Frontiers are redrawn and things usually far apart are brought closer, and vice versa: instead of relating the biological taxonomies to other knowledge of the living being (the theory of germination, or the physiology of animal movement, or the statics of plants), I have compared them with what might have been said at the same time about linguistic signs, the formation of general ideas, the lan­guage of action, the hierarchy of needs, and the exchange of goods.

This had two consequences: I was led to abandon the great divisions that are now familiar to us all. I did not look in the seventeenth and eighteenth centuries for the beginnings of nineteenth-century biology (or philosophy or economics). What I saw was the appearance of figures peculiar to the Classical age: a 'taxonomy' or 'natural history' that was relatively unaffected by the knowledge that then existed in animal or plant physiology; an 'analysis of wealth' that took little account of the assumptions of the 'political arithmetic' that was contemporary with it; and a 'general grammar' that was quite alien to the historical analyses and works of exegesis then being carried out. Epistemological figures, that is, that were not superimposed on the sciences as they were individualized

1 I sometimes use terms like 'thought' or 'Classical science', but they refer practically always to the particular discipline under consideration.

X

FOREWORD TO THE ENGLISH EDITION

and named in the nineteenth century. Moreover, I saw the emergence, between these different figures, of a network of analogies that transcended the traditional proximities: between the classification of plants and the theory of coinage, between the notion of generic character and the analysis of trade, one finds in the Classical sciences isomorphisms that appear to ignore the extreme diversity of the objects under consideration. The space of knowledge was then arranged in a totally different way from that systematized in the nineteenth century by Comte or Spencer. The second risk I took was in having wished to describe not so much the genesis of our sciences as an epistemological space specific to a particular period.

3. I did not operate, therefore, at the level that is usually that of the historian of science -1 should say at the two levels that are usually his. For, on the one hand, the history of science traces the progress of dis­covery, the formulation of problems, and the clash of controversy; it also analyses theories in their internal economy; in short, it describes the pro­cesses and products of the scientific consciousness. But, on the other hand, it tries to restore what eluded that consciousness: the influences that affected it, the implicit philosophies that were subjacent to it, the unformulated thematics, the unseen obstacles; it describes the unconscious of science. This unconscious is always the negative side of science - that which resists it, deflects it, or disturbs it. What I would like to do, how­ever, is to reveal a positive unconscious of knowledge: a level that eludes the consciousness of the scientist and yet is part of scientific discourse, instead of disputing its validity and seeking to diminish its scientific nature. What was common to the natural history, the economics, and the grammar of the Classical period was certainly not present to the consciousness of the scientist; or that part of it that was conscious was superficial, limited, and almost fanciful (Adanson, for example, wished to draw up an artificial denomination for plants; Turgot compared coinage with language); but, unknown to themselves, the naturalists, economists, and grammarians employed the same rules to define the objects proper to their own study, to form their concepts, to build their theories. It is these rules of formation, which were never formulated in their own right, but are to be found only in widely differing theories, concepts, and objects of study, that I have tried to reveal, by isolating, as their specific locus, a level that I have called, somewhat arbitrarily perhaps, archaeological. Taking as an example the period covered in this book, I have tried to determine the basis or arch­aeological system common to a whole series of scientific 'representations'

xi

FOREWORD TO THE ENGLISH EDITION

or 'products' dispersed throughout the natural history, economics, and philosophy of the Classical period.

4. I should like this work to be read as an open site. Many questions are laid out on it that have not yet found answers; and many of the gaps refer either to earlier works or to others that have not yet been completed, or even begun. But I should like to mention three problems.

The problem of change. It has been said that this work denies the very possibility of change. And yet my main concern has been with changes. In fact, two things in particular struck me: the suddenness and thorough­ness with which certain sciences were sometimes reorganized; and the fact that at the same time similar changes occurred in apparently very different disciplines. Within a few years (around 1800), the tradition of general grammar was replaced by an essentially historical philology; natural classifications were ordered according to the analyses of comparative anatomy; and a political economy was founded whose main themes were labour and production. Confronted by such a curious combination of phenomena, it occurred to me that these changes should be examined more closely, without being reduced, in the name of continuity, in either abruptness or scope. It seemed to me at the outset that different kinds of change were taking place in scientific discourse - changes that did not occur at the same level, proceed at the same pace, or obey the same laws;

the way in which, within a particular science, new propositions were pro­duced, new facts isolated, or new concepts built up (the events that make up the everyday life of a science) did not, in all probability, follow the same model as the appearance of new fields of study (and the frequently corresponding disappearance of old ones); but the appearance of new fields of study must not, in turn, be confused with those overall redis­tributions that alter not only the general form of a science, but also its relations with other areas of knowledge. It seemed to me, therefore, that all these changes should not be treated at the same level, or be made to culminate at a single point, as is sometimes done, or be attributed to the genius of an individual, or a new collective spirit, or even to the fecundity of a single discovery; that it would be better to respect such differences, and even to try to grasp them in their specificity. In this way I tried to describe the combination of corresponding transformations that char­acterized the appearance of biology, political economy, philology, a number of human sciences, and a new type of philosophy, at the threshold of the nineteenth century.

The problem of causality. It is not always easy to determine what has xii

FOREWORD TO THE ENGLISH EDITION

caused a specific change in a science. What made such a discovery possible? Why did this new concept appear? Where did this or that theory come from? Questions like these are often highly embarrassing because there are no definite methodological principles on which to base such an analysis. The embarrassment is much greater in the case of those general changes that alter a science as a whole. It is greater still in the case of several corresponding changes. But it probably reaches its highest point in the case of the empirical sciences: for the role of instruments, techniques, institutions, events, ideologies, and interests is very much in evidence; but one does not know how an articulation so complex and so diverse in composition actually operates. It seemed to me that it would not be prudent for the moment to force a solution I felt incapable, I admit, of offering: the traditional explanations - spirit of the time, technological or social changes, influences of various kinds - struck me for the most part as being more magical than effective. In this work, then, I left the problem of causes to one side;1 I chose instead to confine myself to describing the transformations themselves, thinking that this would be an indispensable step if, one day, a theory of scientific change and epistemological causality was to be constructed.

The problem of the subject. In distinguishing between the epistemo­logical level of knowledge (or scientific consciousness) and the archaeo­logical level of knowledge, I am aware that I am advancing in a direction that is fraught with difficulty. Can one speak of science and its history (and therefore of its conditions of existence, its changes, the errors it has per­petrated, the sudden advances that have sent it off on a new course) with­out reference to the scientist himself- and I am speaking not merely of the concrete individual represented by a proper name, but of his work and the particular form of his thought? Can a valid history of science be attempted that would retrace from beginning to end the whole spontane­ous movement of an anonymous body of knowledge? Is it legitimate, is it even useful, to replace the traditional 'X thought that. . .' by a 'it was known that. . .'? But this is not exactly what I set out to do. I do not wish to deny the validity of intellectual biographies, or the possibility of a history of theories, concepts, or themes. It is simply that I wonder whether such descriptions are themselves enough, whether they do justice to the immense density of scientific discourse, whether there do not exist, outside their customary boundaries, systems of regularities that have a decisive

--------------------

1 I had approached this question in connection with psychiatry and clinical medicine in two earlier works.

xiii

FORBWORD TO THE BNGLISH EDITION

role in the history of the sciences. I should like to know whether the subjects responsible for scientific discourse are not determined in their situation, their function, their perceptive capacity, and their practical possibilities by conditions that dominate and even overwhelm them. In short, I tried to explore scientific discourse not from the point of view of the individuals who are speaking, nor from the point of view of the formal structures of what they are saying, but from the point of view of the rules that come into play in the very existence of such discourse: what conditions did Linnaeus (or Petty, or Arnauld) have to fulfil, not to make his discourse coherent and true in general, but to give it, at the time when it was written and accepted, value and practical application as scientific discourse - or, more exactly, as naturalist, economic, or gram­matical discourse?

On this point, too, I am well aware that I have not made much progress. But I should not like the effort I have made in one direction to be taken as a rejection of any other possible approach. Discourse in general, and scientific discourse in particular, is so complex a reality that we not only can, but should, approach it at different levels and with different methods. If there is one approach that I do reject, however, it is that (one might call it, broadly speaking, the phenomenological approach) which gives absolute priority to the observing subject, which attributes a constituent role to an act, which places its own point of view at the origin of all his­toricity - which, in short, leads to a transcendental consciousness. It seems to me that the historical analysis of scientific discourse should, in the last resort, be subject, not to a theory of the knowing subject, but rather to a theory of discursive practice.

5. This last point is a request to the English-speaking reader. In France, certain half-witted 'commentators' persist in labelling me a 'structuralist'. I have been unable to get it into their tiny minds that I have used none of the methods, concepts,, or key terms that characterize structural analysis.

I should be grateful if a more serious public would free me from a connection that certainly does me honour, but that I have not deserved. There may well be certain similarities between the works of the struc­turalists and my own work. It would hardly behove me, of all people, to claim that my discourse is independent of conditions and rules of which I am very largely unaware, and which determine other work that is being done today. But it is only too easy to avoid the trouble of analysing such work by giving it an admittedly impressive-sounding, but inaccurate, label.

xiv

Preface

This book first arose out of a passage in Borges, out of the laughter that shattered, as I read the passage, all the familiar landmarks of my thought - our thought, the thought that bears the stamp of our age and our geography - breaking up all the ordered surfaces and all the planes with which we are accustomed to tame the wild profusion of existing things, and continuing long afterwards to disturb and threaten with collapse our age-old distinction between the Same and the Other. This passage quotes a 'certain Chinese encyclopaedia' in which it is written that 'animals are divided into: (a) belonging to the Emperor, (b) embalmed, (c) tame, (d) sucking pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush, (1) et cetera, (m) having just broken the water pitcher, (n) that from a long way off" look like flies'. In the wonderment of this taxonomy, the thing we apprehend in one great leap, the thing that, by means of the fable, is demonstrated as the exotic charm of another system of thought, is the limitation of our own, the stark impossibility of thinking that.

But what is it impossible to think, and what kind of impossibility are we faced with here? Each of these strange categories can be assigned a precise meaning and a demonstrable content; some of them do certainly involve fantastic entities - fabulous animals or sirens - but, precisely be­cause it puts them into categories of their own, the Chinese encyclopaedia localizes their powers of contagion; it distinguishes carefully between the very real animals (those that are frenzied or have just broken the water pitcher) and those that reside solely in the realm of imagination. The possibility of dangerous mixtures has been exorcized, heraldry and fable have been relegated to their own exalted peaks: no inconceivable amphibi­ous maidens, no clawed wings, no disgusting, squamous epidermis, none xv

PREFACE

of those polymorphous and demoniacal faces, no creatures breathing fire. The quality of monstrosity here does not affect any real body, nor does it produce modifications of any kind in the bestiary of the imagination; it does not lurk in the depths of any strange power. It would not even be present at all in this classification had it not insinuated itself into the empty space, 'the interstitial blanks separating all these entities from one another. It is not the 'fabulous' animals that are impossible, since they are desig­nated as such, but the narrowness of the distance separating them from (and juxtaposing them to) the stray dogs, or the animals that from a long way off look like flies. What transgresses the boundaries of all imagina­tion, of all possible thought, is simply that alphabetical series (a, b, c, d) which links each of those categories to all the others.

Moreover, it is not simply the oddity of unusual Juxtapositions that we are faced with here. We are all familiar with the disconcerting effect of the proximity of extremes, or, quite simply, with the sudden vicinity of things that have no relation to each other; the mere act of enumeration that heaps them all together has a power of enchantment all its own: 'I am no longer hungry,' Eusthenes said. 'Until the morrow, safe from my saliva all the following shall be: Aspics, Acalephs, Acanthocephalates, Amoebocytes, Ammonites, Axolotis, Amblystomas, Aphislions, Anacon­das, Ascarids, Amphisbaenas, Angleworms, Amphipods, Anaerobes, Anne­lids, Anthozoans. . . .' But all these worms and snakes, all these creatures redolent of decay and slime are slithering, like the syllables which desig­nate them, in Eusthenes' saliva: that is where they all have their common locus, like the umbrella and the sewing-machine on the operating table;

startling though their propinquity may be, it is nevertheless warranted by that and, by that in, by that on whose solidity provides proof of the possibility of juxtaposition. It was certainly improbable that arachnids, ammonites, and annelids should one day mingle on Eusthenes' tongue, but, after all, that welcoming and voracious mouth certainly provided them with a feasible lodging, a roof under which to coexist.

The monstrous quality that runs through Borges's enumeration con­sists, on the contrary, in the fact that the common ground on which such meetings are possible has itself been destroyed. What is impossible is not the propinquity of the things listed, but the very site on which their propinquity would be possible. The animals '(i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush' - where could they ever meet, except in the immaterial sound of the voice pronouncing their enumeration, or on the page transcribing it? Where else could they be

xvi

PREFACE

juxtaposed except in the non-place of language? Yet, though language can spread them before us, it can do so only in an unthinkable space. The central category of animals 'included in the present classification', with its explicit reference to paradoxes we are familiar with, is indication enough that we shall never succeed in defining a stable relation of con­tained to container between each of these categories and that which includes them all: if all the animals divided up here can be placed without exception in one of the divisions of this list, then aren't all the other divisions to be found in that one division too? And then again, in what space would that single, inclusive division have its existence? Absurdity destroys the and of the enumeration by making impossible the in where the things enumerated would be divided up. Borges adds no figure to the atlas of the impossible; nowhere does he strike the spark of poetic con­frontation; he simply dispenses with the least obvious, but most com­pelling, of necessities; he does away with the site, the mute ground upon which it is possible for entities to be juxtaposed. A vanishing trick that is masked or, rather, laughably indicated by our alphabetical order, which is to be taken as the clue (the only visible one) to the enumerations of a Chinese encyclopaedia... . What has been removed, in short, is the famous 'operating table'; and rendering to Roussel1 a small part of what is still his due, I use that word 'table' in two superimposed senses: the nickel-plated, rubbery table swathed in white, glittering beneath a glass sun devouring all shadow - the table where, for an instant, perhaps for­ever, the umbrella encounters the sewing-machine; and also a table, a tabula, that enables thought to operate upon the entities of our world, to put them in order, to divide them into classes, to group them according to names that designate their similarities and their differences - the table upon which, since the beginning of time, language has intersected space.

That passage from Borges kept me laughing a long time, though not without a certain uneasiness that I found hard to shake off. Perhaps be­cause there arose in its wake the suspicion that there is a worse kind of disorder than that of the incongruous, the linking together of things that are inappropriate; I mean the disorder in which fragments of a large number of possible orders glitter separately in the dimension, without law or geometry, of the heteroclite; and that word should be taken in its most literal, etymological sense: in such a state, things are 'laid', 'placed', 'arranged' in sites so very different from one another that it is impossible

---------------

1 Raymond Roussel, the French novelist. Cf. Michel Foucault's Raymond Roussel (Paris, I963). [Translator's note.]

xvii

PREFACE

to find a place of residence for them, to define a common locus beneath them all. Utopias afford consolation: although they have no real locality there is nevertheless a fantastic, untroubled region in which they are able to unfold; they open up cities with vast avenues, superbly planted gardens, countries where life is easy, even though the road to them is chimerical. Heterotopias are disturbing, probably because they secretly undermine language, because they make it impossible to name this and that, because they shatter or tangle common names, because they destroy 'syntax' in advance, and not only the syntax with which we construct sentences but also that less apparent syntax which causes words and things (next to and also opposite one another) to 'hold together'. This is why Utopias permit fables and discourse: they run with the very grain of language and are part of the fundamental dimension of the fabula; heterotopias (such as those to be found so often in Borges) desiccate speech, stop words in their tracks, contest the very possibility of grammar at its source; they dissolve our myths and sterilize the lyricism of our sentences.

It appears that certain aphasiacs, when shown various differently coloured skeins of wool on a table top, are consistently unable to arrange them into any coherent pattern; as though that simple rectangle were unable to serve in their case as a homogeneous and neutral space in which things could be placed so as to display at the same time the continuous order of their identities or differences as well as the semantic field of their denomination. Within this simple space in which things are normally arranged and given names, the aphasiac will create a multiplicity of tiny, fragmented regions in which nameless resemblances agglutinate things into unconnected islets; in one corner, they will place the lightest-coloured skeins, in another the red ones, somewhere else those that are softest in texture, in yet another place the longest, or those that have a tinge of purple or those that have been wound up into a ball. But no sooner have they been adumbrated than all these groupings dissolve again, for the field of identity that sustains them, however limited it may be, is still too wide not to be unstable; and so the sick mind continues to infinity, creating groups then dispersing them again, heaping up diverse similari­ties, destroying those that seem clearest, splitting up things that are identical, superimposing different criteria, frenziedly beginning all over again, becoming more and more disturbed, and teetering finally on the brink of anxiety.

The uneasiness that makes us laugh when we read Borges is certainly related to the profound distress of those whose language has been

xviii

PREFACE

destroyed: loss of what is 'common' to place and name. Atopia, aphasia. Yet our text from Borges proceeds in another direction; the mythical homeland Borges assigns to that distortion of classification that prevents us from applying it, to that picture that lacks all spatial coherence, is a precise region whose name alone constitutes for the West a vast reservoir of Utopias. In our dreamworld, is not China precisely this privileged site of space? In our traditional imagery, the Chinese culture is the most meti­culous, the most rigidly ordered, the one most deaf to temporal events, most attached to the pure delineation of space; we think of it as a civiliza­tion of dikes and dams beneath the eternal face of the sky; we see it, spread and frozen, over the entire surface of a continent surrounded by walls. Even its writing does not reproduce the fugitive flight of the voice in horizontal lines; it erects the motionless and still-recognizeable images of things themselves in vertical columns. So much so that the Chinese encyclopaedia quoted by Borges, and the taxonomy it proposes, lead to a kind of thought without space, to words and categories that lack all life and place, but are rooted in a ceremonial space, overburdened with com­plex figures, with tangled paths, strange places, secret passages, and unex­pected communications. There would appear to be, then, at the other extremity of the earth we inhabit, a culture entirely devoted to the order­ing of space, but one that does not distribute the multiplicity of existing things into any of the categories that make it possible for us to name, speak, and think.

When we establish a considered classification, when we say that a cat and a dog resemble each other less than two greyhounds do, even if both are tame or embalmed, even if both are frenzied, even if both have just broken the water pitcher, what is the ground on which we are able to establish the validity of this classification with complete certainty? On what 'table', according to what grid of identities, similitudes, analogies, have we become accustomed to sort out so many different and similar things? What is this coherence - which, as is immediately apparent, is neither determined by an a priori and necessary concatenation, nor im­posed on us by immediately perceptible contents? For it is not a question of linking consequences, but of grouping and isolating, of analysing, of matching and pigeon-holing concrete contents; there is nothing more tentative, nothing more empirical (superficially, at least) than the process of establishing an order among things; nothing that demands a sharper eye or a surer, better-articulated language; nothing that more insistently requires that one allow oneself to be carried along by the proliferation of

xix

PREFACE

qualities and forms. And yet an eye not consciously prepared might well group together certain similar figures and distinguish between others on the basis of such and such a difference: in fact, there is no similitude and no distinction, even for the wholly untrained perception, that is not the result of a precise operation and of the application of a preliminary criterion. A 'system of elements' - a definition of the segments by which the resemblances and differences can be shown, the types of variation by which those segments can be affected, and, lastly, the threshold above which there is a difference and below which there is a similitude - is indispensable for the establishment of even the simplest form of order. Order is, at one and the same time, that which is given in things as their inner law, the hidden network that determines the way they confront one another, and also that which has no existence except in the grid created by a glance, an examination, a language; and it is only in the blank spaces of this grid that order manifests itself in depth as though already there, waiting in silence for the moment of its expression.

The fundamental codes of a culture - those governing its language, its schemas of perception, its exchanges, its techniques, its values, the hier­archy of its practices - establish for every man, from the very first, the empirical orders with which he will be dealing and within which he will be at home. At the other extremity of thought, there are the scientific theories or the philosophical interpretations which explain why order exists in general, what universal law it obeys, what principle can account for it, and why this particular order has been established and not some other. But between these two regions, so distant from one another, lies a domain which, even though its role is mainly an intermediary one, is nonetheless fundamental: it is more confused, more obscure, and prob­ably less easy to analyse. It is here that a culture, imperceptibly deviating from the empirical orders prescribed for it by its primary codes, instituting an initial separation from them, causes them to lose their original trans­parency, relinquishes its immediate and invisible powers, frees itself sufficiently to discover that these orders are perhaps not the only possible ones or the best ones; this culture then finds itself faced with the stark fact that there exists, below the level of its spontaneous orders, things that are in themselves capable of being ordered, that belong to a certain unspoken order; the fact, in short, that order exists. As though emanci­pating itself to some extent from its linguistic, perceptual, and practical grids, the culture superimposed on them another kind of grid which neutralized them, which by this superimposition both revealed and ex-

xx

PREFACE

cluded them at the same time, so that the culture, by this very process, came face to face with order in its primary state. It is on the basis of this newly perceived order that the codes of language, perception, and prac­tice are criticized and rendered partially invalid. It is on the basis of this order, taken as a firm foundation, that general theories as to the ordering of things, and the interpretation that such an ordering involves, will be constructed. Thus, between the already 'encoded' eye and reflexive know­ledge there is a middle region which liberates order itself: it is here that it appears, according to the culture and the age in question, continuous and graduated or discontinuous and piecemeal, linked to space or constituted anew at each instant by the driving force of time, related to a series of variables or defined by separate systems of coherences, composed of re­semblances which are either successive or corresponding, organized around increasing differences, etc. This middle region, then, in so far as it makes manifest the modes of being of order, can be posited as the most fundamental of all: anterior to words, perceptions, and gestures, which are then taken to be more or less exact, more or less happy, ex­pressions of it (which is why this experience of order in its pure primary state always plays a critical role); more solid, more archaic, less dubious, always more 'true' than the theories that attempt to give those expressions explicit form, exhaustive application, or philosophical foundation. Thus, in every culture, between the use of what one might call the ordering codes and reflections upon order itself, there is the pure experience of order and of its modes of being.

The present study is an attempt to analyse that experience. I am con­cerned to show its developments, since the sixteenth century, in the main­stream of a culture such as ours: in what way, as one traces - against the current, as it were - language as it has been spoken, natural creatures as they have been perceived and grouped together, and exchanges as they have been practised; in what way, then, our culture has made manifest the existence- of order, and how, to the modalities of that order, the exchanges owed their laws, the living beings their constants, the words their sequence and their representative value; what modalities of order have been recognized, posited, linked with space and time, in order to create the positive basis of knowledge as we find it employed in grammar and philology, in natural history and biology, in the study of wealth and political economy. Quite obviously, such an analysis does not belong to the history of ideas or of science: it is rather an inquiry whose aim is to rediscover on what basis knowledge and theory became possible; within

xxi

PREFACE

what space of order knowledge was constituted; on the basis of what historical a priori, and in the element of what positivity, ideas could appear, sciences be established, experience be reflected in philosophies, rationalities be formed, only, perhaps, to dissolve and vanish soon after­wards. I am not concerned, therefore, to describe the progress of know­ledge towards an objectivity in which today's science can finally be recog­nized; what I am attempting to bring to light is the epistemological field, the episteme in which knowledge, envisaged apart from all criteria having reference to its rational value or to its objective forms, grounds its positivity and thereby manifests a history which is not that of its growing perfection, but rather that of its conditions of possibility; in this account, what should appear are those configurations within the space of know­ledge which have given rise to the diverse forms of empirical science. Such an enterprise is not so much a history, in the traditional meaning of that word, as an 'archaeology'.1

Now, this archaeological inquiry has revealed two great discontinuities in the episteme of Western culture: the first inaugurates the Classical age (roughly half-way through the seventeenth century) and the second, at the beginning of the nineteenth century, marks the beginning of the modem age. The order on the basis of which we think today does not have the same mode of being as that of the Classical thinkers. Despite the impression we may have of an almost uninterrupted development of the European ratio from the Renaissance to our own day, despite our possible belief that the classifications of Linnaeus, modified to a greater or lesser degree, can still lay claim to some sort of validity, that Condillac's theory of value can be recognized to some extent in nineteenth-century mar­ginalism, that Keynes was well aware of the affinities between his own analyses and those of Cantillon, that the language of general grammar (as exemplified in the authors of Port-Royal or in Bauzee) is not so very far removed from our own - all this quasi-continuity on the level of ideas and themes is doubtless only a surface appearance; on the archaeological level, we see that the system of positivities was transformed in a whole­sale fashion at the end of the eighteenth and beginning of the nineteenth century. Not that reason made any progress: it was simply that the mode of being of things, and of the order that divided them up before present­ing them to the understanding, was profoundly altered. If the natural history of Tournefort, Linnaeus, and Button can be related to anything

1 The problems of method raised by such an 'archaeology' will be examined in a later work.

xxii

PREFACE

at all other than itself, it is not to biology, to Cuvier's comparative anatomy, or to Darwin's theory of evolution, but to Bauzee's general grammar, to the analysis of money and wealth as found in the works of Law, or Veron de Fortbonnais, or Turgot. Perhaps knowledge succeeds in engendering knowledge, ideas in transforming themselves and actively modifying one another (but how? - historians have not yet enlightened us on this point); one thing, in any case, is certain: archaeology, addressing itself to the general space of knowledge, to its configurations, and to the mode of being of the things that appear in it, defines systems of simul­taneity, as well as the series of mutations necessary and sufficient to circumscribe the threshold of a new positivity.

In this way, analysis has been able to show the coherence that existed, throughout the Classical age, between the theory of representation and the theories of language, of the natural orders, and of wealth and value. It is this configuration that, from the nineteenth century onward, changes entirely; the theory of representation disappears as the universal founda­tion of all possible orders; language as the spontaneous tabula, the primary grid of things, as an indispensable link between representation and things, is eclipsed in its turn; a profound historicity penetrates into the heart of things, isolates and defines them in their own coherence, imposes upon them the forms of order implied by the continuity of time; the analysis of exchange and money gives way to the study of production, that of the organism takes precedence over the search for taxonomic characteristics, and, above all, language loses its privileged position and becomes, in its turn, a historical form coherent with the density of its own past. But as things become increasingly reflexive, seeking the principle of their intelli­gibility only in their own development, and abandoning the space of representation, man enters in his turn, and for the first time, the field of Western knowledge. Strangely enough, man - the study of whom is supposed by the naive to be the oldest investigation since Socrates - is prob­ably no more than a kind of rift in the order of things, or, in any case, a configuration whose outlines are determined by the new position he has so recently taken up in the field of knowledge. Whence all the chimeras of the new humanisms, all the facile solutions of an 'anthropology' under­stood as a universal reflection on man, half-empirical, half-philosophical. It is comforting, however, and a source of profound relief to think that man is only a recent invention, a figure not yet two centuries old, a new wrinkle in our knowledge, and that he will disappear again as soon as that knowledge has discovered a new form.

xxiii

PREFACE

It is evident that the present study is, in a sense, an echo of my under­taking to write a history of madness in the Classical age; it has the same articulations in time, taking the end of the Renaissance as its starting-point, then encountering, at the beginning of the nineteenth century, just as my history of madness did, the threshold of a modernity that we have not yet left behind. But whereas in the history of madness I was investi­gating the way in which a culture can determine in a massive, general form the difference that limits it, I am concerned here with observing how a culture experiences the propinquity of things, how it establishes the tabula of their relationships and the order by which they must be con­sidered. I am concerned, in short, with a history of resemblance: on what conditions was Classical thought able to reflect relations of similarity or equivalence between things, relations that would provide a foundation and a justification for their words, their classifications, their systems of exchange? What historical a priori provided the starting-point from which it was possible to define the great checkerboard of distinct identities established against the confused, undefined, faceless, and, as it were, indifferent background of differences? The history of madness would be the history of the Other - of that which, for a given culture, is at once interior and foreign, therefore to be excluded (so as to exorcize the interior danger) but by being shut away (in order to reduce its otherness);

whereas the history of the order imposed on things would be the history of the Same - of that which, for a given culture, is both dispersed and related, therefore to be distinguished by kinds and to be collected together into identities.

And if one considers that disease is at one and the same time disorder -the existence of a perilous otherness within the human body, at the very heart of life - and a natural phenomenon with its own constants, resem­blances, and types, one can see what scope there would be for an archaeo­logy of the medical point of view. From the limit-experience of the Other to the constituent forms of medical knowledge, and from the latter to the order of things and the conceptions of the Same, what is available to archaeological analysis is the whole of Classical knowledge, or rather the threshold that separates us from Classical thought and constitutes our modernity. It was upon this threshold that the strange figure of knowledge called man first appeared and revealed a space proper to the human sciences. In attempting to uncover the deepest strata of Western culture, I am restor­ing to our silent and apparently immobile soil its rifts, its instability, its flaws; and it is the same ground that is once more stirring under our feet.

xxiv


PART 1

CHAPTER I
Las Meninas

1

The painter is standing a little back from his canvas [1]. He is glancing at his model; perhaps he is considering whether to add some finishing touch, though it is also possible that the first stroke has not yet been made. The arm holding the brush is bent to the left, towards the palette; it is motionless, for an instant, between canvas and paints. The skilled hand is suspended in mid-air, arrested in rapt attention on the painter's gaze; and the gaze, in return, waits upon the arrested gesture. Between the fine point of the brush and the steely gaze, the scene is about to yield up its volume.

But not without a subtle system of feints. By standing back a little, the painter has placed himself to one side of the painting on which he is working. That is, for the spectator'at present observing him he is to the right of his canvas, while the latter, the canvas, takes up the whole of the extreme left. And the canvas has its back turned to that spectator: he can see nothing of it but the reverse side, together with the huge frame on which it is stretched. The painter, on the other hand, is perfectly visible in his full height; or at any rate, he is not masked by the tall canvas which may soon absorb him, when, taking a step towards it again, he returns to his task; he has no doubt just appeared, at this very instant, before the eyes of the spectator, emerging from what is virtually a sort of vast cage pro­jected backwards by the surface he is painting. Now he can be seen, caught in a moment of stillness, at the neutral centre of this oscillation. His dark torso and bright face are half-way between the visible and the invisible: emerging from that canvas beyond our view, he moves into our gaze; but when, in a moment, he makes a step to the right, removing himself from our gaze, he will be standing exactly in front of the canvas he is painting; he will enter that region where his painting, neglected for an instant, will, for him, become visible once more, free of shadow and

3

THB ORDER OF THINGS

free of reticence. As though the painter could not at the same time be seen on the picture where he is represented and also see that upon which he is representing something. He rules at the threshold of those two in­compatible visibilities.

The painter is looking, his face turned slightly and his head leaning towards one shoulder. He is staring at a point to which, even though it is invisible, we, the spectators, can easily assign an object, since it is we, our­selves, who are that point: our bodies, our faces, our eyes. The spectacle he is observing is thus doubly invisible: first, because it is not represented within the space of the painting, and, second, because it is situated pre­cisely in that blind point, in that essential hiding-place into which our gaze disappears from ourselves at the moment of our actual looking. And yet, how could we fail to see that invisibility, there in front of our eyes, since it has its own perceptible equivalent, its sealed-in figure, in the painting itself? We could, in effect, guess what it is the painter is looking at if it were possible for us to glance for a moment at the canvas he is working on; but all we can see of that canvas is its texture, the horizontal and vertical bars of the stretcher, and the obliquely rising foot of the easel. The tall, monotonous rectangle occupying the whole left portion of the real picture, and representing the back of the canvas within the picture, reconstitutes in the form of a surface the invisibility in depth of what the artist is observing: that space in which we are, and which we are. From the eyes of the painter to what he is observing there runs a compelling line that we, the onlookers, have no power of evading: it runs through the real picture and emerges from its surface to join the place from which we see the painter observing us; this dotted line reaches out to us ineluctably, and links us to the representation of the picture. In appearance, this locus is a simple one; a matter of pure reciprocity:

we are looking at a picture in which the painter is in turn looking out at us. A mere confrontation, eyes catching one another's glance, direct looks superimposing themselves upon one another as they cross. And yet this slender line of reciprocal visibility embraces a whole complex net­work of uncertainties, exchanges, and feints. The painter is turning his eyes towards us only in so far as we happen to occupy the same position as his subject. We, the spectators, are an additional factor. Though greeted by that gaze, we are also dismissed by it, replaced by that which was always there before we were: the model itself. But, inversely, the painter's gaze, addressed to the void confronting him outside the picture, accepts as many models as there are spectators; in this precise but neutral place, the observer 4

LAS MENINAS

and the observed take part in a ceaseless exchange. No gaze is stable, or rather in the neutral furrow of the gaze piercing at a right angle through the canvas, subject and object, the spectator and the model, reverse their roles

infinity. And here the great canvas with its back to us on the extreme left of the picture exercises its second function: stubbornly invisible, it prevents the relation of these gazes from ever being discoverable or de­finitely established. The opaque fixity that it establishes on one side renders forever unstable the play of metamorphoses established in the centre be­tween spectator and model. Because we can see only that reverse side, we do not know who we are, or what we are doing. Seen or seeing? The painter is observing a place which, from moment to moment, never ceases to change its content, its form, its face, its identity. But the attentive immobility of his eyes refers us back to another direction which they have often followed already, and which soon, there can be no doubt, they will take again: that of the motionless canvas upon which is being traced, has already been traced perhaps, for a long time and forever, a portrait that will never again be erased. So that the painter's sovereign gaze commands a virtual triangle whose outline defines this picture of a picture: at the top - the only visible corner - the painter's eyes; at one of the base angles, the invisible place occupied by the model; at the other base angle, the figure probably sketched out on the invisible surface of the canvas.

As soon as they place the spectator in the field of their gaze, the painter's eyes seize hold of him, force him to enter the picture, assign him a place at once privileged and inescapable, levy their luminous and visible tribute from him, and project it upon the inaccessible surface of the canvas within the picture. He sees his invisibility made visible to the painter and trans­posed into an image forever invisible to himself. A shock that is aug­mented and made more inevitable still by a marginal trap. At the extreme right, the picture is lit by a window represented in very sharp perspective; so sharp that we can see scarcely more than the embrasure; so that the flood of light streaming through it bathes at the same time, and with equal generosity, two neighboring spaces, overlapping but irreducible: the surface of the painting, together with the volume it represents (which is to say, the painter's studio, or the salon in which his easel is now set up), and, in front of that surface, the real volume occupied by the spectator (or again, the unreal site of the model). And as it passes through the room from right to left, this vast flood of golden light carries both the spectator towards the painter and the model towards the canvas; it is this light too, which, washing over the painter, makes him visible to the spectator and

5

THE ORDER OF THINGS

turns into golden lines, in the model's eyes, the frame of that enigmatic canvas on which his image, once transported there, is to be imprisoned. This extreme, partial, scarcely indicated window frees a whole flow of daylight which serves as the common locus of the representation. It balances the invisible canvas on the other side of the picture: just as that canvas, by turning its back to the spectators, folds itself in against the picture representing it, and forms, by the superimposition of its reverse and visible side upon the surface of the picture depicting it, the ground, inaccessible to us, on which there shimmers the Image par excellence, so does the window, a pure aperture, establish a space as manifest as the other is hidden; as much the common ground of painter, figures, models, and spectators, as the other is solitary (for no one is looking at it, not even the painter). From the right, there streams in through an invisible window the pure volume of a light that renders all representation visible; to the left extends the surface that conceals, on the other side of its all too visible woven texture, the representation it bears. The light, by flooding the scene (I mean the room as well as the canvas, the room represented on the canvas, and the room in which the canvas stands), envelops the figures and the spectators and carries them with it, under the painter's gaze, towards the place where his brush will represent them. But that place is concealed from us. We are observing ourselves being observed by the painter, and made visible to his eyes by the same light that enables us to see him. And just as we are about to apprehend ourselves, transcribed by his hand as though in a mirror, we find that we can in fact apprehend nothing of that mirror but its lustreless back. The other side of a psyche.

Now, as it happens, exactly opposite the spectators - ourselves - on the wall forming the far end of the room, Velazquez has represented a series of pictures; and we see that among all those hanging canvases there is one that shines with particular brightness. Its frame is wider and darker than those of the others; yet there is a fine white line around its inner edge diffusing over its whole surface a light whose source is not easy to deter­mine; for it comes from nowhere, unless it be from a space within itself. In this strange light, two silhouettes are apparent, while above them, and a little behind them, is a heavy purple curtain. The other pictures reveal little more than a few paler patches buried in a darkness without depth. This particular one, on the other hand, opens onto a perspective of space in which recognizable forms recede from us in a light that belongs only to itself. Among all these elements intended to provide representations, while impeding them, hiding them, concealing them because of their

6

LAS MENINAS

position or their distance from us, this is the only one that fulfils its function in all honesty and enables us to see what it is supposed to show. Despite its distance from us, despite the shadows all around it. But it isn't a picture: it is a mirror. It offers us at last that enchantment of the double that until now has been denied us, not only by the distant paintings but also by the light in the foreground with its ironic canvas.

Of all the representations represented in the picture this is the only one visible; but no one is looking at it. Upright beside his canvas, his attention entirely taken up by his model, the painter is unable to see this looking-glass shining so softly behind him. The other figures in the picture are also, for the most part, turned to face what must be taking place in front -towards the bright invisibility bordering the canvas, towards that balcony of light where their eyes can gaze at those who are gazing back at them, and not towards that dark recess which marks the far end of the room in which they are represented. There are, it is true, some heads turned away from us in profile: but not one of them is turned far enough to see, at the back of the room, that solitary mirror, that tiny glowing rectangle which is nothing other than visibility, yet without any gaze able to grasp it, to render it actual, and to enjoy the suddenly ripe fruit of the spectacle it offers.

It must be admitted that this indifference is equalled only by the mirror's own. It is reflecting nothing, in fact, of all that is there in the same space as itself: neither the painter with his back to it, nor the figures in the centre of the room. It is not the visible it reflects, in those bright depths. In Dutch painting it was traditional for mirrors to play a duplicating role:

they repeated the original contents of the picture, only inside an unreal, modified, contracted, concave space. One saw in them the same things as one saw in the first instance in the painting, but decomposed and re-composed according to a different law. Here, the mirror is saying nothing that has already been said before. Yet its position is more or less completely central: its upper edge is exactly on an imaginary line running half-way between the top and the bottom of the painting, it hangs right in the middle of the far wall (or at least in the middle of the portion we can see); it ought, therefore, to be governed by the same lines of perspective as the picture itself; we might well expect the same studio, the same painter, the same canvas to be arranged within it according to an identical space;

it could be the perfect duplication.

In fact, it shows us nothing of what is represented in the picture itself. Its motionless gaze extends out in front of the picture, into that necessarily

7

THB ORDER OF THINGS

invisible region which forms its exterior face, to apprehend the figures arranged in that space. Instead of surrounding visible objects, this mirror cuts straight through the whole field of the representation, ignoring all it might apprehend within that field, and restores visibility to that which resides outside all view. But the invisibility that it overcomes in this way is not the invisibility of what is hidden: it does not make its way around any obstacle, it is not distorting any perspective, it is addressing itself to what is invisible both because of the picture's structure and because of its existence as painting. What it is reflecting is that which all the figures within the painting are looking at so fixedly, or at least those who are looking straight ahead; it is therefore what the spectator would be able to see if the painting extended further forward, if its bottom edge were brought lower until it included the figures the painter is using as models. But it is also, since the picture does stop there, displaying only the painter and his studio, what is exterior to the picture, in so far as it is a picture - in other words, a rectangular fragment of lines and colours intended to rep­resent something to the eyes of any possible spectator. At the far end of the room, ignored by all, the unexpected mirror holds in its glow the figures that the painter is looking at (the painter in his represented, ob­jective reality, the reality of the painter at his work); but also the figures that are looking at the painter (in that material reality which the lines and the colours have laid out upon the canvas). These two groups of figures are both equally inaccessible, but in different ways: the first be­cause of an effect of composition peculiar to the painting; the second be­cause of the law that presides over the very existence of all pictures in general. Here, the action of representation consists in bringing one of these two forms of invisibility into the place of the other, in an unstable superimposition - and in rendering them both, at the same moment, at the other extremity of the picture - at that pole which is the very height of its representation: that of a reflected depth in the far recess of the painting's depth. The mirror provides a metathesis of visibility that affects both the space represented in the picture and its nature as representation;

it allows us to see, in the centre of the canvas, what in the painting is of necessity doubly invisible.

A strangely literal, though inverted, application of the advice given, so it is said, to his pupil by the old Pachero when the former was working in his studio in Seville: 'The image should stand out from the frame.'

8

LAS MENINAS II

But perhaps it is time to give a name at last to that image which appears in the depths of the mirror, and which the painter is contemplating in front of the picture. Perhaps it would be better, once and for all, to determine the identities of all the figures presented or indicated here, so as to avoid embroiling ourselves forever in those vague, rather abstract designations, so constantly prone to misunderstanding and duplication, 'the painter', 'the characters', 'the models', 'the spectators', 'the images'. Rather than pursue to infinity a language inevitably inadequate to the visible fact, it would be better to say that Velazquez composed a picture; that in this picture he represented himself, in his studio or in a room of the Escurial, in the act of painting two figures whom the Infanta Margarita has come there to watch, together with an entourage of duennas, maids of honour, courtiers, and dwarfs; that we can attribute names to this group of people with great precision: tradition recognizes that here we have Dona Maria Agustina Sarmiente, over there Nieto, in the foreground Nicolaso Pertusato, an Italian jester. We could then add that the two personages serving as models to the painter are not visible, at least directly; but that we can see them in a mirror; and that they are, without any doubt, King Philip IV and his wife, Mariana.

These proper names would form useful landmarks and avoid ambiguous designations; they would tell us in any case what the painter is looking at, and the majority of the characters in the picture along with him. But the relation of language to painting is an infinite relation. It is not that words are imperfect, or that, when confronted by the visible, they prove insuperably inadequate. Neither can be reduced to the other's terms: it is in vain that we say what we see; what we see never resides in what we say. And it is in vain that we attempt to show, by the use of images, metaphors, or similes, what we are saying; the space where they achieve their splendour is not that deployed by our eyes but that defined by the sequential elements of syntax. And the proper name, in this particular context, is merely an artifice: it gives us a finger to point with, in other words, to pass surreptitiously from the space where one speaks to the space where one looks; in other words, to fold one over the other as though they were equivalents. But if one wishes to keep the relation of language to vision open, if one wishes to treat their incompatibility as a starting-point for speech instead of as an obstacle to be avoided, so as to stay as close as possible to both, then one must erase those proper names

9

THE OKDBR OF THINGS

and preserve the infinity of the task. It is perhaps through the medium of this grey, anonymous language, always over-meticulous and repetitive be­cause too broad, that the painting may, little by little, release its illumina­tions.

We must therefore pretend not to know who is to be reflected in the depths of that mirror, and interrogate that reflection in its own terms.

First, it is the reverse of the great canvas represented on the left. The reverse, or rather the right side, since it displays in full face what the canvas, by its position, is hiding from us. Furthermore, it is both in opposition to the window and a reinforcement of it. Like the window, it provides a ground which is common to the painting and to what lies outside it. But the window operates by the continuous movement of an effusion which, flowing from right to left, unites the attentive figures, the painter, and the canvas, with the spectacle they are observing; whereas the mirror, on the other hand, by means of a violent, instantaneous move­ment, a movement of pure surprise, leaps out from the picture in order to reach that which is observed yet invisible in front of it, and then, at the far end of its fictitious depth, to render it visible yet indifferent to every gaze. The compelling tracer line, joining the reflection to that which it is reflecting, cuts perpendicularly through the lateral flood of light. Lastly -and this is the mirror's third function - it stands adjacent to a doorway which forms an opening, like the mirror itself, in the far wall of the room. This doorway too forms a bright and sharply defined rectangle whose soft light does not shine through into the room. It would be nothing but a gilded panel if it were not recessed out from the room by means of one leaf of a carved door, the curve of a curtain, and the shadows of several steps. Beyond the steps, a corridor begins; but instead of losing itself in obscurity, it is dissipated in a yellow dazzle where the light, without com­ing in, whirls around on itself in dynamic repose. Against this background, at once near and limitless, a man stands out in full-length silhouette; he is seen in profile; with one hand he is holding back the weight of a curtain; his feet are placed on different steps; one knee is bent. He may be about to enter the room; or he may be merely observing what is going on inside it, content to surprise those within without being seen him­self. Like the mirror, his eyes are directed towards the other side of the scene; nor is anyone paying any more attention to him than to the mirror. We do not know where he has come from: it could be that by following uncertain corridors he has just made his way around the outside of the room in which these characters are collected and the painter is at work;

10

LAS MENINAS

perhaps he too, a short while ago, was there in the forefront of the scene, in the invisible region still being contemplated by all those eyes in the pic­ture. Like the images perceived in the looking-glass, it is possible that he too is an emissary from that evident yet hidden space. Even so, there is a difference: he is there in flesh and blood; he has appeared from the outside, on the threshold of the area represented; he is indubitable - not a probable reflection but an irruption. The mirror, by making visible, beyond even the walls of the studio itself, what is happening in front of the picture, creates, in its sagittal dimension, an oscillation between the interior and the exterior. One foot only on the lower step, his body entirely in pro­file, the ambiguous visitor is coming in and going out at the same time, like a pendulum caught at the bottom of its swing. He repeats on the spot, but in the dark reality of his body, the instantaneous movement of those images flashing across the room, plunging into the mirror, being reflected there, and springing out from it again like visible, new, and identical species. Pale, minuscule, those silhouetted figures in the mirror are challenged by the tall, solid stature of the man appearing in the doorway. But we must move down again from the back of the picture towards the front of the stage; we must leave that periphery whose volute we have just been following. Starting from the painter's gaze, which constitutes an off-centre centre to the left, we perceive first of all the back of the canvas, then the paintings hung on the wall, with the mirror in their centre, then the open doorway, then more pictures, of which, because of the sharpness of the perspective, we can see no more than the edges of the frames, and finally, at the extreme right, the window, or rather the groove in the wall from which the light is pouring. This spiral shell presents us with the entire cycle of representation: the gaze, the palette and brush, the canvas innocent of signs (these are the material tools of representation), the paintings, the reflections, the real man (the completed representation, but as it were freed from its illusory or truthful contents, which are juxtaposed to it); then the representation dissolves again: we can see only the frames, and the light that is flooding the pictures from outside, but that they, in return, must reconstitute in their own kind, as though it were coming from elsewhere, passing through their dark wooden frames. And we do, in fact, see this light on the painting, ap­parently welling out from the crack of the frame; and from there it moves over to touch the brow, the cheekbones, the eyes, the gaze of the painter, who is holding a palette in one hand and in the other a fine brush . . . And so the spiral is closed, or rather, by means of that light, is opened.

11

THE ORDER OF THINGS

This opening is not, like the one in the back wall, made by pulling back a door; it is the whole breadth of the picture itself, and the looks that pass across it are not those of a distant visitor. The frieze that occupies the foreground and the middle ground of the picture represents - if we include the painter - eight characters. Five of these, their heads more or less bent, turned or inclined, are looking straight out at right angles to the surface of the picture. The centre of the group is occupied by the little Infanta, with her flared pink and gray dress. The princess is turning her head towards the right side of the picture, while her torso and the big panniers other dress slant away slightly towards the left; but her gaze is directed absolutely straight towards the spectator standing in front of the painting. A vertical line dividing the canvas into two equal halves would pass between the child's eyes. Her face is a third of the total height of the picture above the lower frame. So that here, beyond all question, resides the principal theme of the composition; this is the very object of this painting. As though to prove this and to emphasize it even more, Velazquez has made use of a traditional visual device: beside the principal figure he has placed a secondary one, kneeling and looking in towards the central one. Like a donor in prayer, like an angel greeting the Virgin, a maid of honour on her knees is stretching out her hands towards the princess. Her face stands out in perfect profile against the background. It is at the same height as that of the child. This attendant is looking at the princess and only at the princess. A little to the right, there stands another maid of honour, also turned towards the Infanta, leaning slightly over her, but with her eyes clearly directed towards the front, towards the same spot already being gazed at by the painter and the princess. Lastly, two other groups made up of two figures each: one of these groups is further away; the other, made up of the two dwarfs, is right in the fore­ground. One character in each of these pairs is looking straight out, the other to the left or the right. Because of their positions and their size, these two groups correspond and themselves form a pair: behind, the courtiers (the woman, to the left, looks to the right); in front, the dwarfs (the boy, who is at the extreme right, looks in towards the centre of the picture). This group of characters, arranged in this manner, can be taken to constitute, according to the way one looks at the picture and the centre of reference chosen, two different figures. The first would be a large X: the top left-hand point of this X would be the painter's eyes; the top right-hand one, the male courtier's eyes; at the bottom left-hand comer there is the comer of the canvas represented with its back towards us (or,

12

LAS MBNINAS

more exactly, the foot of the easel); at the bottom right-hand corner, the dwarf (his foot on the dog's back). Where these two lines intersect, at the centre of the X, are the eyes of the Infanta. The second figure would be more that of a vast curve, its two ends determined by the painter on the left and the male courtier on the right - both these extremities occurring high up in the picture and set back from its surface; the centre of the curve, much nearer to us, would coincide with the princess's face and the look her maid of honour is directing towards her. This curve describes a shallow hollow across the centre of the picture which at once contains and sets off the position of the mirror at the back.

There are thus two centres around which the picture may be organized, according to whether the fluttering attention of the spectator decides to settle in this place or in that. The princess is standing upright in the centre of a St Andrew's cross, which is revolving around her with its eddies of courtiers, maids of honour, animals, and fools. But this pivoting move­ment is frozen. Frozen by a spectacle that would be absolutely invisible if those same characters, suddenly motionless, were not offering us, as though in the hollow of a goblet, the possibility of seeing in the depths of a mirror the unforeseen double of what they are observing. In depth, it is the princess who is superimposed on the mirror; vertically, it is the re­flection that is superimposed on the face. But, because of the perspective, they are very close to one another. Moreover, from each of them there springs an ineluctable line: the line issuing from the mirror crosses the whole of the depth represented (and even more, since the mirror forms a hole in the back wall and brings a further space into being behind it);

the other line is shorter: it comes from the child's eyes and crosses only the foreground. These two sagittal lines converge at a very sharp angle, and the point where they meet, springing out from the painted surface, occurs in front of the picture, more or less exactly at the spot from which we are observing it. It is an uncertain point because we cannot see it; yet it is an inevitable and perfectly defined point too, since it is determined by those two dominating figures and confirmed further by other, adjacent dotted lines which also have their origin inside the picture and emerge from it in a similar fashion.

What is there, then, we ask at last, in that place which is completely inaccessible because it is exterior to the picture, yet is prescribed by all the lines of its composition? What is the spectacle, what are the faces that are reflected first of all in the depths of the Infanta's eyes, then in the courtiers' and the painter's, and finally in the distant glow of the mirror? But the i3

THE ORDER OF THINGS

question immediately becomes a double one: the face reflected in the mirror is also the face that is contemplating it; what all the figures in the picture are looking at are the two figures to whose eyes they too present a scene to be observed. The entire picture is looking out at a scene for which it is itself a scene. A condition of pure reciprocity manifested by the observing and observed mirror, the two stages of which are uncoupled at the two lower corners of the picture: on the left the canvas with its back to us, by means of which the exterior point is made into pure spectacle;

to the right the dog lying on the floor, the only element in the picture that is neither looking at anything nor moving, because it is not intended, with its deep reliefs and the light playing on its silky hair, to be anything but an object to be seen.

Our first glance at the painting told us what it' is' that creates this spectacle-as-observation. It is the two sovereigns. One can sense their presence already in the respectful gaze of the figures in the picture, in the astonishment of the child and the dwarfs. We recognize them, at the far end of the picture, in the two tiny silhouettes gleaming out from the looking-glass. In the midst of all those attentive faces, all those richly dressed bodies, they are the palest, the most unreal, the most compromised of all the painting's images: a movement, a little light, would be sufficient to eclipse them. Of all these figures represented before us, they are also the most ignored, since no one is paying the slightest attention to that reflection which has slipped into the room behind them all, silently occupying its unsuspected space; in so far as they are visible, they are the frailest and the most distant form of all reality. Inversely, in so far as they stand outside the picture and are therefore withdrawn from it in an essential invisibility, they provide the centre around which the entire representation is ordered: it is they who are being faced, it is towards them that everyone is turned, it is to their eyes that the princess is being presented in her holiday clothes; from the canvas with its back to us to the Infanta, and from the Infanta to the dwarf playing on the extreme right, there runs a curve (or again, the lower fork of the X opens) that orders the whole arrangement of the picture to their gaze and thus makes apparent the true centre of the composition, to which the Infanta's gaze and the image in the mirror are both finally subject.

In the realm of the anecdote, this centre is symbolically sovereign, since it is occupied by King Philip IV and his wife. But it is so above all because of the triple function it fulfils in relation to the picture. For in it there occurs an exact superimposition of the model's gaze as it is being

14

LAS MENINAS

painted, of the spectator's as he contemplates the painting, and of the painter's as he is composing his picture (not the one represented, but the one in front of us which we are discussing). These three 'observing' functions come together in a point exterior to the picture: that is, an ideal point in relation to what is represented, but a perfectly real one too, since it is also the starting-point that makes the representation possible. Within that reality itself, it cannot not be invisible. And yet, that reality is pro­jected within the picture - projected and diffracted in three forms which correspond to the three functions of that ideal and real point. They are: on the left, the painter with his palette in his hand (a self-portrait of Velazquez); to the right, the visitor, one foot on the step, ready to enter the room; he is taking in the scene from the back, but he can see the royal couple, who are the spectacle itself, from the front; and lastly, in the centre, the reflection of the king and the queen, richly dressed, motionless, in the attitude of patient models.

A reflection that shows us quite simply, and in shadow, what all those in the foreground are looking at. It restores, as if by magic, what is lack­ing in every gaze: in the painter's, the model, which his represented double is duplicating over there in the picture; in the king's, his portrait, which is being finished off on that slope of the canvas that he cannot perceive from where he stands; in that of the spectator, the real centre of the scene, whose place he himself has taken as though by usurpation. But perhaps this generosity on the part of the mirror is feigned; perhaps it is hiding as much as and even more than it reveals. That space where the king and his wife hold sway belongs equally well to the artist and to the spectator:

in the depths of the mirror there could also appear - there ought to appear - the anonymous face of the passer-by and that of Velazquez. For the function of that reflection is to draw into the interior of the picture what is intimately foreign to it: the gaze which has organized it and the gaze for which it is displayed. But because they are present within the picture, to the right and to the left, the artist and the visitor cannot be given a place in the mirror: just as the king appears in the depths of the looking-glass precisely because he does not belong to the picture.

In the great volute that runs around the perimeter of the studio, from the gaze of the painter, with his motionless hand and palette, right round to the finished paintings, representation came into being, reached completion, only to dissolve once more into the light; the cycle was complete. The lines that run through the depth of the picture, on the other hand, are not complete; they all lack a segment of their trajectories. This gap is

15

THE ORDER OF THINGS

caused by the absence of the king - an absence that is an artifice on the part of the painter. But this artifice both conceals and indicates another vacancy which is, on the contrary, immediate: that of the painter and the spectator when they are looking at or composing the picture. It may be that, in this picture, as in all the representations of which it is, as it were, the manifest essence, the profound invisibility of what one sees is in­separable from the invisibility of the person seeing - despite all mirrors, reflections, imitations, and portraits. Around the scene are arranged all the signs and successive forms of representation; but the double relation of the representation to its model and to its sovereign, to its author as well as to the person to whom it is being offered, this relation is neces­sarily interrupted. It can never be present without some residuum, even in a representation that offers itself as a spectacle. In the depth that traverses the picture, hollowing it into a fictitious recess and projecting it forward in front of itself, it is not possible for the pure felicity of the image ever to present in a full light both the master who is representing and the sovereign who is being represented.

Perhaps there exists, in this painting by Velazquez, the representation as it were, of Classical representation, and the definition of the space it opens up to us. And, indeed, representation undertakes to represent itself here in all its elements, with its images, the eyes to which it is offered, the faces it makes visible, the gestures that call it into being. But there, in the midst of this dispersion which it is simultaneously grouping to­gether and spreading out before us, indicated compellingly from every side, is an essential void: the necessary disappearance of that which is its foundation - of the person it resembles and the person in whose eyes it is only a resemblance. This very subject - which is the same - has been elided. And representation, freed finally from the relation that was im­peding it, can offer itself as representation in its pure form.

NOTES

[1] Sec frontispiece.

16

CHAPTER 2
The Prose of the World
I THE FOUR SIMILITUDES

Up to the end of the sixteenth century, resemblance played a constructive role in the knowledge of Western culture. It was resemblance that largely guided exegesis and the interpretation of texts; it was resemblance that organized the play of symbols, made possible knowledge of things visible and invisible, and controlled the art of representing them. The universe was folded in upon itself: the earth echoing the sky, faces seeing them­selves reflected in the stars, and plants holding within their stems the secrets that were of use to man. Painting imitated space. And representa­tion - whether in the service of pleasure or of knowledge - was posited as a form of repetition: the theatre of life or the mirror of nature, that was the claim made by all language, its manner of declaring its existence and of formulating its right of speech.

We must pause here for a while, at this moment in time when resem­blance was about to relinquish its relation with knowledge and disappear, in part at least, from the sphere of cognition. How, at the end of the sixteenth century, and even in the early seventeenth century, was simili­tude conceived? How did it organize the figures of knowledge? And if the things that resembled one another were indeed infinite in number, can one, at least, establish the forms according to which they might resemble one another?

The semantic web of resemblance in the sixteenth century is extremely rich: Amicitia, Aequalitas (contractus, consensus, matrimonium, societas, pax, et similia), Consonantia, Concertus, Continuum, Pantos, Proportio, Similitudo, Conjunctio, Copula[1]. And there are a great many other notions that intersect, overlap, reinforce, or limit one another on the surface of thought. It is enough for the moment to indicate the principal figures that deter­mine the knowledge of resemblance with their articulations. There are four of these that are, beyond doubt, essential.

17

THE ORDER OF THINGS

First of all, convenientia. This word really denotes the adjacency of places more strongly than it does similitude. Those things are 'convenient' which come sufficiently close to one another to be in juxtaposition; their edges touch, their fringes intermingle, the extremity of the one also de­notes the beginning of the other. In this way, movement, influences, passions, and properties too, are communicated. So that in this hinge between two things a resemblance appears. A resemblance that becomes double as soon as one attempts to unravel it: a resemblance of the place, the site upon which nature has placed the two things, and thus a similitude of properties; for in this natural container, the world, adjacency is not an exterior relation between things, but the sign of a relationship, obscure though it may be. And then, from this contact, by exchange, there arise new resemblances; a common regimen becomes necessary; upon the simi­litude that was the hidden reason for their propinquity is superimposed a resemblance that is the visible effect of that proximity. Body and soul, for example, are doubly 'convenient': the soul had to be made dense, heavy, and terrestrial for God to place it in the very heart of matter. But through this propinquity, the soul receives the movements of the body and assimilates itself to that body, while 'the body is altered and corrupted by the passions of the soul'[2]. In the vast syntax of the world, the differ­ent beings adjust themselves to one another; the plant communicates with the animal, the earth with the sea, man with everything around him. Resemblance imposes adjacencies that in their turn guarantee further re­semblances. Place and similitude become entangled: we see mosses grow­ing on the outsides of shells, plants in the antlers of stags, a sort of grass on the faces of men; and the strange zoophyte, by mingling together the properties that make it similar to the plants as well as to the animals, also juxtaposes them[3]. All so many signs of 'convenience'.

Convenientia is a resemblance connected with space in the form of a graduated scale of proximity. It is of the same order as conjunction and adjustment. This is why it pertains less to the things themselves than to the world in which they exist. The world is simply the universal 'convenience' of things; there are the same number of fishes in the water as there are animals, or objects produced by nature or man, on the land (are there not fishes called Episcopus, others called Catena, and others called Priapus?);

the same number of beings in the water and on the surface of the earth as there are in the sky, the inhabitants of the former corresponding with those of the latter; and lastly, there are the same number of beings in the whole of creation as may be found eminently contained in God himself,

18

THE PROSE OF THE WORLD

'the Sower of Existence, of Power, of Knowledge and of Love'[4]. Thus, by this linking of resemblance with space, this 'convenience' that brings like things together and makes adjacent things similar, the world is linked to­gether like a chain. At each point of contact there begins and ends a link that resembles the one before it and the one after it; and from circle to circle, these similitudes continue, holding the extremes apart (God and matter), yet bringing them together in such a way that the will of the Almighty may penetrate into the most unawakened comers. It is this immense, taut, and vibrating chain, this rope of 'convenience', that Porta evokes in a passage from his Magie naturelle:

As with respect to its vegetation the plant stands convenient to the brute beast, so through feeling does the brutish animal to man, who is conformable to the rest of the stars by his intelligence; these links pro­ceed so strictly that they appear as a rope stretched from the first cause as far as the lowest and smallest of things, by a reciprocal and continu­ous connection; in such wise that the superior virtue, spreading its beams, reaches so far that if we touch one extremity of that cord it will make tremble and move all the rest [5].

The second form of similitude is aemulatio: a sort of'convenience' that has been freed from the law of place and is able to function, without motion, from a distance. Rather as though the spatial collusion of con­venientia had been broken, so that the links of the chain, no longer con­nected, reproduced their circles at a distance from one another in accord­ance with a resemblance that needs no contact. There is something in emulation of the reflection and the mirror: it is the means whereby things scattered through the universe can answer one another. The human face, from afar, emulates the sky, and just as man's intellect is an imperfect re­flection of God's wisdom, so his two eyes, with their limited brightness, are a reflection of the vast illumination spread across the sky by sun and moon; the mouth is Venus, since it gives passage to kisses and words of love; the nose provides an image in miniature of Jove's sceptre and Mer­cury's staff[6]. The relation of emulation enables things to imitate one another from one end of the universe to the other without connection or proximity: by duplicating itself in a mirror the world abolishes the dis­tance proper to it; in this way it overcomes the place alloted to each thing. But which of these reflections coursing through space are the original images? Which is the reality and which the projection? It is often not possible to say, for emulation is a sort of natural twinship existing in

19

THE ORDER OF THINGS

things; it arises from a fold in being, the two sides of which stand im­mediately opposite to one another. Paracelsus compares this fundamental duplication of the world to the image of two twins 'who resemble one another completely, without its being possible for anyone to say which of them brought its similitude to the other* [7].

However, emulation does not leave the two reflected figures it has confronted in a merely inert state of opposition. One may be weaker, and therefore receptive to the stronger influence of the other, which is thus reflected in his passive mirror. Are not the stars, for example, dominant over the plants of the earth, of which they are the unchanged model, the unalterable form, and over which they have been secretly empowered to pour the whole dynasty of their influences? The dark earth is the mirror of the star-sown sky, but the two rivals are neither of equal value nor of equal dignity in that tournament. The bright colours of the flowers reproduce, without violence, the pure form of the sky. As Crollius says:

The stars are the matrix of all the plants and every star in the sky is only the spiritual prefiguration of a plant, such that it represents that plant, and just as each herb or plant is a terrestrial star looking up at the sky, so also each star is a celestial plant in spiritual form, which differs from the terrestrial plants in matter alone . . . , the celestial plants and herbs are turned towards the earth and look directly down upon the plants they have procreated, imbuing them with some particular virtue[8].

But the lists may remain open, and the untroubled mirror reflect only the image of'two wrathful soldiers'. Similitude then becomes the combat of one form against another - or rather of one and the same form separ­ated from itself by the weight of matter or distance in space. Man as Paracelsus describes him is, like the firmament, 'constellated with stars', but he is not bound to it like 'the thief to his galley-oar, the murderer to the wheel, the fish to the fisherman, the quarry to the huntsman'. It pertains to the firmament of man to be 'free and powerful', to 'bow to no order', and 'not to be ruled by any other created beings'. His inner sky may remain autonomous and depend only upon itself, but on condition that by means of his wisdom, which is also knowledge, he comes to re­semble the order of the world, takes it back into himself and thus re­creates in his inner firmament the sway of that other firmament in which he sees the glitter of the visible stars. If he does this, then the wisdom of the mirror will in turn be reflected back to envelop the world in which it has been placed; its great ring will spin out into the depths of the heavens,

20

THE PROSE OF THE WORLD

and beyond; man will discover that he contains 'the stars within him­self ..., and that he is thus the bearer of the firmament with all its in­fluences'[9].

Emulation is posited in the first place in the form of a mere reflection, furtive and distant; it traverses the spaces of the universe in silence. But the distance it crosses is not annulled by the subtle metaphor of emulation; it remains open to the eye. And in this duel, the two confronting figures seize upon one another. Like envelops like, which in turn surrounds the other, perhaps to be enveloped once more in a duplication which can continue ad infinitum. The links of emulation, unlike the elements of convenientia, do not form a chain but rather a series of concentric circles reflecting and rivalling one another.

The third form of similitude is analogy. An old concept already familiar to Greek science and medieval thought, but one whose use has probably become different now. In this analogy, convenientia and aemulatio arc superimposed. Like the latter, it makes possible the marvellous confron­tation of resemblances across space; but it also speaks, like the former, of adjacencies, of bonds and joints. Its power is immense, for the similitudes of which it treats are not the visible, substantial ones between things them­selves; they need only be the more subtle resemblances of relations. Disencumbered thus, it can extend, from a single given point, to an end­less number of relationships. For example, the relation of the stars to the sky in which they shine may also be found: between plants and the earth, between living beings and the globe they inhabit, between minerals such as diamonds and the rocks in which they are buried, between sense organs and the face they animate, between skin moles and the body of which they are the secret marks. An analogy may also be turned around upon itself without thereby rendering itself open to dispute. The old analogy of plant to animal (the vegetable is an animal living head down, its mouth - or roots - buried in the earth), is neither criticized nor disposed of by Cesalpino; on the contrary, he gives it added force, he multiplies it by itself when he makes the discovery that a plant is an upright animal, whose nutritive principles rise from the base up to the summit, channelled along a stem that stretches upwards like a body and is topped by a head -spreading flowers and leaves: a relation that inverts but does not contradict the initial analogy, since it places 'the root in the lower part of the plant and the stem in the upper part, for the venous network in animals also begins in the lower part of the belly, and the principal vein rises up to the heart and head'[10].

21

THE ORDER OF THINGS

This reversibility and this polyvalency endow analogy with a universal field of application. Through it, all the figures in the whole universe can be drawn together. There does exist, however, in this space, furrowed in every direction, one particularly privileged point: it is saturated with analogies (all analogies can find one of their necessary terms there), and as they pass through it, their relations may be inverted without losing any of their force. This point is man: he stands in proportion to the heavens, just as he does to animals and plants, and as he does also to the earth, to metals, to stalactites or storms. Upright between the surfaces of the universe, he stands in relation to the firmament (his face is to his body what the face of heaven is to the ether; his pulse beats in his veins as the stars circle the sky according to their own fixed paths; the seven orifices in his head are to his face what the seven planets are to the sky); but he is also the fulcrum upon which all these relations turn, so that we find them again, their similarity unimpaired, in the analogy of the human animal to the earth it inhabits: his flesh is a glebe, his bones are rocks, his veins great rivers, his bladder is the sea, and his seven principal organs arc the metals hidden in the shafts ofmines[11]. Man's body is always the possible half of a universal atlas. It is well known how Pierre Belon drew, and drew in the greatest detail, the first comparative illustration of the human skeleton and that of birds: in it, we see

the pinion called the appendix which is in proportion to the wing and in the same place as the thumb on the hand; the extremity of the pinion which is like the fingers in us ...; the bone given as legs to the bird corresponding to our heel; just as we have four toes on our feet, so the birds have four fingers of which the one behind is proportionate to the big toe in us[12].

So much precision is not, however, comparative anatomy except to an eye armed with nineteenth-century knowledge. It is merely that the grid through which we permit the figures of resemblance to enter our know­ledge happens to coincide at this point (and at almost no other) with that which sixteenth-century learning had laid over things.

In fact, Belon's description has no connection with anything but the positivity which, in his day, made it possible. It is neither more rational nor more scientific than an observation such as Aldrovandi's comparison of man's baser parts to the fouler parts of the world, to Hell, to the dark­ness of Hell, to the damned souls who are like the excrement of the Universe [13]; it belongs to the same analogical cosmography as the

22

THE PROSE OF THE WORLD

comparison, classic in Crollius's time, between apoplexy and tempests: the storm begins when the air becomes heavy and agitated, the apoplectic attack at the moment when our thoughts become heavy and disturbed; then the clouds pile up, the belly swells, the thunder explodes and the bladder bursts; the lightning flashes and the eyes glitter with a terrible brightness, the rain falls, the mouth foams, the thunderbolt is unleashed and the spirits burst open breaches in the skin; but then the sky becomes clear again, and in the sick man reason regains ascendancy [14]. The space occupied by analogies is really a space of radiation. Man is surrounded by it on every side; but, inversely, he transmits these resemblances back into the world from which he receives them. He is the great fulcrum of proportions - the centre upon which relations are concentrated and from which they are once again reflected.

Lastly, the fourth form of resemblance is provided by the play of sympathies. And here, no path has been determined in advance, no distance laid down, no links prescribed. Sympathy plays through the depths of the universe in a free state. It can traverse the vastest spaces in an instant: it falls like a thunderbolt from the distant planet upon the man ruled by that planet; on the other hand, it can be brought into being by a simple contact - as with those 'mourning roses that have been used at obsequies' which, simply from their former adjacency with death, will render all persons who smell them 'sad and moribund' [15]. But such is its power that sympathy is not content to spring from a single contact and speed through space; it excites the things of the world to movement and can draw even the most distant of them together. It is a principle of mobility: it attracts what is heavy to the heaviness of the earth, what is light up towards the weightless ether; it drives the root towards the water, and it makes the great yellow disk of the sunflower turn to follow the curving path of the sun. Moreover, by drawing things towards one another in an exterior and visible movement, it also gives rise to a hidden interior movement - a displacement of qualities that take over from one another in a series of relays: fire, because it is warm and light, rises up into the air, towards which its flames untiringly strive; but in doing so it loses its dryness (which made it akin to the earth) and so acquires humidity (which links it to water and air); it disappears therefore into light vapour, into blue smoke, into clouds: it has become air. Sympathy is an instance of the Same so strong and so insistent that it will not rest content to be merely one of the forms of likeness; it has the dangerous power of assimilating, of rendering things identical to one another, of mingling

23

THE ORDER OF THINGS

them, of causing their individuality to disappear - and thus of rendering them foreign to what they were before. Sympathy transforms. It alters, but in the direction of identity, so that if its power were not counter­balanced it would reduce the world to a point, to a homogeneous mass, to the featureless form of the Same: all its parts would hold together and communicate with one another without a break, with no distance be­tween them, like those metal chains held suspended by sympathy to the attraction of a single magnet [16].

This is why sympathy is compensated for by its twin, antipathy. Antipathy maintains the isolation of things and prevents their assimilation; it encloses every species within its impenetrable difference and its pro­pensity to continue being what it is:

It is fairly widely known that the plants have hatreds between them­selves ... it is said that the olive and the vine hate the cabbage; the cucumber flies from the olive . . . Since they grow by means of the sun's warmth and the earth's humour, it is inevitable that any thick and opaque tree should be pernicious to the others, and also the tree that has several roots[17].

And so to infinity, through all time, the world's beings will hate one another and preserve their ferocious appetites in opposition to all sym­pathy.

The rat of India is pernicious to the crocodile, since Nature has created them enemies; in such wise that when that violent reptile takes his pleasure in the sun, the rat lays an ambush for it of mortal subtlety;

perceiving that the crocodile, lying unaware for delight, is sleeping with its jaws agape, it makes its way through them and slips down the wide throat into the crocodile's belly, gnawing through the entrails of which, it emerges at last from the slain beast's bowel.

But the rat's enemies are lying in wait for it in their turn: for it lives in discord with the spider, and 'battling with the aspic it oft so dies'. Through this play of antipathy, which disperses them, yet draws them with equal force into mutual combat, makes them into murderers and then exposes them to death in their turn, things and animals and all the forms of the world remain what they are.

The identity of things, the fact that they can resemble others and be drawn to them, though without being swallowed up or losing their singularity - this is what is assured by the constant counterbalancing of

24

THE PROSE OF THE WORLD

sympathy and antipathy. It explains how things grow, develop, inter­mingle, disappear, die, yet endlessly find themselves again; in short, how there can be space (which is nevertheless not without landmarks or repeti­tions, not without havens of similitude) and time (which nevertheless allows the same forms, the same species, the same elements to reappear indefinitely).

Though yet of themselves the four bodies (water, air, fire, earth) be simple and possessed of their distinct qualities, yet forasmuch as the Creator has ordained that the elementary bodies shall be composed of mingled elements, therefore arc their harmonies and discordancies remarkable, as we may know from their qualities. The element of fire is hot and dry; it has therefore an antipathy to those of water, which is cold and damp. Hot air is humid, cold earth is dry, which is an antipathy. That they may be brought into harmony, air has been placed between fire and water, water between earth and air. Inasmuch as the air is hot, it marches well with fire and its humidity goes well with that of water. The humidity of water is heated by the heat of the air and brings relief to the cold dryness of the earth [18].

Because of the movement and the dispersion created by its laws, the sovereignty of the sympathy-antipathy pair gives rise to all the forms of resemblance. The first three similitudes are thus all resumed and explained by it. The whole volume of the world, all the adjacencies of 'convenience', all the echoes of emulation, all the linkages of analogy, are supported, maintained, and doubled by this space governed by sympathy and antipathy, which are ceaselessly drawing things together and holding them apart. By means of this interplay, the world remains identical; re­semblances continue to be what they are, and to resemble one another. The same remains the same, riveted onto itself.
II  SIGNATURES

And yet the system is not closed. One aperture remains: and through it the whole interplay of resemblances would be in danger of escaping from itself, or of remaining hidden in darkness, if there were not a further form of similitude to close the circle - to render it at once perfect and manifest.

Convenioitia, aemulntio, analogy, and sympathy tell us how the world must fold in upon itself, duplicate itself, reflect itself, or form a chain with

25

THE ORDER OF THINGS

itself so that things can resemble one another. They tell us what the paths of similitude are and the directions they take; but not where it is, how one sees it, or by what mark it may be recognized. Now there is a possi­bility that we might make our way through all this marvellous teeming abundance of resemblances without even suspecting that it has long been prepared by the order of the world, for our greater benefit. In order that we may know that aconite will cure our eye disease, or that ground walnut mixed with spirits of wine will case a headache, there must of course be some mark that will make us aware of these things: otherwise, the secret would remain indefinitely dormant. Would we ever know that there is a relation of twinship or rivalry between a man and his planet, if there were no sign upon his body or among the wrinkles on his face that he is an emulator of Mars or akin to Saturn? These buried similitudes must be indicated on the surface of things; there must be visible marks for the invisible analogies. Is not any resemblance, after all, both the most ob­vious and the most hidden of things? Because it is not made up of juxta­posed fragments, some identical and others different, it is all of a piece, a similitude that can be seen and yet not seen. It would thus lack any criterion if it did not have within it - or above it or beside it - a decisive clement to transform its uncertain glimmer into bright certainty.

There are no resemblances without signatures. The world of similarity can only be a world of signs. Paracelsus says:

It is not God's will that what he creates for man's benefit and what he has given us should remain hidden . . . And even though he has hidden certain things, he has allowed nothing to remain without exterior and visible signs in the form of special marks -just as a man who has buried a hoard of treasure marks the spot that he may find it again [19].

A knowledge of similitudes is founded upon the unearthing and decipher­ment of these signatures. It is useless to go no further than the skin or bark of plants if you wish to know their nature; you must go straight to their marks - 'to the shadow and image of God that they bear or to their internal virtue, which has been given to them by heaven as a natural dowry, ... a virtue, I say, that is to be recognized rather by its signa­ture'[20]. The system of signatures reverses the relation of the visible to the invisible. Resemblance was the invisible form of that which, from the depths of the world, made things visible; but in order that this form may be brought out into the light in its turn there must be a visible figure that will draw it out from its profound invisibility. This is why

26

THE PROSE OF THE WORLD

the face of the world is covered with blazons, with characters, with ciphers and obscure words - with 'hieroglyphics', as Turner called them. And the space inhabited by immediate resemblances becomes like a vast open book; it bristles with written signs; every page is seen to be filled with strange figures that intertwine and in some places repeat themselves. All that remains is to decipher them: 'Is it not true that all herbs, plants, trees and other things issuing from the bowels of the earth are so many magic books and signs?' [21] Thegreat untroubled mirror in whose depths things gazed at themselves and reflected their own images back to one another is, in reality, filled with the murmur of words. The mute re­flections all have corresponding words which indicate them. And by the grace of one final form of resemblance, which envelops all the others and encloses them within a single circle, the world may be compared to a man with the power of speech:

Just as the secret movements of his understanding are manifested by his voice, so it would seem that the herbs speak to the curious physician through their signatures, discovering to him . . . their inner virtues hidden beneath nature's veil ofsilence[22].

But we must pause a little here to examine this language! itself. To examine the signs of which it is made up and the way in which these signs refer back to what they indicate.

There exists a sympathy between aconite and our eyes. This unexpected affinity would remain in obscurity if there were not some signature on the plant, some mark, some word, as it were, telling us that it is good for diseases of the eye. This sign is easily legible in its seeds: they are tiny dark globes set in white skinlike coverings whose appearance is much like that of eyelids covering an eye [23]. It is the same with the affinity of the walnut and the human head: what cures 'wounds of the pericranium' is the thick green rind covering the bones - the shell - of the fruit; but internal head ailments may be prevented by use of the nut itself 'which is exactly like the brain in appearance'[24]. The sign of affinity, and what renders it visible, is quite simply analogy; the cipher of sympathy resides in the proportion.

But what signature can the proportion itself bear in order to make it­self recognizable? How is one to know that the lines of a hand or the furrows on a brow are tracing on a man's body the tendencies, accidents, or obstacles present in the whole vast fabric of his life? How indeed, if not because we know that sympathy creates communication between our

27

THE ORDER OF THINGS

bodies and the heavens, and transmits the movement of the planets to the affairs of men. And if not, too, because the shortness of a line reflects the simple image of a short life, the intersection of two furrows an ob­stacle in one's path, the upward direction of a wrinkle a man's rise to success. Breadth is a sign of wealth and importance; continuity denotes good fortune, discontinuity ill fortune[25]. The great analogy between body and destiny has its sign in the whole system of mirrors and attrac­tions. It is sympathies and emulations that indicate analogies.

Emulation may be recognized by analogy: the eyes are stars because they spread light over our faces just as stars light up the darkness, and because blind people exist in the world like clairvoyants in the darkest of nights. It can also be recognized through convenicntia: we have known, ever since the Greeks, that the strongest and bravest animals have large and well-developed extremities to their limbs, as though their strength had communicated itself to the most distant parts of their bodies. In the same way, man's face and hands must resemble the soul to which they are joined. The recognition of the most visible similitudes occurs, therefore, against a background of the discovery that things in general arc 'convenient' among themselves. And if one then considers that conveniency is not always defined by actual localization, but that many beings separated in space are also 'convenient' (as with a disease and its remedy, man and his stars, or a plant and the soil it needs), then again a sign of their conveniency is essential. And what other sign is there that two things are linked to one another unless it is that they have a mutual attraction for each other, as do the sun and the sunflower, or water and a cucumber shoot, that there is an affinity and, as it were, a sympathy between them?

And so the circle is closed. Though it is apparent what a complicated system of duplications was necessary to achieve this. Resemblances re­quire a signature, for none of them would ever become observable were it not legibly marked. But what are these signs? How, amid all the aspects of the world and so many interlacing forms, does one recognize that one is faced at any given moment with a character that should give one pause because it indicates a secret and essential resemblance? What form constitutes a sign and endows it with its particular value as a sign? - Re­semblance does. It signifies exactly in so far as it resembles what it is indicating (that is, a similitude). But what it indicates is not the homology; for its distinct existence as a signature would then be indistinguishable from the face of which it is the sign; it is another resemblance, an adjacent

28

THE PROSE OF THE WORLD

similitude, one of another type which enables us to recognize the first, and which is revealed in its turn by a third. Every resemblance receives a signature; but this signature is no more than an intermediate form of the same resemblance. As a result, the totality of these marks, sliding over the great circle of similitudes, forms a second circle which would be an exact duplication of the first, point by point, were it not for that tiny degree of displacement which causes the sign of sympathy to reside in an analogy, that of analogy in emulation, that of emulation in convenience, which in turn requires the mark of sympathy for its recognition. The signature and what it denotes are of exactly the same nature; it is merely that they obey a different law of distribution; the pattern from which they are cut is the same.

The form making a sign and the form being signalized are resemblances, but they do not overlap. And it is in this respect that resemblance in sixteenth-century knowledge is without doubt the most universal thing there is: at the same time that which is most clearly visible, yet something that one must nevertheless search for, since it is also the most hidden;

what determines the form of knowledge (for knowledge can only follow the paths of similitude), and what guarantees its wealth of content (for the moment one lifts aside the signs and looks at what they indicate, one allows Resemblance itself to emerge into the light of day and shine with its own inner light).

Let us call the totality of the learning and skills that enable one to make the signs speak and to discover their meaning, hermeneutics; let us call the totality of the learning and skills that enable one to distinguish the location of the signs, to define what constitutes them as signs, and to know how and by what laws they are linked, semiology: the sixteenth century superimposed hermeneutics and semiology in the form of similitude. To search for a meaning is to bring to light a resemblance. To search for the law governing signs is to discover the things that are alike. The gram­mar of beings is an exegesis of these things. And what the language they speak has to tell us is quite simply what the syntax is that binds them together. The nature of things, their coexistence, the way in which they are linked together and communicate is nothing other than their re­semblance. And that resemblance is visible only in the network of signs that crosses the world from one end to the other. 'Nature' is trapped in the thin layer that holds semiology and hermeneutics one above the other; it is neither mysterious nor veiled, it offers itself to our cognition, which it sometimes leads astray, only in so far as this superimposition

29

THE ORDER OF THINGS

necessarily includes a slight degree of non-coincidence between the re­semblances. As a result, the grid is less easy to see through; its transparency is clouded over from the very first. A dark space appears which must be made progressively clearer. That space is where 'nature' resides, and it is what one must attempt to know. Everything would be manifest and im­mediately knowable if the hermeneutics of resemblance and the semiology of signatures coincided without the slightest parallax. But because the similitudes that form the graphics of the world are one 'cog' out of alignment with those that form its discourse, knowledge and the infinite labour it involves find here the space that is proper to them: it is their task to weave their way across this distance, pursuing an endless zigzag course from resemblance to what resembles it.
III  THE LIMITS OF THE WORLD

Such, sketched in its most general aspects, is the sixteenth-century episteme. This configuration carries with it a certain number of conse­quences.

First and foremost, the plethoric yet absolutely poverty-stricken charac­ter of this knowledge. Plethoric because it is limitless. Resemblance never remains stable within itself; it can be fixed only if it refers back to another similitude, which then, in turn, refers to others; each resemblance, there­fore, has value only from the accumulation of all the others, and the whole world must be explored if even the slightest of analogies is to be justified and finally take on the appearance of certainty. It is therefore a knowledge that can, and must, proceed by the infinite accumulation of confirmations all dependent on one another. And for this reason, from its very foun­dations, this knowledge will be a thing of sand. The only possible form of link between the elements of this knowledge is addition. Hence those immense columns of compilation, hence their monotony. By positing resemblance as the link between signs and what they indicate (thus making resemblance both a third force and a sole power, since it resides in both the mark and the content in identical fashion), sixteenth-century knowledge condemned itself to never knowing anything but the same thing, and to knowing that thing only at the unattainable end of an end­less journey.

And it is here that we find that only too well-known category, the microcosm, coming into play. This ancient notion was no doubt revived, during the Middle Ages and at the beginning of the Renaissance, by a

30

THE PROSE OF THE WORLD

certain neo-Platonist tradition. But by the sixteenth century it had come to play a fundamental role in the field of knowledge. It hardly matters whether it was or was not, as was once claimed, a world view or Welt-enschawng. The fact is that it had one, or rather two, precise functions in the epistemological configuration of this period. As a category of thought, it applies the interplay of duplicated resemblances to all the realms of nature; it provides all investigation with an assurance that everything will find its mirror and its macrocosmic justification on another and larger scale; it affirms, inversely, that the visible order of the highest spheres will be found reflected in the darkest depths of the earth. But, under­stood as a general configuration of nature, it poses real and, as it were, tan­gible limits to the indefatigable to-and-fro of similitudes relieving one another. It indicates that there exists a greater world, and that its perimeter defines the limit of all created things; that at the far extremity of this great world there exists a privileged creation which reproduces, within its restricted dimensions, the immense order of the heavens, the stars, the mountains, rivers, and storms; and that it is between the effective limits of this constituent analogy that the interplay of resemblances takes place. By this very fact, however immense the distance from microcosm to macrocosm may be, it cannot be infinite; the beings that reside within it may be extremely numerous, but in the end they can be counted; and, Consequently, the similitudes that, through the action of the signs they require, always rest one upon another, can cease their endless night. They have a perfectly closed domain to support and buttress them. Nature, like the interplay of signs and resemblances, is closed in upon itself in conformity with the duplicated form of the cosmos.

We must therefore be careful not to invert the relations here. There is no doubt that the idea of the microcosm was, as we say, 'important' in the sixteenth century; it would probably have been one of the most frequently mentioned terms in the results of any poll taken at the time. But we are not concerned here with a study of opinions, which could be undertaken only by a statistical analysis of contemporary records. If, on the other hand, one investigates sixteenth-century knowledge at its archaeological level - that is, at the level of what made it possible - then the relations of macrocosm and microcosm appear as a mere surface affect. It was not because people believed in such relations that they set about trying to hunt down all the analogies in the world. But there was a necessity lying at the heart of their knowledge: they had to find an adjustmtent between the infinite richness of a resemblance introduced as a third

31

THE ORDER OF THINGS

term between signs and their meaning, and the monotony that imposed the same pattern of resemblance upon the sign and what it signified. In an episteme in which signs and similitudes were wrapped around one another in an endless spiral, it was essential that the relation of microcosm to macrocosm should be conceived as both the guarantee of that know­ledge and the limit of its expansion.

It was this same necessity that obliged knowledge to accept magic and erudition on the same level. To us, it seems that sixteenth-century learn­ing was made up of an unstable mixture of rational knowledge, notions derived from magical practices, and a whole cultural heritage whose power and authority had been vastly increased by the rediscovery of Greek and Roman authors. Perceived thus, the learning of that period appears structurally weak: a common ground where fidelity to the Ancients, a taste for the supernatural, and an already awakened awareness of that sovereign rationality in which we recognize ourselves, confronted one another in equal freedom. And this tripartite period would conse­quently be reflected in the mirror of each work and each divided mind occurring within it. ... In fact, it is not from an insufficiency of structure that sixteenth-century knowledge suffers. On the contrary, we have al­ready seen how very meticulous the configurations arc that define its space. It is this very rigour that makes the relation of magic to erudition inevitable - they arc not selected contents but required forms. The world is covered with signs that must be deciphered, and those signs, which re­veal resemblances and affinities, are themselves no more than forms of similitude. To know must therefore be to interpret: to find a way from the visible mark to that which is being said by it and which, without that mark, would lie like unspoken speech, dormant within things.

But we men discover all that is hidden in the mountains by signs and outward correspondences; and it is thus that we find out all the pro­perties of herbs and all that is in stones. There is nothing in the depths of the seas, nothing in the heights of the firmament that man is not capable of discovering. There is no mountain so vast that it can hide from the gaze of man what is within it; it is revealed to him by cor­responding signs [26].

Divination is not a rival form of knowledge; it is part of the main body of knowledge itself. Moreover, these signs that must be interpreted in­dicate what is hidden only in so far as they resemble it; and it is not possible to act upon those marks without at the same time operating upon

32

THE PROSE OF THE WORLD

that which is secretly indicated by them. This is why the plants that re­present the head, or the eyes, or the heart, or the liver, will possess an efficacity in regard to that organ; this is why the animals themselves will react to the marks that designate them. Paracelsus asks:

Tell me, then, why snakes in Helvetia, Algeria, Swedland understand the Greek words Osy, Osya, Osy ... In what academies did they learn them, so that scarcely have they heard the word than they immediately turn tail in order not to hear it again? Scarcely do they hear the word when, notwithstanding their nature and their spirit, they remain im­mobile and poison no one with their venomous wounds.

And let no one say that this is merely the effect of the sound made by the words when pronounced: 'if you write these words alone on vellum, parchment or paper at a favourable time, then place them in front of the serpent, it will stay no less motionless than if you had pronounced them aloud.' The project of elucidating the 'Natural Magics', which occupies an important place at the end of the sixteenth century and survives into the middle of the seventeenth, is not a vestigial phenomenon in the European consciousness; it was revived - as Campanella expressly tells us [27] - and for contemporary reasons: because the fundamental con­figuration of knowledge consisted of the reciprocal cross-reference of signs and similitudes. The form of magic was inherent in this way of knowing.

And by the same token, so was erudition: for, in the treasure handed down to us by Antiquity, the value of language lay in the fact that it was the sign of things. There is no difference between the visible marks that God has stamped upon the surface of the earth, so that we may know its inner secrets, and the legible words that the Scriptures, or the sages of Antiquity, have set down in the books preserved for us by tradition. The relation to these texts is of the same nature as the relation to things: in both cases there are signs that must be discovered. But God, in order to exercise our wisdom, merely sowed nature with forms for us to decipher (and it is in this sense that knowledge should be divinatio), whereas the Ancients have already provided us with interpretations, which we need do no more than gather together. Or which we would need only to gather together, were it not for the necessity of learning their language, reading their texts, and understanding what they have said. The heritage of Antiquity, like nature itself, is a vast space requiring interpretation; in both cases there arc signs to be discovered and then, little by little,

33

THE ORDER OF THINGS

made to speak. In other words, divinatio and cruditio arc both part of the same hermeneutics; but this develops, following similar forms, on two different levels: one moves from the mute sign to the thing itself (and makes nature speak); the other moves from the unmoving graphism to clear speech (it restores sleeping languages to life). But just as natural signs arc linked to what they indicate by the profound relation of resem­blance, so the discourse of the Ancients is in the image of what it expresses;

if it has the value of a precious sign, that is because, from the depth of its being, and by means of the light that has never ceased to shine through it since its origin, it is adjusted to things themselves, it forms a mirror for them and emulates them; it is to eternal truth what signs are to the secrets of nature (it is the mark whereby the word may be deciphered); and it possesses an ageless affinity with the things that it unveils. It is useless therefore to demand its title to authority; it is a treasury of signs linked by similitude to that which they are empowered to denote. The only difference is that we are dealing with a treasure-hoard of the second degree, one that refers to the notations of nature, which in their turn indicate obscurely the pure gold of things themselves. The truth of all these marks - whether they are woven into nature itself or whether they exist in lines on parchments and in libraries - is everywhere the same: coeval with the institution of God.

There is no difference between marks and words in the sense that there is between observation and accepted authority, or between verifiable fact and tradition. The process is everywhere the same: that of the sign and its likeness, and this is why nature and the word can intertwine with one another to infinity, forming, for those who can read it, one vast single text.
IV THE WRITING OF THINGS

In the sixteenth century, real language is not a totality of independent signs, a uniform and unbroken entity in which things could be reflected one by one, as in a mirror, and so express their particular truths. It is rather an opaque, mysterious thing, closed in upon itself, a fragmented mass, its enigma renewed in every interval, which combines here and there with the forms of the world and becomes interwoven with them: so much so that all these elements, taken together, form a network of marks in which each of them may play, and does in fact play, in relation to all the others, the role of content or of sign, that of secret or of indicator. 34

THE PROSE OF THE WORLD

In its raw, historical sixteenth-century being, language is not an arbitrary system; it has been set down in the world and forms a part of it, both because things themselves hide and manifest their own enigma like a language and because words offer themselves to men as things to be de­ciphered. The great metaphor of the book that one opens, that one pores over and reads in order to know nature, is merely the reverse and visible side of another transference, and a much deeper one, which forces lan­guage to reside in the world, among the plants, the herbs, the stones, and the animals.

Language partakes in the world-wide dissemination of similitudes and signatures. It must, therefore, be studied itself as a thing in nature. Like animals, plants, or stars, its elements have their laws of affinity and con­venience, their necessary analogies. Ramus divided his grammar into two parts. The first was devoted to etymology, which means that one looked in it to discover, not the original meanings of words, but the in­trinsic 'properties' of letters, syllables, and, finally, whole words. The second part dealt with syntax: its purpose was to teach 'the building of words together by means of their properties', and it consisted 'almost entirely in the convenience and mutual communion of properties, as of the noun with the noun or with the verb, of the adverb with all the words to which it is adjoined, of the conjunction in the order of things con­joined'[28]. Language is not what it is because it has a meaning; its representative content, which was to have such importance for gram­marians of the seventeenth and eighteenth centuries that it provided them with the guiding thread of their analyses, has no role to play here. Words group syllables together, and syllables letters, because there are virtues placed in individual letters that draw them towards each other or keep them apart, exactly as the marks found in nature also repel or attract one another. The study of grammar in the sixteenth century is based upon the same epistemological arrangement as the science of nature or the esoteric disciplines. The only differences are that there is only one nature and there are several languages; and that in the esoteric field the properties of words, syllables, and letters are discovered by another discourse which always remains secret, whereas in grammar it is the words and phrases of every­day life that themselves express their properties. Language stands half­way between the visible forms of nature and the secret conveniences of esoteric discourse. It is a fragmented nature, divided against itself and deprived of its original transparency by admixture; it is a secret that carries within itself, though near the surface, the decipherable signs of what it

35

THE ORDER OF THINGS

is trying to say. It is at the same time a buried revelation and a revelation that is gradually being restored to ever greater clarity.

In its original form, when it was given to men by God himself, language was an absolutely certain and transparent sign for things, because it re­sembled them. The names of things were lodged in the things they desig­nated, just as strength is written in the body of the lion, regality in the eye of the eagle, just as the influence of the planets is marked upon the brows of men: by the form of similitude. .This transparency was des­troyed at Babel as a punishment for men. Languages became separated and incompatible with one another only in so far as they had previously lost this original resemblance to the things that had been the prime reason for the existence of language. All the languages known to us are now spoken only against the background of this lost similitude, and in the space that it left vacant. There is only one language that retains a memory of that similitude, because it derives in direct descent from that first vocabu­lary which is now forgotten; because God did not wish men to forget the punishment inflicted at Babel; because this language had to be used in order to recount God's ancient Alliance with his people; and lastly, because it was in this language that God addressed himself to those who listened to him. Hebrew therefore contains, as if in the form of fragments, the marks of that original name-giving. And those words pronounced by Adam as he imposed them upon the various animals have endured, in part at least, and still carry with them in their density, like an embedded fragment of silent knowledge, the unchanging properties of beings:

Thus the stork, so greatly lauded for its charity towards its father and its mother, is called in Hebrew Chasida, which is to say, meek, charit­able, endowed with pity . . . The horse is named Sus, thought to be from the verb Hasas, unless that verb is rather derived from the noun, and it signifies to rise up, for among all four-footed animals the horse is most proud and brave, as Job depicts it in Chapter 39 [29].

But these arc no more than fragmentary monuments; all other languages have lost these radical similitudes, which have been preserved in Hebrew only in order to show that it was once the common language of God, Adam, and the animals of the newly created earth.

But though language no longer bears an immediate resemblance to the things it names, this does not mean that it is separate from the world; it still continues, in another form, to be the locus of revelations and to be included in the area where truth is both manifested and expressed. True,

36

THE PROSE OF THE WORLD

it is no longer nature in its primal visibility, but neither is it a mysterious instrument with powers known only to a few privileged persons. It is rather the figuration of a world redeeming itself, lending its ear at last to the true word. This is why it was God's wish that Latin, the language of his Church, should spread over the whole of the terrestrial globe. And it is also why all the languages of the world, as it became possible to know them through this conquest, make up together the image of the truth. Their interlacing and the space in which they are deployed free the sign of the redeemed world, just as the arrangement of the first names bore a likeness to the things that God had given to Adam for his use. Claude Duret points out that the Hebrews, the Canaans, the Samaritans, the Chaldeans, the Syrians, the Egyptians, the Carthaginians, the Phoe­nicians, the Arabs, the Saracens, the Turks, the Moors, the Persians, and the Tartars all write from right to left, following 'the course and daily movement of the first heaven, which is most perfect, according to the opinion of the great Aristotle, tending towards unity'; the Greeks, the Georgians, the Maronites, the Serbians, the Jacobites, the Copts, the Poznanians, and of course the Romans and all Europeans write from left to right, following 'the course and movement of the second heaven, home of the seven planets'; the Indians, Cathayans, Chinese, and Japanese write from top to bottom, in conformity with the 'order of nature, which has given men heads at the tops of their bodies and feet at the bottom';

'in opposition to the aforementioned', the Mexicans write either from bottom to top or else in 'spiral lines, such as those made by the sun in its annual journey through the Zodiac'. And thus 'by these five diverse sorts of writing the secrets and mysteries of the world's frame and the form of the cross, the unity of the heaven's rotundity and that of the earth, are properly denoted and expressed' [30]. The relation of languages to the world is one of analogy rather than of signification; or rather, their value as signs and their duplicating function are superimposed; they speak the heaven and the earth of which they are the image; they repro­duce in their most material architecture the cross whose coming they announce - that coming which establishes its existence in its own turn through the Scriptures and the "Word. Language possesses a symbolic function; but since the disaster at Babel we must no longer seek for it -with rare exceptions [31] - in the words themselves but rather in the very existence of language, in its total relation to the totality of the world, in the intersecting of its space with the loci and forms of the cosmos. Hence the form of the encyclopedic project as it appears at the end of

37

THE ORDER OF THINGS

the sixteenth century or in the first years of the seventeenth: not to reflect what one knows in the neutral element of language - the use of the alpha­bet as an arbitrary but efficacious encyclopaedic order does not appear until the second half of the seventeenth century [32] - but to reconstitute the very order of the universe by the way in which words are linked together and arranged in space. It is this project that we find in Gregoire's Syntaxeon artis mirabilis (1610), and in Alstedius's Encyclopaedia (1630); or again in the Tableau de tous les arts liberaux by Christophe de Savigny, who contrives to spatialize acquired knowledge both in accordance with the cosmic, unchanging, and perfect form of the circle and in accor­dance with the sublunary, perishable, multiple, and divided form of the tree; it is also to be found in the work of La Croix du Maine, who en­visages a space that would be at once an Encyclopaedia and a Library, and would permit the arrangement of written texts according to the forms of adjacency, kinship, analogy, and subordination prescribed by the world itself [3 3]. But in any case, such an interweaving of language and things, in a space common to both, presupposes an absolute privilege on the part of writing.

This privilege dominated the entire Renaissance, and was no doubt one of the great events in Western culture. Printing, the arrival in Europe of Oriental manuscripts, the appearance of a literature no longer created for the voice or performance and therefore not governed by them, the pre­cedence given to the interpretation of religious texts over the tradition and magisterium of the Church - all these things bear witness, without its being possible to indicate causes and effects, to the fundamental place accorded in the West to Writing. Henceforth, it is the primal nature of language to be written. The sounds made by voices provide no more than a transitory and precarious translation of it. What God introduced into the world was written words; Adam, when he imposed their first names upon the animals, did no more than read those visible and silent marks; the Law was entrusted to the Tables, not to men's memories;

and it is in a book that the true Word must be found again. Vigenere and Duret[34] both said-and in almost identical terms - that the written had always preceded the spoken, certainly in nature, and perhaps even in the knowledge of men. For it was very possible that before Babel, before the Flood, there had already existed a form of writing composed of the marks of nature itself, with the result that its characters would have had the power to act upon things directly, to attract them or repel them, to represent their properties, their virtues, and their secrets. A primitively

38

THB PROSE OF THB WORLD

natural writing, of which certain forms of esoteric knowledge, and the cabala first and foremost, may perhaps have preserved the scattered memory and were now attempting to retrieve its long-dormant powers. Bsoterism in the sixteenth century is a phenomenon of the written word, not the spoken word. At all events, the latter is stripped of all its powers;

it is merely the female part of language, Vigenere and Duret tell us, just as its intellect is passive; Writing, on the other hand, is the active intellect, the 'male principle' of language. It alone harbours the truth.

This primacy of the written word explains the twin presence of two forms which, despite their apparent antagonism, are indissociable in sixteenth-century knowledge. The first of these is a non-distinction be­tween what is seen and what is read, between observation and relation, which results in the constitution of a single, unbroken surface in which observation and language intersect to infinity. And the second, the inverse of the first, is an immediate dissociation of all language, duplicated, with­out any assignable term, by the constant reiteration of commentary.

Later, Buffon was to express astonishment at finding in the work of a naturalist like Aldrovandi such an inextricable mixture of exact descrip­tions, reported quotations, fables without commentary, remarks dealing indifferently with an animal's anatomy, its use in heraldry, its habitat, its mythological values, or the uses to which it could be put in medicine or magic. And indeed, when one goes back to take a look at the Historia serpentum et draconum, one finds the chapter 'On the serpent in general' arranged under the following headings: equivocation (which means the various meanings of the word serpent), synonyms and etymologies, differences, form and description, anatomy, nature and habits, tempera­ment, coitus and generation, voice, movements, places, diet, physiog­nomy, antipathy, sympathy, modes of capture, death and wounds caused by the serpent, modes and signs of poisoning, remedies, epithets, de­nominations, prodigies and presages, monsters, mythology, gods to which it is dedicated, fables, allegories and mysteries, hieroglyphics, em­blems and symbols, proverbs, coinage, miracles, riddles, devices, heraldic signs, historical facts, dreams, simulacra and statues, use in human diet, use in medicine, miscellaneous uses. Whereupon Buffon comments: 'Let it be judged after that what proportion of natural history is to be found in such a hotch-potch of writing. There is no description here, only legend.' And indeed, for Aldrovandi and his contemporaries, it was all legenda -things to be read. But the reason for this was not that they preferred the authority of men to the precision of an unprejudiced eye, but that nature,

39

THE ORDER OF THINGS

in itself, is an unbroken tissue of words and signs, of accounts and charac­ters, of discourse and forms. When one is faced with the task of writing an animal's history, it is useless and impossible to choose between the profes­sion of naturalist and that of compiler: one has to collect together into one and the same form of knowledge all that has been seen and heard, all that has been recounted, either by nature or by men, by the language of the world, by tradition, or by the poets. To know an animal or a plant, or any terrestrial thing whatever, is to gather together the whole dense layer of signs with which it or they may have been covered; it is to re­discover also all the constellations of forms from which they derive their value as heraldic signs. Aldrovandi was neither a better nor a worse observer than Buffon; he was neither more credulous than he, nor less attached to the faithfulness of the observing eye or to the rationality of things. His observation was simply not linked to things in accordance with the same system or by the same arrangement of the episteme. For Aldrovandi was meticulously contemplating a nature which was, from top to bottom, written.

Knowledge therefore consisted in relating one form of language to another form of language; in restoring the great, unbroken plain of words and things; in making everything speak. That is, in bringing into being, at a level above that of all marks, the secondary discourse of commentary. The function proper to knowledge is not seeing or demonstrating; it is interpreting. Scriptural commentary, commentaries on Ancient authors, commentaries on the accounts of travelers, commentaries on legends and fables: none of these forms of discourse is required to justify its claim to be expressing a truth before it is interpreted; all that is required of it is the possibility of talking about it. Language contains its own inner prin­ciple of proliferation. 'There is more work in interpreting interpretations than in interpreting things; and more books about books than on any other subject; we do nothing but write glosses on one another'[35]. These words are not a statement of the bankruptcy of a culture buried beneath its own monuments; they are a definition of the inevitable re­lation that language maintained with itself in the sixteenth century. This relation enabled language to accumulate to infinity, since it never ceased to develop, to revise itself, and to lay its successive forms one over another. Perhaps for the first time in Western culture, we find revealed the absolutely open dimension of a language no longer able to halt itself, because, never being enclosed in a definitive statement, it can express its truth only in some future discourse and is wholly intent on what it will

40

THE PROSE OF THE WORLD

have said; but even this future discourse itself does not have the power to halt the progression, and what it says is enclosed within it like a promise, a bequest to yet another discourse.... The task of commentary can never, by definition, be completed. And yet commentary is directed entirely towards the enigmatic, murmured element of the language being com­mented on: it calls into being, below the existing discourse, another discourse that is more fundamental and, as it were, 'more primal', which it sets itself the task of restoring. There can be no commentary unless, below the language one is reading and deciphering, there runs the sover­eignty of an original Text. And it is this text which, by providing a foundation for the commentary, offers its ultimate revelation as the promised reward of commentary. The necessary proliferation of the exegesis is therefore measured, ideally limited, and yet ceaselessly ani­mated, by this silent dominion. The language of the sixteenth century -understood not as an episode in the history of any one tongue, but as a global cultural experience - found itself caught, no doubt, between these interacting elements, in the interstice occurring between the primal Text and the infinity of Interpretation. One speaks upon the basis of a writing that is part of the fabric of the world; one speaks about it to infinity, and each of its signs becomes in turn written matter for further discourse;

but each of these stages of discourse is addressed to that primal written word whose return it simultaneously promises and postpones.

It will be seen that the experience of language belongs to the same archaeological network as the knowledge of things and nature. To know those things was to bring to light the system of resemblances that made them close to and dependent upon one another; but one could discover the similitudes between them only in so far as there existed, on their sur­face, a totality of signs forming the text of an unequivocal message. But then, these signs themselves were no more than a play of resemblances, and they referred back to the infinite and necessarily uncompleted task of knowing what is similar. In the same way, though the analogy is in­verted, language sets itself the task of restoring an absolutely primal dis­course, but it can express that discourse only by trying to approximate to it, by attempting to say things about it that are similar to it, thereby bringing into existence the infinity of adjacent and similar fidelities of interpretation. The commentary resembles endlessly that which it is commenting upon and which it can never express; just as the knowledge of nature constantly finds new signs for resemblance because resemblance cannot be known in itself, even though the signs can never be anything but

41

THE ORDER OF THINGS

similitudes. And just as this infinite play within nature finds its link, its form, and its limitation in the relation of the microcosm to the macro­cosm, so does the infinite task of commentary derive its strength from the promise of an effectively written text which interpretation will one day reveal in its entirety.
V THE BEING OF LANGUAGE

Ever since the Stoics, the system of signs in the Western world had been a ternary one, for it was recognized as containing the significant, the sig­nified, and the 'conjuncture' (the tvyxavov). From the seventeenth cen­tury, on the other hand, the arrangement of signs was to become binary, since it was to be defined, with Port-Royal, as the connection of a signi­ficant and a signified. At the Renaissance, the organization is different, and much more complex: it is ternary, since it requires the formal domain of marks, the content indicated by them, and the similitudes that link the marks to the things designated by them; but since resemblance is the form of the signs as well as their content, the three distinct elements of this articulation are resolved into a single form.

This arrangement, together with the interplay it authorizes, is found also, though inverted, in the experience of language. In fact, language exists first of all, in its raw and primitive being, in the simple, material form of writing, a stigma upon things, a mark imprinted across the world which is a part of its most ineffaceable forms. In a sense, this layer of language is unique and absolute. But it also gives rise to two other forms of discourse which provide it with a frame: above it, there is commentary, which recasts the given signs to serve a new purpose, and below it, the text, whose primacy is presupposed by commentary to exist hidden be­neath the marks visible to all. Hence there are three levels of language, all based upon the single being of the written word. It is this complex interaction of elements that was to disappear with the end of the Renais­sance. And in two ways: because the forms oscillating endlessly between one and three terms were to be fixed in a binary form which would render them stable; and because language, instead of existing as the material writing of things, was to find its area of being restricted to the general organization of representative signs.

This new arrangement brought about the appearance of a new problem, unknown until then: in the sixteenth century, one asked oneself how it was possible to know that a sign did in fact designate what it signified;

42

THE PROSE OF THE WORLD

from the seventeenth century, one began to ask how a sign could be linked to what it signified. A question to which the Classical period was to reply by the analysis of representation; and to which modern thought was to reply by the analysis of meaning and signification. But given the fact itself, language was never to be anything more than a particular case of representation (for the Classics) or of signification (for us). The pro­found kinship of language with the world was thus dissolved. The primacy of the written word went into abeyance. And that uniform layer, in which the seen and the read, the visible and the expressible, were endlessly interwoven, vanished too. Things and words were to be separated from one another. The eye was thenceforth destined to see and only to see, the ear to hear and only to hear. Discourse was still to have the task of speaking that which is, but it was no longer to be anything more than what it said.

This involved an immense reorganization of culture, a reorganization of which the Classical age was the first and perhaps the most important stage, since it was responsible for the new arrangement in which we are still caught - since it is the Classical age that separates us from a culture in which the signification of signs did not exist, because it was reabsorbed into the sovereignty of the Like; but in which their enigmatic, monoton­ous, stubborn, and primitive being shone in an endless dispersion.

There is nothing now, either in our knowledge or in our reflection, that still recalls even the memory of that being. Nothing, except perhaps literature - and even then in a fashion more allusive and diagonal than direct. It may be said in a sense that 'literature', as it was constituted and so designated on the threshold of the modern age, manifests, at a time when it was least expected, the reappearance, of the living being of lan­guage. In the seventeenth and eighteenth centuries, the peculiar existence and ancient solidity of language as a thing inscribed in the fabric of the world were dissolved in the functioning of representation; all language had value only as discourse. The art of language was a way of 'making a sign' - of simultaneously signifying something and arranging signs around that thing; an art of naming, therefore, and then, by means of a reduplication both demonstrative and decorative, of capturing that name, of enclosing and concealing it, of designating it in turn by other names that were the deferred presence of the first name, its secondary sign, its figuration, its rhetorical panoply. And yet, throughout the nineteenth century, and right up to our own day - from Holderlin to Mallarme and on to Antonin Artaud - literature achieved autonomous existence, and

43

THE ORDER OF THINGS

separated itself from all other language with a deep scission, only by form­ing a sort of 'counter-discourse', and by finding its way back from the representative or signifying function of language to this raw being that had been forgotten since the sixteenth century.

It is possible to believe that one has attained the very essence of litera­ture when one is no longer interrogating it at the level of what it says but only in its significant form: in doing so, one is limiting one's view of language to its Classical status. In the modern age, literature is that which compensates for (and not that which confirms) the signifying function of language. Through literature, the being of language shines once more on the frontiers of Western culture - and at its centre - for it is what has been most foreign to that culture since the sixteenth century; but it has also, since this same century, been at the very centre of what Western culture has overlain. This is why literature is appearing more and more as that which must be thought; but equally, and for the same reason, as that which can never, in any circumstance, be thought in accordance with a theory of signification. Whether one analyses it from the point of view of what is signified (of what it is trying to say, of its 'ideas', of what it promises, or of what it commits one to) or from the point of view of that which signifies (with the help of paradigms borrowed from linguis­tics or psychoanalysis) matters little: all that is merely incidental. In both cases one would be searching for it outside the ground in which, as re­gards our culture, it has never ceased for the past century and a half to come into being and to imprint itself. Such modes of decipherment be­long to a Classical situation of language - the situation that predominated during the seventeenth century, when the organization of signs became binary, and when signification was reflected in the form of the representa­tion; for at that time literature really was composed of a signifying element and a signified content, so that it was proper to analyse it accordingly. But from the nineteenth century, literature began to bring language back to light once more in its own being: though not as it had still appeared at the end of the Renaissance. For now we no longer have that primary, that absolutely initial, word upon which the infinite movement of discourse was founded and by which it was limited;

henceforth, language was to grow with no point of departure, no end, and no promise. It is the traversal of this futile yet fundamental space that the text of literature traces from day to day.

44

THE PROSE OF THE WORLD NOTES

[1] P. Gregoire, Syntaxeon artis mirabilis (Cologne, 1610, p. 28).

[2] G. Porta, La Physionomie humaine (Fr. trans. 1655, p. i).

[3] U. Aldrovandi, Monstrorum historia (Bononiae, 1647, p. 663).

[4] T. Campanella, Realis philosophia (Frankfurt, 1623, p. 98).

[5] G. Porta, Magie naturelle (Fr. trans. Rouen, 1650, p. 22).

[6] Aldrovandi, Monstrorum historia, p. 3.

[7] Paracelsus, Liber Paramirum (trans. Grillot de Givry, Paris, 1913, p. 3).

[8] 0. Crollius, Traite des signatures (Fr. trans. Lyon, 1624, p. 18).

[9] Paracelsus, loc. cit.

[10] Cesalpino, De plantis libri, XVI (1583).

[11] Crollius, Traite des signatures, p. 88.

[12] P. Belon, Histoire de la nature des oiscaux (Paris, 1555, p. 37).

[13] Aldrovandi, Monstrorum historia, p. 4.

[14] Crollius, Traite des signatures, p. 87.

[15] Porta, Magic naturelle, p. 72.

[16] Ibid.

[17] J. Cardan, De la subtilite (Fr. trans. Paris, 1656, p. 154).

[18] S.G.S. Annotations au Grand Miroir du Monde de Duchesne, p. 498.

[19] Paracelsus, Die 9 Bucher der Natura Rerum (Works, ed. Suhdorff, vol. IX, P. 393).

[20] Crollius, Traite des signatures, p. 4.

[21] Ibid., p. 6.

[22] Ibid., p. 6.

[23] Ibid., p. 33.

[24] Ibid., pp. 33-4.

[25] J. Cardan, Metoposcopie (1658 edn., pp. iii-viii).

[26] Paracelsus, Archidoxis magica (Fr. trans. 1909, pp. 21-3).

[27] T. Campanella, De sensu rerum et magia (Frankfurt, 1620).

[28] P. Ramus, Grammaire (Paris, 1572, p. 3 and pp. 125-6).

[29] Claude Duret, Tresor de I'histoire des langues (Cologne, 1613, p. 40).

[30] Duret, loc. cit.

[31] In Mithridates, J. M. Gesner cites onomatopeias of course, but only as exceptions to a rule (2nd edn., Tiguri, 1610, pp. 3-4).

[32] Except with regard to languages, since the alphabet is the raw material of  language. Cf. Chapter II of Gesner's Mithridates. The first alphabetical encyclopaedia is L. Moreri's Grand Dictionnaire nistoriaue of 1674.

[33] La Croix du Maine, Les cents Buffets pour dresser une bibliotheque parfaite (1583).

[34] Blaise de Vigenere, Traitedes chiffres (Paris, 1537, pp. i and2); C. Duret, Tresor de I'histoire des langues, pp. 19 and 20.

[35] Montaigne, Essais (1580-8, livre III, chap. XIII). 45

CHAPTER 3
Representing
I DON QUIXOTE

With all their twists and turns, Don Quixote's adventures form the boundary: they mark the end of the old interplay between resemblance and signs and contain the beginnings of new relations. Don Quixote is not a man given to extravagance, but rather a diligent pilgrim breaking his journey before all the marks of similitude. He is the hero of the Same. He never manages to escape from the familiar plain stretching out on all sides of the Analogue, any more than he does from his own small pro­vince. He travels endlessly over that plain, without ever crossing the clearly defined frontiers of difference, or reaching the heart of identity. Moreover, he is himself like a sign, a long, thin graphism, a letter that has just escaped from the open pages of a book. His whole being is noth­ing but language, text, printed pages, stories that have already been written down. He is made up of interwoven words; he is writing itself, wandering through the world among the resemblances of things. Yet not entirely so: for in his reality as an impoverished hidalgo he can be­come a knight only by listening from afar to the age-old epic that gives its form to Law. The book is not so much his existence as his duty. He is constantly obliged to consult it in order to know what to do or say, and what signs he should give himself and others in order to show that he really is of the same nature as the text from which he springs. The chivalric romances have provided once and for all a written prescription for his adventures. And every episode, every decision, every exploit will be yet another sign that Don Quixote is a true likeness of all the signs that he has traced from his book. But the fact that he wishes to be like them means that he must put them to the test, that the (legible) signs no longer resemble (visible) people. All those written texts, all those extravagant romances are, quite literally, unparalleled: no one in the world ever did

46

REPRESENTING

resemble them; their timeless language remains suspended, unfulfilled by any similitude; they could all be burned in their entirety and the form of the world would not be changed. If he is to resemble the texts of which he is the witness, the representation, the real analogue, Don Quixote must also furnish proof and provide the indubitable sign that they are telling die truth, that they really are the language of the world. It is incumbent upon him to fulfil the promise of the books. It is his task to recreate the epic, though by a reverse process: the epic recounted (or claimed to re­count) real exploits, offering them to our memory; Don Quixote, on the other hand, must endow with reality the signs-without-content of the narrative. His adventures will be a deciphering of the world: a diligent search over the entire surface of the earth for the forms that will prove that what the books say is true. Each exploit must be a proof: it consists, not in a real triumph - which is why victory is not really important - but in an attempt to transform reality into a sign. Into a sign that the signs of language really are in conformity with things themselves. Don Quixote reads the world in order to prove his books. And the only proofs he gives himself are the glittering reflections of resemblances.

His whole journey is a quest for similitudes: the slightest analogies are pressed into service as dormant signs that must be reawakened and made to speak once more. Flocks, serving girls, and inns become once more the language of books to the imperceptible degree to which they resemble castles, ladies, and armies - a perpetually untenable resemblance which transforms the sought-for proof into derision and leaves the words of the books forever hollow. But non-similitude itself has its model, and one that it imitates in the most servile way: it is to be found in the trans­formations performed by magicians. So all the indices of non-resemblance, all the signs that prove that the written texts are not telling the truth, resemble the action of sorcery, which introduces difference into the in­dubitable existence of similitude by means of deceit. And since this magic has been foreseen and described in the books, the illusory difference that it introduces can never be anything but an enchanted similitude, and, there­fore, yet another sign that the signs in the books really do resemble the truth.

Don Quixote is a negative of the Renaissance world; writing has ceased to be the prose of the world; resemblances and signs have dissolved their former alliance; similitudes have become deceptive and verge upon the visionary or madness; things still remain stubbornly within their ironic identity: they are no longer anything but what they are; words wander

47

THE ORDER OF THINGS

off on their own, without content, without resemblance to fill their emptiness; they are no longer the marks of things; they lie sleeping be­tween the pages of books and covered in dust. Magic, which permitted the decipherment of the world by revealing the secret resemblances be­neath its signs, is no longer of any use except as an explanation, in terms of madness, of why analogies arc always proved false. The erudition that once read nature and books alike as parts of a single text has been relegated to the same category as its own chimeras: lodged in the yellowed pages of books, the signs of language no longer have any value apart from the slender fiction which they represent. The written word and things no longer resemble one another. And between them, Don Quixote wanders off on his own.

Yet language has not become entirely impotent. It now possesses new powers, and powers peculiar to it alone. In the second part of the novel, Don Quixote meets characters who have read the first part of his story and recognize him, the real man, as the hero of the book. Cervantes's text turns back upon itself, thrusts itself back into its own density, and becomes the object of its own narrative. The first part of the hero's adventures plays in the second part the role originally assumed by the chivalric romances. Don Quixote must remain faithful to the book that he has now become in reality; he must protect it from errors, from counterfeits, from apocryphal sequels; he must fill in the details that have been left out; he must preserve its truth. But Don Quixote himself has not read this book, and docs not have to read it, since he is the book in flesh and blood. Having first read so many books that he became a sign, a sign wandering through a world that did not recognize him, he has now, despite himself and without his knowledge, become a book that contains his truth, that records exactly all that he has done and said and seen and thought, and that at last makes him recognizable, so closely does he re­semble all those signs whose ineffaceable imprint he has left behind him. Between the first and second parts of the novel, in the narrow gap between those two volumes, and by their power alone, Don Quixote has achieved his reality - a reality he owes to language alone, and which resides entirely inside the words. Don Quixote's truth is not in the re­lation of the words to the world but in that slender and constant relation woven between themselves by verbal signs. Tlic hollow fiction of epic exploits has become the representative power of language. Words have swallowed up their own nature as signs.

Don Quixote is the first modern work of literature, because in it we see

48

REPRESENTING

the cruel reason of identities and differences make endless sport of signs and similitudes; because in it language breaks off its old kinship with things and enters into that lonely sovereignty from which it will reappear, in its separated state, only as literature; because it marks the point where resemblance enters an age which is, from the point of view of resemblance, one of madness and imagination. Once similitude and signs are sundered from each other, two experiences can be established and two characters appear face to face. The madman, understood not as one who is sick but as an established and maintained deviant, as an indispensable cultural function, has become, in Western experience, the man of primitive re­semblances. This character, as he is depicted in the novels or plays of the Baroque age, and as he was gradually institutionalized right up to the advent of nineteenth-century psychiatry, is the man who is alienated in analogy. He is the disordered player of the Same and the Other. He takes things for what they are not, and people one for another; he cuts his friends and recognizes complete strangers; he thinks he is unmasking when, in fact, he is putting on a mask. He inverts all values and all proportions, because he is constantly under the impression that he is deciphering signs: for him, the crown makes the king. In the cultural perception of the madman that prevailed up to the end of the eighteenth century, he is Different only in so far as he is unaware of Difference; he sees nothing but resemblances and signs of resemblance everywhere; for him all signs resemble one another, and all resemblances have the value of signs. At the other end of the cultural area, but brought close by sym­metry, the poet is he who, beneath the named, constantly expected differences, rediscovers the buried kinships between things, their scattered resemblances. Beneath the established signs, and in spite of them, he hears another, deeper, discourse, which recalls the time when words glittered in the universal resemblance of things; in the language of the poet, the Sovereignty of the Same, so difficult to express, eclipses, the distinction existing between signs.

This accounts, no doubt, for the confrontation of poetry and madness in modern Western culture. But it is no longer the old Platonic theme of inspired madness. It is the mark of a new experience of language and things. At the fringes of a knowledge that separates beings, signs, and similitudes, and as though to limit its power, the madman fulfils the function of homosemanticism: he groups all signs together and leads them with a resemblance that never ceases to proliferate. The poet fulfils the opposite function: his is tlic allegorical role; beneath the language of

49

THE ORDER OF THINGS

signs and beneath the interplay of their precisely delineated distinctions, he strains his ears to catch that 'other language', the language, without words or discourse, of resemblance. The poet brings similitude to the signs that speak it, whereas the madman loads all signs with a resemblance that ultimately erases them. They share, then, on the outer edge of our culture and at the point nearest to its essential divisions, that 'frontier' situation - a marginal position and a profoundly archaic silhouette -where their words unceasingly renew the power of their strangeness and the strength of their contestation. Between them there has opened up a field of knowledge in which, because of an essential rupture in the Western world, what has become important is no longer resemblances but identities and differences.
II  ORDER

Establishing discontinuities is not an easy task even for history in general. And it is certainly even less so for the history of thought. We may wish to draw a dividing-line; but any limit we set may perhaps be no more than an arbitrary division made in a constantly mobile whole. We may wish to mark off a period; but have we the right to establish symmetrical breaks at two points in time in order to give an appearance of continuity and unity to the system we place between them? Where, in that case, would the cause of its existence lie? Or that of its subsequent disappearance and fall? What rule could it be obeying by both its existence and its disappearance? If it contains a principle of coherence within itself, whence could come the foreign element capable of rebutting it? How can a thought melt away before anything other than itself? Generally speaking, what does it mean, no longer being able to think a certain thought? Or to introduce a new thought?

Discontinuity - the fact that within the space of a few years a culture sometimes ceases to think as it had been thinking up till then and begins to think other things in a new way - probably begins with an erosion from outside, from that space which is, for thought, on the other side, but in which it has never ceased to think from the very beginning. Ultimately, the problem that presents itself is that of the relations between thought and culture: how is it that thought has a place in the space of the world, that it has its origin there, and that it never ceases, in this place or that, to begin anew? But perhaps it is not yet time to pose this problem; perhaps we should wait until the archaeology of thought has been

50

REPRESENTING

established more firmly, until it is better able to gauge what it is capable of describing directly and positively, until it has defined the particular systems and internal connections it has to deal with, before attempting to encompass thought and to investigate how it contrives to escape itself. For the moment, then, let it suffice that we accept these discontinuities in the simultaneously manifest and obscure empirical order wherever they posit themselves.

At the beginning of the seventeenth century, during the period that has been termed, rightly or wrongly, the Baroque, thought ceases to move in the element of resemblance. Similitude is no longer the form of knowledge but rather the occasion of error, the danger to which one exposes oneself when one does not examine the obscure region of con­fusions. 'It is a frequent habit,' says Descartes, in the first lines of his Regulae, 'when we discover several resemblances between two things, to attribute to both equally, even on points in which they arc in reality different, that which we have recognized to be true of only one of them'[1]. The age of resemblance is drawing to a close. It is leaving noth­ing behind it but games. Games whose powers of enchantment grow out of the new kinship between resemblance and illusion; the chimeras of similitude loom up on all sides, but they are recognized as chimeras; it is the privileged age of trompe-l'aeil painting, of the comic illusion, of the play that duplicates itself by representing another play, of the quid pro quo, of dreams and visions; it is the age of the deceiving senses; it is the age in which the poetic dimension of language is defined by metaphor, simile, and allegory. And it was also in the nature of things that the know­ledge of the sixteenth century should leave behind it the distorted memory of a muddled and disordered body of learning in which all the things in the world could be linked indiscriminately to men's experiences, tradi­tions, or credulities. From then on, the noble, rigorous, and restrictive figures of similitude were to be forgotten. And the signs that designated them were to be thought of as the fantasies and charms of a knowledge that had not yet attained the age of reason.

We already find a critique of resemblance in Bacon - an empirical critique that concerns, not the relations of order and equality between dungs, but the types of mind and the forms of illusion to which they might be subject. We are dealing with a doctrine of the quid pro quo. Bacon does not dissipate similitudes by means of evidence and its attendant rules. He shows them, shimmering before our eyes, vanishing as one draws near, then re-forming again a moment later, a little further off. They are idols.

51

THE ORDER OF THINGS

The idols of the den and the idols of the theatre make us believe that things resemble what we have learned and the theories we have formed for ourselves; other idols make us believe that things are linked by resem­blances between themselves.

The human Intellect, from its peculiar nature, easily supposes a greater order and equality in things than it actually finds; and, while there are many things in Nature unique, and quite irregular, still it feigns paral­lels, correspondents, and relations that have no existence. Hence that fiction, 'that among the heavenly bodies all motion takes place by perfect circles'.

Such arc the idols of the tribe, spontaneous fictions of the mind; to which are added - as effects and sometimes as causes - the confusions of lan­guage: one and the same name being applied indifferently to things that are not of the same nature. These are the idols of the market[2]. Only prudence on the part of the mind can dissipate them, if it abjures its natural haste and levity in order to become 'penetrating' and ultimately perceive the differences inherent in nature.

The Cartesian critique of resemblance is of another type. It is no longer sixteenth-century thought becoming troubled as it contemplates itself and beginning to jettison its most familiar forms; it is Classical thought excluding resemblance as the fundamental experience and primary form of knowledge, denouncing it as a confused mixture that must be analysed in terms of identity, difference, measurement, and order. Though Des­cartes rejects resemblance, he does so not by excluding the act of com­parison from rational thought, nor even by seeking to limit it, but on the contrary by universalizing it and thereby giving it its purest form. Indeed, it is by means of comparison that we discover 'form, extent, movement and other such things' - that is to say, simple natures - in all subjects in which they may be present. And, moreover, in a deduction of the type 'all of A is B, all of B is C, therefore all of A is C', it is clear that the mind 'makes a comparison between the term sought and the term given, to wit A and C, with'relation to the knowledge that both are B'. In con­sequence, if one makes an exception of the intuition one may have of a single thing, one can say that all knowledge 'is obtained by the comparison of two or more things with each other'[3]. But in fact, there can be no true knowledge except by intuition, that is, by a singular act of pure and attentive intelligence, and by deduction, which links the observed evi­dence together. How then can comparison, which is required for the 52

REPRESENTING

acquisition of almost all knowledge and which, by definition, is neither an isolated observation nor a deduction, stand as an authority for a true thought? 'Almost all the labour accomplished by human reason consists without doubt in rendering this operation possible'[4].   There exist two forms of comparison, and only two: the comparison of measurement and that of order. One can measure sizes or multiplicities, in other words continuous sizes or discontinuous sizes; but in both cases the use of measurement presupposes that, unlike calculation, which proceeds from elements towards a totality, one considers the whole first and then divides it up into parts. This division results in a number of units, of which some are merely conventional or 'borrowed' (in the case of continuous size) and others (in the case of multiplicities or discontinuous sizes) are the units of arithmetic. The comparison of two sizes or two multiplicities requires, in any case, that they both be analysed according to a common unit; so that comparison effected according to measure­ment is reducible, in every case, to the arithmetical relations of equality and inequality. Measurement enables us to analyse like things according to the calculable form of identity and difference [5].

Order, on the other hand, is established without reference to an ex­terior unit: 'I can recognize, in effect, what the order is that exists between A and B without considering anything apart from those two outer terms'; one cannot know the order of things 'in their isolated nature', but by discovering that which is the simplest, then that which is the next simplest, one can progress inevitably to the most complex things of all. Whereas comparison by measurement requires a division to begin from, then the application of a common unit, here, comparison and order are one and the same thing: comparison by means of order is a simple act which enables us to pass from one term to another, then to a third, etc., by means of an 'absolutely uninterrupted'[6] movement. In this way we establish series in which the first term is a nature that we may intuit independently of any other nature; and in which the other terms are established according to increasing differences.

Such, then, are the two types of comparison: the one analyses into units in order to establish relations of equality and inequality; the other estab­lishes elements, the simplest that can be found, and arranges differences according to the smallest possible degrees. Now, it is possible to use the measurement of sizes and multiplicities in establishing an order; arith­metical values can always be arranged according to a series; a multiplicity of units can therefore 'be arranged according to an order such that the

53

THB ORDER OF THINGS

difficulty, which previously lay in the knowing of measurement, comes finally to depend solely on the consideration of order'[7]. And it is pre­cisely in this that the method and its 'progress' consist: the reduction of all measurement (all determination by equality and inequality) to a serial arrangement which, beginning from the simplest, will show up all differences as degrees of complexity. After being analysed according to a given unit and the relations of equality or inequality, the like is analysed according to its evident identity and differences: differences that can be thought in the order of inferences. However, this order or generalized form of comparison can be established only according to its position in the body of our acquired knowledge; the absolute character we recognize in what is simple concerns not the being of things but rather the manner in which they can be known. A thing can be absolute according to one relation yet relative according to others [8]; order can be at once neces­sary and natural (in relation to thought) and arbitrary (in relation to things), since, according to the way in which we consider it, the same thing may be placed at differing points in our order.

All this was of the greatest consequence to Western thought. Resem­blance, which had for long been the fundamental category of knowledge - both the form and the content of what we know - became dissociated in an analysis based on terms of identity and difference; moreover, whether indirectly by the intermediary of measurement, or directly and, as it were, on the same footing, comparison became a function of order; and, lastly, comparison ceased to fulfil the function of revealing how the world is ordered, since it was now accomplished according to the order laid down by thought, progressing naturally from the simple to the complex. As a result, the entire episteme of Western culture found its fundamental arrangements modified. And, in particular, the empirical domain which sixteenth-century man saw as a complex of kinships, resemblances, and affinities, and in which language and things were endlessly interwoven -this whole vast field was to take on a new configuration. This new con­figuration may, I suppose, be called 'rationalism'; one might say, if one's mind is filled with ready-made concepts, that the seventeenth century marks the disappearance of the old superstitious or magical beliefs and the entry of nature, at long last, into the scientific order. But what we must grasp and attempt to reconstitute are the modifications that affected knowledge itself, at that archaic level which makes possible both know­ledge itself and the mode of being of what is to be known.

These modifications may be summed up as follows. First, the substi-

54

                                                REPRESENTING

tution of analysis for the hierarchy of analogies: in the sixteenth century, the fundamental supposition was that of a total system of correspondence (earth and sky, planets and faces, microcosm and macrocosm), and each particular similitude was then lodged within this overall relation. From now on, every resemblance must be subjected to proof by comparison, that is, it will not be accepted until its identity and the scries of its differ­ences have been discovered by means of measurement with a common unit, or, more radically, by its position in an order. Furthermore, the interplay of similitudes was hitherto infinite: it was always possible to discover new ones, and the only limitation came from the fundamental ordering of things, from the finitude of a world held firmly between the macrocosm and the microcosm. A complete enumeration will now be possible: whether in the form of an exhaustive census of all the elements constituting the envisaged whole, or in the form of a categorical arrange­ment that will articulate the field of study in its totality, or in the form of an analysis of a certain number of points, in sufficient number, taken along the whole length of a series. Comparison, then, can attain to perfect certainty: the old system of similitudes, never complete and always open to fresh possibilities, could, it is true, through successive confirmations, achieve steadily increasing probability; but it was never certain. Complete enumeration, and the possibility of assigning at each point the necessary connection with the next, permit an absolutely certain knowledge of identities and differences: 'Enumeration alone, whatever the question to which we are applying ourselves, will permit us always to deliver a true and certain judgement upon it'[9]. The activity of the mind - and this is the fourth point - will therefore no longer consist in drawing things to­gether, in setting out on a quest for everything that might reveal some sort of kinship, attraction, or secretly shared nature within them, but, on the contrary, in discriminating, that is, in establishing their identities, then the inevitability of the connections with all the successive degrees of a series. Ih this sense, discrimination imposes upon comparison the primary and fundamental investigation of difference: providing oneself by intuition with a distinct representation of things, and apprehending clearly the inevitable connection between one element in a series and that which immediately follows it. Lastly, a final consequence, since to know is to discriminate, history and science will become separated from one another. On the one hand there will be erudition, the perusal of written works, the interplay of their authors' opinions; this interplay may well, in some cases, possess an indicative value, not so much because of the agreement

55

THE ORDER OF THINGS

it produces as because of the disagreement: 'When the question at issue is a difficult one, it is more probable that there were few rather than many to discover the truth about it.' Over against this history, and lacking any common unit of measurement with it, are the confident judgements we are able to make by means of intuitions and their serial connection. These and these alone are what constitute science, and even if we had 'read all the arguments of Plato and Aristotle, . . . what we would have learned would not be sciences, it appears, but history' [10]. This being so, the written word ceases to be included among the signs and forms of truth; language is no longer one of the figurations of the world, or a signature stamped upon things since the beginning of time. The manifestation and sign of truth are to be found in evident and distinct perception. It is the task of words to translate that truth if they can; but they no longer have the right to be considered a mark of it. Language has withdrawn from the midst of beings themselves and has entered a period of trans­parency and neutrality.

This is a general phenomenon in seventeenth-century culture - a more general one than the particular fortunes of Cartesianism.

We must, in fact, distinguish between three things. On the one hand, there was the mechanism that, for what was really a fairly short period (not quite the last fifty years of the seventeenth century), offered a theoret­ical model to certain fields of knowledge such as medicine or physiology. There was also an attempt, rather diverse in the forms it took, to mathematicize empirical knowledge; though constant and continuous in the case of astronomy and part of physics, it was only sporadic in other fields -sometimes actually attempted (as with Condorcet), sometimes suggested as a universal ideal and a horizon for research (as with Condillac or Destutt), and sometimes, too, rejected even as a possibility (by Buffon, for example). But neither this endeavour nor the attempts of mechanism should be confused with the relation that all Classical knowledge, in its most general form, maintains with the mathesis, understood as a universal science of measurement and order. Under cover of the empty and ob­scurely incantatory phrases 'Cartesian influence' or 'Newtonian model', our historians of ideas are in the habit of confusing these three things and defining Classical rationalism as the tendency to make nature mechanical and calculable. Others are slightly more perceptive, and go to a great deal of trouble to discover beneath this rationalism a play of'contrary forces':

the forces of nature and life refusing to let themselves be reduced either to algebra or to dynamics, and thus preserving, in the depths of Classicism

56

REPRESENTING

itself, the natural resources of the non-rationalizable. These two forms of analysis are equally inadequate; for the fundamental element of the Classical episteme is neither the success or failure of mechanism, nor the right to mathematicize or the impossibility of mathematicizing nature, but rather a link with the mathesis which, until the end of the eighteenth century, remains constant and unaltered. This link has two essential characteristics. The first is that relations between beings are indeed to be conceived in the form of order and measurement, but with this funda­mental imbalance, that it is always possible to reduce problems of measure­ment to problems of order. So that the relation of all knowledge to the mathesis is posited as the possibility of establishing an ordered succession between things, even non-measurable ones. In this sense, analysis was very quickly to acquire the value of a universal method; and the Leibnizian project of establishing a mathematics of qualitative orders is situated at the very heart of Classical thought; its gravitational centre. But, on the other hand, this relation to the mathesis as a general science of order does not signify that knowledge is absorbed into mathematics, or that the latter becomes the foundation for all possible knowledge; on the con­trary, in correlation with the quest for a mathesis, we perceive the appear­ance of a certain number of empirical fields now being formed and defined for the very first time. In none of these fields, or almost none, is it possible to find any trace of mechanism or mathematicization; and yet they all rely for their foundation upon a possible science of order. Although they were all dependent upon analysis in general, their particular instrument was not the algebraic method but the system of signs. So there first appeared general grammar, natural history, and the analysis of wealth, all sciences of order in the domain of words, beings, and needs; and none of these empirical studies, new in the Classical period and co-extensive with it in duration (their chronological frontiers are marked by Lancelot and Bopp, Ray and Cuvier, Petty and Ricardo, the first group writing around 1660 and the second around 1800-10), could have been founded without the relation that the entire episteme of Western culture maintained at that time with a universal science of order.

This relation to Order is as essential to the Classical age as the relation to Interpretation was to the Renaissance. And just as interpretation in the sixteenth century, with its superimposition of a semiology upon a hermeneutics, was essentially a knowledge based upon similitude, so the odering of things by means of signs constitutes all empirical forms of knowledge as knowledge based upon identity and difference. The

57

THE ORDER OF THINGS

simultaneously endless and closed, full and tautological world of resem­blance now finds itself dissociated and, as it were, split down the middle: on the one side, we shall find the signs that have become tools of analysis, marks of identity and difference, principles whereby things can be re­duced to order, keys for a taxonomy; and, on the other, the empirical and murmuring resemblance of things, that unreacting similitude that lies beneath thought and furnishes the infinite raw material for divisions and distributions. On the one hand, the general theory of signs, divisions, and classifications; on the other, the problem of immediate resemblances, of the spontaneous movement of the imagination, of nature's repetitions. And between the two, the new forms of knowledge that occupy the area opened up by this new split.
Ill THE REPRESENTATION OF THE SIGN

What is a sign in the Classical age? For what was altered in the first half of the seventeenth century, and for a long time to come - perhaps right up to our own day - was the entire organization of signs, the conditions under which they exercise their strange function; it is this, among so many other things one knows or sees, that causes them to emerge sud­denly as signs; it is their very being. On the threshold of the Classical age, the sign ceases to be a form of the world; and it ceases to be bound to what it marks by the solid and secret bonds of resemblance or affinity.

Classical thought defines it according to three variables [11]. First, the certainty of the relation: a sign may be so constant that one can be sure of its accuracy (in the sense that breathing denotes life), but it may also be simply probable (in the sense that pallor probably denotes pregnancy). Second, the type of relation: a sign may belong to the whole that it de­notes (in the sense that a healthy appearance is part of the health it denotes) or be separate from it (in the sense that the figures of the Old Testament are distant signs of the Incarnation and Redemption). Third, the origin of the relation: a sign may be natural (in the sense that a reflection in a mirror denotes that which it reflects) or conventional (in the sense that a word may signify an idea to a given group of men). None of these forms of relation necessarily implies resemblance; even the natural sign does not require that: a cry is a spontaneous sign of fear, but not analogous to it; or again, as Berkeley puts it, visual sensations are signs of touch established in us by God, yet they do not resemble it in any way[12].

58

REPRESENTING

These three variables replace resemblance in defining the sign's efficacity in the domains of empirical knowledge.

l. The sign, since it is always either certain or probable, should find its area of being within knowledge. In the sixteenth century, signs were thought to have been placed upon things so that men might be able to uncover their secrets, their nature or their virtues; but this discovery was merely the ultimate purpose of signs, the justification of their presence; it was a possible way of using them, and no doubt the best; but they did not need to be known in order to exist: even if they remained silent, even if no one were to perceive them, they were just as much there. It was not knowledge that gave them their signifying function, but the very lan­guage of things. From the seventeenth century onward, the whole domain of the sign is divided between the certain and the probable: that is to say, there can no longer be an unknown sign, a mute mark. This is not because men are in possession of all the possible signs, but because there can be no sign until there exists a known possibility of substitution between two known elements. The sign does not wait in silence for the coming of a man capable of recognizing it: it can be constituted only by an act of knowing.

It is here that knowledge breaks off its old kinship with divinatio. The latter always presupposed signs anterior to it: so that knowledge always resided entirely in the opening up of a discovered, affirmed, or secretly transmitted, sign. Its task was to uncover a language which God had previously distributed across the face of the earth; it is in this sense that it was the divination of an essential implication, and that the object of its divination was divine. From now on, however, it is within knowledge itself that the sign is to perform its signifying function; it is from know­ledge that it will borrow its certainty or its probability. And though God still employs signs to speak to us through nature, he is making use of our knowledge, and of the relations that are set up between our im­pressions, in order to establish in our minds a relation of signification. Such is the role of feeling in Malebranche or of sensation in Berkeley; in natural judgement, in feeling, in visual impressions, and in the perception of the third dimension, what we are dealing with are hasty and confused, but pressing, inevitable, and obligatory kinds of knowledge serving as signs for discursive kinds of knowledge which we humans, because we are not pure intelligences, no longer have the time or the permission to attain to ourselves and by the unaided strength of our own minds. In Malebranche and Berkeley, the sign arranged by God is the cunning and

59

THE ORDER OF THINGS

thoughtful superimposition of two kinds of knowledge. There is no longer any divinatio involved - no insertion of knowledge in the enig­matic, open, and sacred area of signs - but a brief and concentrated kind of knowledge: the contraction of a long sequence of judgements into the rapidly assimilated form of the sign. And it will also be seen how, by a reversal of direction, knowledge, having enclosed the signs within its own space, is now able to accommodate probability: between one im­pression and another the relation will be that of sign to signified, in other words, a relation which, like that of succession, will progress from the weakest probability towards the greatest certainty.

The connection of ideas does not imply the relation of cause and effect, but only of a mark or sign with the thing signified. The fire which I sec is not the cause of the pain I suffer upon my approaching it, but the mark that forewarns me of it [13].

The knowledge that divined, at random, signs that were absolute and older than itself has been replaced by a network of signs built up step by step in accordance with a knowledge of what is probable. Hume has become possible.

2. The second variable of the sign: the form of its relation with what it signifies. By means of the interplay of conveniency, emulation, and above all sympathy, similitude was able in the sixteenth century to triumph over space and time; for it was within the power of the sign to draw things together and unite them. "With the advent of Classical thought, on the other hand, the sign becomes characterized by its essential dispersion. The circular world of converging signs is replaced by an infinite pro­gression. Within this space, the sign can have one of two positions: either it can be claimed, as an element, to be part of that which it serves to designate; or else it is really and actually separated from what it serves to designate. The truth is, however, that this alternative is not a radical one, since the sign, in order to function, must be simultaneously an insertion in that which it signifies and also distinct from it. For the sign to be, in effect, what it is, it must be presented as an object of knowledge at the same time as that which it signifies. As Condillac points out, a sound could never become the verbal sign of something for a child unless the child had heard it at least once at the moment of perceiving the objcct[14]. But if one element of a perception is to become a sign for it, it is not enough merely for that element to be part of the perception; it must be differentiated qua element and be distinguished from the total impression

60

REPRESENTING

with which it is confusedly linked; consequently, that total impression itself must have been divided up, and attention must have been directed towards one of the intermingled regions composing it, in order to isolate one of them. The constitution of the sign is thus inseparable from analysis. Indeed, it is the result of it, since without analysis the sign could not be­come apparent. But it is also the instrument of analysis, since once defined and isolated it can be applied to further impressions; and in relation to them it plays the role of a grid, as it were. Because the mind analyses, the sign appears. Because the mind has signs at its disposal, analysis never ceases. It is understandable why, from Condillac to Destutt de Tracy and Gerando, the general theory of signs and the definition of the power of analysis of thought were so exactly superimposed to form a single and unbroken theory of knowledge.

When the Logique de Port-Royal states that a sign can be inherent in what it designates or separate from it, it is demonstrating that the sign, in the Classical age, is charged no longer with the task of keeping the world close to itself and inherent in its own forms, but, on the contrary, with that of spreading it out, of juxtaposing it over an indefinitely open surface, and of taking up from that point the endless deployment of the substitutes in which we conceive of it. And it is by this means that it is offered simultaneously to analysis and to combination, and can be ordered from beginning to end. The sign in Classical thought does not erase dis­tances or abolish time: on the contrary, it enables one to unfold them and to traverse them step by step. It is the sign that enables things to become distinct, to preserve themselves within their own identities, to dissociate themselves or bind themselves together. Western reason is entering the age of judgement.

3. There remains a third variable: the one that can assume the two values of nature and of convention. It had long been known - and well before Plato's Cratylus - that signs can be either given by nature or established by man. Nor was the sixteenth century ignorant of this fact, since it recognized human languages to be instituted signs. But the artificial signs owed their power only to their fidelity to natural signs. These latter, even at a remove, were the foundation of all others. From

the seventeenth century, the values allotted to nature and convention in this field are inverted: if natural, a sign is no more than an element selected from the world of things and constituted as a sign by our know­ledge. It is therefore strictly limited, rigid, inconvenient, and impossible for the mind to master. When, on the other hand, one establishes a

61

THE ORDER OF THINGS

conventional sign, it is always possible (and indeed necessary) to choose it in such a way that it will be simple, easy to remember, applicable to an indefinite number of elements, susceptible of subdivision within itself and of combination with other signs; the man-made sign is the sign at the peak of its activity. It is the man-made sign that draws the dividing-line between man and animal; that transforms imagination into voluntary memory, spontaneous attention into reflection, and instinct into rational knowledge [15]. It is also what Itard found lacking in the 'wild man of Aveyron'[i6]. Natural signs are merely rudimentary sketches for these conventional signs, the vague and distant design that can be realized only by the establishment of arbitrariness.

But this arbitrariness is measured by its function; and has its rules very exactly defined by that function. An arbitrary system of signs must per­mit the analysis of things into their simplest elements; it must be capable of decomposing them into their very origins; but it must also demon­strate how combinations of those elements are possible, and permit the ideal genesis of the complexity of things. 'Arbitrary' stands in opposition to 'natural' only if one is attempting to designate the manner in which signs have been established. But this arbitrariness is also the grid of analysis and the combinative space through which nature is to posit itself as that which it is - at the level of primal impressions and in all the possible forms of their combination. In its perfect state, the system of signs is that simple, absolutely transparent language which is capable of naming what is elementary; it is also that complex of operations which defines all possible conjunctions. To our eyes, this search for origins and this calculus of combinations appear incompatible, and we are only too ready to in­terpret them as an ambiguity in seventeenth- and eighteenth-century thought. The same is true of the interaction between the system and nature. In fact, there is no contradiction at all for thought at that time. More precisely, there exists a single, necessary arrangement running through the whole of the Classical episteme: the association of a universal calculus and a search for the elementary within a system that is artificial and is, for that very reason, able to make nature visible from its primary elements right to the simultaneity of all their possible combinations. In the Classical age, to make use of signs is not, as it was in preceding centuries, to attempt to rediscover beneath them the primitive text of a discourse sustained, and retained, forever; it is an attempt to discover the arbitrary language that will authorize the deployment of nature within its space, the final terms of its analysis and the laws of its composition. It is no longer the task of

62

REPRESENTING

knowledge to dig out the ancient Word from the unknown places where it may be hidden; its job now is to fabricate a language, and to fabricate it well - so that, as an instrument of analysis and combination, it will really be the language of calculation.

It is now possible to define the instruments laid down for the use of Classical thought by the sign system. It was this system that introduced into knowledge probability, analysis, and combination, and the justified arbitrariness of the system. It was the sign system that gave rise simul­taneously to the search for origins and to calculability; to the constitution of tables that would fix the possible compositions, and to the restitution of a genesis on the basis of the simplest elements; it was the sign system that linked all knowledge to a language, and sought to replace all languages with a system of artificial symbols and operations of a logical nature. At the level of the history of opinions, all this would appear, no doubt, as a tangled network of influences in which the individual parts played by Hobbes, Berkeley, Leibniz, Condillac, and the 'Ideologues' would be re­vealed. But if we question Classical thought at the level of what, archaeologically, made it possible, we perceive that the dissociation of the sign and resemblance in the early seventeenth century caused these new forms -probability, analysis, combination, and universal language system - to emerge, not as successive themes engendering one another or driving one another out, but as a single network of necessities. And it was this net­work that made possible the individuals we term Hobbes, Berkeley, Hume, or Condillac.
IV DUPLICATED REPRESENTATION

However, the property of signs most fundamental to the Classical episteme has not yet been mentioned. Indeed, the very fact that the sign can be more or less probable, more or less distant from what it signifies, that it can be either natural or arbitrary, without its nature or its value as a sign being affected - all this shows clearly enough that the relation of the sign to its content is not guaranteed by the order of things in them­selves. The relation of the sign to the signified now resides in a space in which there is no longer any intermediary figure to connect them: what connects them is a bond established, inside knowledge, between the idea of one thing and the idea of another. The Logique de Port-Royal states this as follows: 'The sign encloses two ideas, one of the thing representing, the Other of the thing represented; and its nature consists in exciting the first 63

THE ORDER OF THINGS

by means of the second'[17]. This dual theory of the sign is in unequi­vocal opposition to the more complex organization of the Renaissance; at that time, the theory of the sign implied three quite distinct elements: that which was marked, that which did the marking, and that which made it possible to see in the first the mark of the second; and this last element was, of course, resemblance: the sign provided a mark exactly in so far as it was 'almost the same thing' as that which it designated. It is this unitary and triple system that disappears at the same time as 'thought by resemblance', and is replaced by a strictly binary organization.

But there is one condition that must be fulfilled if the sign is indeed to be this pure duality. In its simple state as an idea, or an image, or a per­ception, associated with or substituted for another, the signifying element is not a sign. It can become a sign only on condition that it manifests, in addition, the relation that links it to what it signifies. It must represent;

but that representation, in turn, must also be represented within it. This is a condition indispensable to the binary organization of the sign, and one that the Logique de Port-Royal sets forth even before telling us what a sign is: 'When one looks at a certain object only in so far as it represents another, the idea one has of it is the idea of? sign, and that first object is called a sign'[18]. The signifying idea becomes double, since superim­posed upon the idea that is replacing another there is also the idea of its representative power. This appears to give us three terms: the idea signi­fied, the idea signifying, and, within this second term, the idea of its role as representation. What we are faced with here is not, however, a surrep­titious return to a ternary system, but rather an inevitable displacement within the two-term figure, which moves backward in relation to itself and comes to reside entirely within the signifying element. In fact, the signifying element has no content, no function, and no determination other than what it represents: it is entirely ordered upon and transparent to it. But this content is indicated only in a representation that posits itself as such, and that which is signified resides, without residuum and without opacity, within the representation of the sign. It is characteristic that the first example of a sign given by the Logique de Port-Royal is not the word, nor the cry, nor the symbol, but the spatial and graphic representation - the drawing as map or picture. This is because the pic­ture has no other content in fact than that which it represents, and yet that content is made visible only because it is represented by a representa­tion. The binary arrangement of the sign, as it appears in the seventeenth century, replaces an organization which, in different modes, had been

64

REPRESENTING

ternary ever since the time of the Stoics, and even since the first Greek grammarians; and this new binary arrangement presupposes that the sign is a duplicated representation doubled over upon itself. An idea can be the sign of another, not only because a bond of representation can be established between them, but also because this representation can always be represented within the idea that is representing. Or again, because representation in its peculiar essence is always perpendicular to itself: it is at the same time indication and appearance; a relation to an object and a manifestation of itself. From the Classical age, the sign is the representativity of the representation in so far as it is representable.

This has very considerable consequences. First, the importance of signs in Classical thought. Before, they were means of knowing and the keys to knowledge; now, they are co-extensive with representation, that is, with thought as a whole; they reside within it but they run through its entire extent. Whenever one representation is linked to another and rep­resents that link within itself, there is a sign: the abstract idea signifies the concrete perception from which it has been formed (Condillac); the general idea is no more than a particular idea serving as a sign for other particular ideas (Berkeley); imaginings arc signs of the perceptions from which they arose (Hume, Condillac); sensations are signs of one another (Berkeley, Condillac); and, finally, it is possible that sensations may them­selves be (as in Berkeley) signs of what God wishes to tell us, which would make them, as it were, signs for a complex of signs. Analysis of represen­tation and the theory of signs interpenetrate one another absolutely; and when the day came, at the end of the eighteenth century, for Ideology to raise the question of whether the idea or the sign should be accorded primacy, when Destutt could reproach Gerando for having created a theory of signs before defining the idea[19], this meant that their immediate link was already becoming confused, and that idea and sign would soon cease to be perfectly transparent to one another.

A second consequence: this universal extension of the sign within the field of representation precludes even the possibility of a theory of signification. For to ask ourselves questions about what signification is pre­supposes that it is a determinate form in our consciousness. But if pheno­mena are posited only in a representation that, in itself and because of its own representability, is wholly a sign, then signification cannot consti­tute a problem. Moreover, it is not even visible. All representations are interconnected as signs; all together, they form, as it were, an immense

65

THE ORDBR OF THINGS

network; each one posits itself in its transparency as the sign of what it represents; and yet - or rather, by this very fact - no specific activity of consciousness can ever constitute a signification. No doubt it is because Classical thought about representation excludes any analysis of significa­tion that we today, who conceive of signs only upon the basis of such an analysis, have so much trouble, despite the evidence, in recognizing that Classical philosophy, from Malebranche to Ideology, was through and through a philosophy of the sign.

No meaning exterior or anterior to the sign; no implicit presence of a previous discourse that must be reconstituted in order to reveal the autochthonous meaning of things. Nor, on the other hand, any act constitutive of signification or any genesis interior to consciousness. This is because there is no intermediary element, no opacity intervening between the sign and its content. Signs, therefore, have no other laws than those that may govern their contents: any analysis of signs is at the same time, and without need for further inquiry, the decipherment of what they are trying to say. Inversely, the discovery of what is signified is nothing more than a reflection upon the signs that indicate it. As in the sixteenth century, 'semiology' and 'hermeneutics' are superimposed - but in a different form. In the Classical age they no longer meet and join in the third element of resemblance; their connection lies in that power proper to representation of representing itself. There will therefore be no theory of signs separate and differing from an analysis of meaning. Yet the system does grant a certain privilege to the former over the latter; since it does not accord that which is signified a nature different from that accorded to the sign, meaning cannot be anything more than the totality of the signs arranged in their progression; it will be given in the complete table of signs. But, on the other hand, the complete network of signs is linked together and articulated according to patterns proper to meaning. The table of the signs will be the image of the things. Though the meaning itself is entirely on the side of the sign, its functioning is entirely on the side of that which is signified. This is why the analysis of language, from Lancelot to Destutt de Tracy, is conducted on the basis of an abstract theory of verbal signs and in the form of a general grammar: but it always takes the meaning of words as its guiding thread; it is also why natural history manifests itself as an analysis of the characters of living beings, and why, nevertheless, the taxonomies used, artificial though they may be, are always intended to unite with the natural order, or at least to dissociate it as little as possible; it is also why the analysis of wealth

66

REPRESENTING

is conducted on the basis of money and exchange, but value is always based upon need. In the Classical age, the pure science of signs has value as the direct discourse of that which is signified.

Finally, a third consequence, which probably extends up to our own time: the binary theory of the sign, the theory upon which the whole general science of the sign has been founded since the seventeenth century, is linked according to a fundamental relation with a general theory of representation. If the sign is the pure and simple connection between what signifies and what is signified (a connection that may be arbitrary or not, voluntary or imposed, individual or collective), then the relation can be established only within the general element of representation: the sig­nifying element and the signified element are linked only in so far as they are (or have been or can be) represented, and in so far as the one actually represents the other. It was therefore necessary that the Classical theory of the sign should provide itself with an 'ideology' to serve as its foundation and philosophical justification, that is, a general analysis of all forms of representation, from elementary sensation to the abstract and complex idea. It was also necessary that Saussure, rediscovering the project of a general semiology, should have given the sign a definition 'that could seem 'psychologistic' (the linking of a concept and an image): this is because he was in fact rediscovering the Classical condition for con­ceiving of the binary nature of the sign.
V THE IMAGINATION OF RESEMBLANCE

So signs are now set free from that teeming world throughout which the 'Renaissance had distributed them. They are lodged henceforth within the confines of representation, in the interstices of ideas, in that narrow space in which they interact with themselves in a perpetual state of decom­position and recomposition. As for similitude, it is now a spent force, outside the realm of knowledge. It is merely empiricism in its most un­refined form; like Hobbes, one can no longer 'regard it as being a part of philosophy', unless it has first been erased in its inexact form of re­semblance and transformed by knowledge into a relationship of equality or order. And yet similitude is still an indispensable border of knowledge. For no equality or relation of order can be established between two things Unless their resemblance has at least occasioned their comparison. Hume placed the relation of identity among those 'philosophical' relations that presuppose reflection; whereas, for him, resemblance belonged to natural

67

TUB ORDBR OF THINGS

relations, to those that constiain our minds by means of an inevitable but 'calm force'.

Let the philosopher pride himself on his precision as much as he will... I nevertheless dare defy him to make a single step in his progress without the aid of resemblance. Throw but one glance upon the metaphysical aspect of the sciences, even the least abstract of them, and then tell me whether the general inductions that are derived from particular facts, or rather the kinds themselves, the species and all abstract notions, can be formed otherwise than by means of resemblance [20].

At the border of knowledge, similitude is that barely sketched form, that rudimentary relation which knowledge must overlay to its full extent, but which continues, indefinitely, to reside below knowledge in the manner of a mute and ineffaceable necessity.

As in the sixteenth century, resemblance and sign respond inevitably to one another, but in a new way. Whereas similitude once required a mark in order for its secret to be uncovered, it is now the undifferentiated, shifting, unstable base upon which knowledge can establish its relations, its measurements, and its identities. This results in a double reversal: first, because it is the sign - and with it the whole of discursive knowledge -that requires a basis of similitude, and, second, because it is no longer a question of making a previous content manifest to knowledge but of providing a content that will be able to offer a ground upon which forms of knowledge can be applied. Whereas in the sixteenth century resem­blance was the fundamental relation of being to itself, and the hinge of the whole world, in the Classical age it is the simplest form in which what is to be known, and what is furthest from knowledge itself, appears. It is through resemblance that representation can be known, that is, com­pared with other representations that may be similar to it, analysed into elements (elements common to it and other representations), combined with those representations that may present partial identities, and finally laid out into an ordered table. Similitude in Classical philosophy (that is, in a philosophy of analysis) plays a role parallel to that which will be played by diversity in critical thought and the philosophies of judgement.

In this limiting and conditional position (that without which and beyond which one cannot know), resemblance is situated on the side of imagin­ation, or, more exactly, it can be manifested only by virtue of imagina­tion, and imagination, in turn, can be exercised only with the aid of resemblance. And, in effect, if we suppose in the uninterrupted chain of

68

REPRESENTING

representation certain impressions, the very simplest that can be, without the slightest degree of resemblance between them, then there would be no possibility whatever of the second recalling the first, causing it to reappear, and thus authorizing its representation in the imagination; those im­pressions would succeed one another in the most total differentiation - so total that it could not even be perceived, since no representation would be able to immobilize itself in one place, reanimate a former one, and juxtapose itself to it so as to give rise to a comparison; even that tiny overlap of identity necessary for all differentiation would not be provided. Perpetual change would pass before'us without guidelines and in per­petual monotony. If representation did not possess the obscure power of making a past impression present once more, then no impression would ever appear as either similar to or dissimilar from a previous one. This power of recall implies at least the possibility of causing two impressions to appear as quasi-likenesses (as neighbours or contemporaries, existing in almost the same way) when one of those impressions only is present, while the other has ceased, perhaps a long time ago, to exist. Without imagination, there would be no resemblance between things.

The double requisite is patent. There must be, in the things represented, the insistent murmur of resemblance; there must be, in the representation, the perpetual possibility of imaginative recall. And neither of these requisites can dispense with the other, which completes and confronts it. Hence the two directions of analysis followed throughout the Classical age, consistently drawing closer and closer together until finally, in the second half of the eighteenth century, they were able to express their common truth in Ideology. On the one hand, we find the analysis that provides an account of the inversion of the scries of representations to form a non-actual but simultaneous table of comparisons: the analysis of im­pressions, of reminiscence, of imagination, of memory, of all that in­voluntary background which is, as it were, the mechanics of the image in time. And, on the other hand, there is the analysis that gives an account of the resemblance between things - of their resemblance before their reduction to order, their decomposition into identical and different ele•ments, the tabular redistribution of their unordered similitudes. Why is it, then, that things are given in an overlapping mixture, in an interpenetrating jumble in which their essential order is confused, yet still visible enough to show through in the form of resemblances, vague similitudes, and allusive opportunities for a memory on the alert? The first series of problems corresponds roughly with the analytic of imagination,

69

THE ORDER OF THINGS

as a positive power to transform the linear time of representation into a simultaneous space containing virtual elements; the second corresponds roughly with the analysis of nature, including the lacunae, the disorders that confuse the tabulation of beings and scatter it into a series of representations that vaguely, and from a distance, resemble one another.

Now, these two opposing stages (the first the negative one of the dis­order in nature and in our impressions, the other the positive one of the power to reconstitute order out of those impressions) are united in the idea of a'genesis'. And this in two possible ways. Either the negative stage (that of disorder and vague resemblance) is attributed to the imagination itself, which then exercises a double function: if it is able to restore order solely by duplicating representation, it is able to do so only in so far as it would prevent us from perceiving directly, and in their analytic truth, the identities and differences of things. The power of imagination is only the inverse, the other side, of its defect. It exists within man, at the suture of body and soul. It is there that Descartes, Malebranche, and Spinoza analysed it, both as the locus of error and as the power of attaining to truth, even mathematical truth; they recognized in it the stigma of finitude, whether as the sign of a fall outside the area of intelligibility or as the mark of a limited nature. Alternatively, the positive stage of im­agination can be attributed to shifting resemblances and the vague mur­mur of similitudes. It is the disorder of nature due to its own history, to its catastrophes, or perhaps merely to its jumbled plurality, which is no longer capable of providing representation with anything but things that resemble one another. So that representation, perpetually bound to con­tents so very close to one another, repeats itself, recalls itself, duplicates itself quite naturally, causes almost identical impressions to arise again and again, and engenders imagination. It was in just this proliferation of a nature that is multiple, yet obscurely and irrationally re-created, in the enigmatic fact of a nature that prior to all order resembles itself, that Condillac and Hume sought for the link between resemblance and imagination. Their solutions were strictly contradictory, but they were both answers to the same problem. It is in any case understandable that the second type of analysis should have so easily been deployed in the mythical form of the first man (Rousseau), or that of the awakening consciousness (Condillac), or that of the stranger suddenly thrust into the world (Hume): this genesis functioned exactly instead of and in place of Genesis itself.

70

REPRESENTING

One further remark. Though the notions of nature and human nature have a certain importance in the Classical age, this is not because the hidden and inexhaustibly rich source of power which we call nature had suddenly been discovered as a field for empirical inquiry; nor is it be­cause a tiny, singular, and complex subregion called human nature had been isolated within this vast field of nature. In fact, these two concepts function in such a way as to guarantee the kinship, the reciprocal bond, between imagination and resemblance. It is true that imagination is apparently only one of the properties of human nature, and resemblance one of the effects of nature; but if we follow the archaeological network that provides Classical thought with its laws, we see quite clearly that human nature resides in that narrow overlap of representation which permits it to represent itself to itself (all human nature is there: just enough outside representation for it to present itself again, in the blank space that separates the presence of representation and the 're-' of its repetition); and that nature is nothing but the impalpable confusion within representation that makes the resemblance there perceptible before the order of the identities is yet visible. Nature and human nature, within the general configuration of the episteme, permit the reconciliation of resemblance and imagination that provides a foundation for, and makes possible, all the empirical sciences of order.

In the sixteenth century, resemblance was linked to a system of signs; and it was the interpretation of those signs that opened up the field of concrete knowledge. From the seventeenth century, resemblance was pushed out to the boundaries of knowledge, towards the humblest and basest of its frontiers. There, it links up with imagination, with doubtful repetitions, with misty analogies. And instead of opening up the way to a science of interpretation, it implies a genesis that leads from those un­refined forms of the Same to the great tables of knowledge developed according to the forms of identity, of difference, and of order. The project of a science of order, with a foundation such as it had in the seventeenth century, carried the implication that it had to be paralleled by an accom­panying genesis of consciousness, as indeed it was, effectively and uninterruptedly, from Locke to the 'Ideologues'.
VI MATHESIS AND ' TAXINOMIA'

The project of a general science of order; a theory of signs analysing representation; the arrangement of identities and differences into ordered

71

THE ORDER OF THINGS

tables: these constituted an area of empiricity in the Classical age that had not existed until the end of the Renaissance and that was destined to dis­appear early in the nineteenth century. It is so difficult for us to reinstate now, and so thickly overlaid by the system of positivities to which our own knowledge belongs, that it has for long passed unpcrccived. It is distorted and masked by the use of categories and patterns that are our own. An attempt is apparently being made to reconstitute what the 'sciences of life', of 'nature' or 'man', were, in the seventeenth and eight­eenth centuries, while it is quite simply forgotten that man and life and nature arc none of them domains that present themselves to the curiosity of knowledge spontaneously and passively.

What makes the totality of the Classical episteme possible is primarily the relation to a knowledge of order. When dealing with the ordering of simple natures, one has recourse to a mathesis, of which the universal method is algebra. When dealing with the ordering of complex natures (representations in general, as they are given in experience), one has to constitute a taxinomia, and to do that one has to establish a system of signs. These signs are to the order of composite natures what algebra is to the order of simple natures. But in so far as empirical representations must be analysable into simple natures, it is clear that the taxinomia relates wholly to the mathesis; on the other hand, since the perception of proofs is only one particular case of representation in general, one can equally well say that mathesis is only one particular case of taxinomia. Similarly, the signs established by thought itself constitute, as it were, an algebra of complex representations; and algebra, inversely, is a method of providing simple natures with signs and of operating upon those signs. We therefore have the arrangement shown below:

But that is not all. Taxinomia also implies a certain continuum of things (a non-discontinuity, a plenitude of being) and a certain power of the imagination that renders apparent what is not, but makes possible, by this very fact, the revelation of that continuity. The possibility of a science of empirical orders requires, therefore, an analysis of knowledge - an

72

REPRESENTING

analysis that must show how the hidden (and as it were confused) con­tinuity of being can be reconstituted by means of the temporal connection provided by discontinuous representations. Hence the necessity, constantly manifested throughout the Classical age, of questioning the origin of knowledge. In fact, these empirical analyses are not in opposition to the project of a universal mathesis, in the sense that scepticism is to rational­ism; they were already included in the requisites of a knowledge that is no longer posited as experience of the Same but as the establishment of Order. Thus, at the two extremities of the Classical episteme, we have a mathesis as the science of calculable order and a genesis as the analysis of the constitution of orders on the basis of empirical series. On the one hand, we have a utilization of the symbols of possible operations upon identities and differences; on the other, we have an analysis of the marks progres­sively imprinted in the mind by the resemblances between things and the retrospective action of imagination. Between the mathesis and the genesis there extends the region of signs - of signs that span the whole domain of empirical representation, but never extend beyond it. Hedged in by calculus and genesis, we have the area of the table. This kind of knowledge involves the allotting of a sign to all that our representation can present us with: perceptions, thoughts, desires; these signs must have a value as characters, that is, they must articulate the representation as a whole into distinct subregions, all separated from. one another by assignable character­istics; in this way they authorize the establishment of a simultaneous system according to which the representations express their proximity and their distance, their adjacency and their separateness - and therefore the network, which, outside chronology, makes patent their kinship and reinstates their relations of order within a permanent area. In this manner the table of identities and differences may be drawn up.

It is in this area the we encounter natural history - the science of the characters that articulate the continuity and the tangle of nature. It is also in this area that we encounter the theory of money and the theory of value - the science of the signs that authorize exchange and permit the establish­ment of equivalences between men's needs or desires. Lastly, it is also in this region that we find general grammar - the science of the signs by means of which men group together their individual perceptions and pattern the continuous flow of their thoughts. Despite their differences, these three domains existed in the Classical age only in so far as the fundamental area of the ordered table was established between the cal­culation of equalities and the genesis of representations.

73

THE ORDER OF THINGS

It is patent that these three notions - mathesis, taxinomia, genesis - desig­nate not so much separate domains as a solid grid of kinships that defines the general configuration of knowledge in the Classical age. Taxinomia is not in opposition to mathesis: it resides within it and is distinguished from it; for it too is a science of order - a qualitative mathesis. But under­stood in the strict sense mathesis is a science of equalities, and therefore of attributions and judgements; it is the science of truth. Taxinomia, on the other hand, treats of identities and differences; it is the science of articulations and classifications; it is the knowledge of beings. In the same way, genesis is contained within taxinomia, or at least finds in it its primary possibility. But taxinomia establishes the table of visible differences;

genesis presupposes a progressive series; the first treats of signs in their spatial simultaneity, as a syntax; the second divides them up into an analogon of time, as a chronology. In relation to mathesis, taxinomia functions as an ontology confronted by an apophantics; confronted by genesis, it functions as a semiology confronted by history. It defines, then, the general law of beings, and at the same time the conditions under which it is possible to know them. Hence the fact that the theory of signs in the Classical period was able to support simultaneously both a science with a dogmatic approach, which purported to be a knowledge of nature itself, and a philosophy of representation, which, in the course of time, became more and more nominalist and more and more sceptical. Hence, too, the fact that such an arrangement has disappeared so completely that later ages have lost even the memory of its existence; this is because after the Kantian critique, and all that occurred in Western culture at the end of the eighteenth century, a new type of division was established: on the one hand mathesis was regrouped so as to constitute an apophantics and an ontology, and it is in this form that it has dominated the formal disciplines right up to our day; on the other hand, history and semiology (the latter absorbed, moreover, by the former) united to form those interpretative disciplines whose power has extended from Schleiermacher to Nietzsche and Freud.

In any case, the Classical episteme can be defined in its most general arrangement in terms of the articulated system of a mathesis, a taxinomia, and a genetic analysis. The sciences always carry within themselves the project, however remote it may be, of an exhaustive ordering of the world;

they are always directed, too, towards the discovery of simple elements and their progressive combination; and at their centre they form a table on which knowledge is displayed in a system contemporary with itself.

74

REPRESENTING

The centre of knowledge, in the seventeenth and eighteenth centuries, is the table. As for the great controversies that occupied men's minds, these arc accommodated quite naturally in the folds of this organization.

It is quite possible to write a history of thought in the Classical period using these controversies as starting-points or themes. But one would then be writing only a history of opinions, that is, of the choices operated according to individuals, environments, social groups; and a whole method of inquiry is thereby implied. If one wishes to undertake an archaeological analysis of knowledge itself, it is not these celebrated controversies that ought to be used as the guidelines and articulation of such a project. One must reconstitute the general system of thought whose network, in its positivity, renders an interplay of simultaneous and ap­parently contradictory opinions possible. It is this network that defines the conditions that make a controversy or problem possible, and that bears the historicity of knowledge. If the Western world did battle with itself in order to know whether life was nothing but movement or whether nature was sufficiently well ordered to prove the existence of God, it was not because a problem had been opened up; it was because, after dispersing the undefined circle of signs and resemblances, and before organizing the series of causality and history, the episteme of Western culture had opened up an area to form a table over which it wandered endlessly, from the calculable forms of order to the analysis of the most complex representations. And we see the marks of this movement on the historical surface of the themes, controversies, problems, and preferences of opinion. Acquired learning spanned from one end to the other a 'space of know­ledge' which had suddenly appeared in the seventeenth century and which was not to be closed again until a hundred and fifty years later.

We must now undertake the analysis of this tabulated space, in those subregions in which it is visible in its clearest form, that is, in the theories of language, classification, and money.

It may be objected that the mere fact of attempting to analyse general grammar, natural history, and economics simultaneously and en bloc - by relating them to a general theory of signs and representation - presupposes a question that could originate only in our own century. It is true that the Classical age was no more able than any other culture to circumscribe or name its own general system of knowledge. But that system was in fact sufficiently constricting to cause the visible forms of knowledge to trace their kinships upon it themselves, as though methods, concepts, types of analysis, acquired experiences, minds, and finally men themselves,

75

THB ORDER OF THINGS

had all been displaced at the behest of a fundamental network defining the implicit but inevitable unity of knowledge. History has provided us with innumerable examples of these displacements. The connecting paths between the theories of knowledge, of signs, and of grammar were trodden so many times: Port-Royal produced its Grammaire as a comple­ment and natural sequel to its Logique, the former being connected to the latter by a common analysis of signs; Condillac, Destutt de Tracy, and Gerando articulated one upon the other the decomposition of knowledge into its conditions or 'elements', and the reflection upon those signs of which language forms only the most visible application and use. There is also a well-trodden connection between the analysis of representation and signs and the analysis of wealth: Quesnay the physiocrat wrote the article on 'evidence' for the Encyclopedie; Condillac and Destutt included in their theory of knowledge and language that of trade and economics, which for them possessed political and also moral value; it is well known that Turgot wrote the article on 'etymologic' for the Encyclopedic and the first systematic parallel between money and words; that Adam .Smith, in addition to his great work on economics, wrote a treatise on the origin of languages. There is a connecting path between the theory of natural classifications and theories of language: Adanson did not merely attempt to create, in the botanical field, a nomenclature that was both artificial and coherent; he aimed at (and in part carried out) a whole reorganization of writing in terms of the phonetic data of language;

Rousseau left among his posthumous works some rudiments of botany and a treatise on the origin of languages.

Such, traced out, as it were, in dotted lines, was the great grid of empiri­cal knowledge: that of non-quantitative orders. And perhaps the de­ferred but insistent unity of a Taxinomia universalis appeared in all clarity in the work of Linnaeus, when he conceived the project of discovering in all the concrete domains of nature or society the same distributions and the same order[21]. The limit of knowledge would be the perfect trans­parency of representations to the signs by which they are ordered.

NOTES

[1] Descartes, CEuvres philosophiques (Paris, 1963 edn., t.1, p. 77).

[2] F. Bacon, Novum Organum (1620, book I, xlv and lix).

[3] Descartes, Regulae, XIV, p. 168.

[4] Ibid., XIV, p. 168.

[5] Ibid., XIV, p. 182.



76

REPRESENTING

[6] Ibid., VI, p. 102; VII, p. 109.

[7] Ibid., XIV, p. 182.

[8] Ibid., VI, p. 103.

[9] Ibid., VII, p. no.

[10] Ibid., Ill, p. 86.

[11] Logique de Port-Royal, lere partie, chap. IV.

[12] G. Berkeley, An essay towards a new theory of vision (1709, CXLVII).

[13] G. Berkeley, A treatise concerning the principles of human knowledge (1710, LXV).

[14] Condillac, Essai sur I'origine des connaissances humaines (CEuvres, Paris, 1798,1.1, pp. 188-208). [15] Ibid., p. 75.

[16] J. Itard, Rapport sur les nouvcaux developpements de Victor de I'Aveyron (1806); reprinted in L. Malson, Les Enfants sauvages (Paris, 1964). [17] Logique de Port-Royal, lere partie, chap. IV. [18]Ibid.

[19] Destutt de Tracy, Elements d'ldeologie (Paris, year XI, t. II, p. 1).

[20] Merian, Reflexions philosophiques sur la ressemblance (1767, pp. 3 and 4).

 [21] Linnaeus, Philosophie botanique, sections 155 and 256.



77

CHAPTER 4
Speaking
I CRITICISM AND COMMENTARY

The existence of language in the Classical age is both pre-eminent and unobtrusive.

Pre-eminent, because words have been allotted the task and the power of 'representing thought'. But representing in this case does not mean translating, giving a visible version of, fabricating a material double that will be able, on the external surface of the body, to reproduce thought in its exactitude. Representing must be understood in the strict sense: language represents thought as thought represents itself. To constitute language or give it life from within, there is no essential and primitive act of signification, but only, at the heart of representation, the power that it possesses to represent itself, that is, to analyse itself by juxtaposing itself to itself, part by part, under the eye of reflection, and to delegate it­self in the form of a substitute that will be an extension of it. In the Classi­cal age, nothing is given that is not given to representation; but, by that very fact, no sign ever appears, no word is spoken, no proposition is ever directed at any content except by the action of a representation that stands back from itself, that duplicates and reflects itself in another representa­tion that is its equivalent. Representations are not rooted in a world that gives them meaning; they open of themselves on to a space that is their own, whose internal network gives rise to meaning. And language exists in the gap that representation creates for itself. Words do not, then, form a thin film that duplicates thought on the outside; they recall thought, they indicate it, but inwards first of all, among all those representations that represent other representations. The language of the Classical age is much closer to the thought it is charged with expressing than is generally supposed; but it is not parallel to it; it is caught in the grid of thought, woven into the very fabric it is unrolling. It is not an exterior effect of thought, but thought itself.

78

SPEAKING

And, because of this, it makes itself invisible, or almost so. In any case, it has become so transparent to representation that its very existence ceases to be a problem. The Renaissance came to a halt before the brute fact that language existed: in the density of the world, a graphism mingling with things or flowing beneath them; marks made upon manuscripts or the pages of books. And all these insistent marks summoned up a secon­dary language - that of commentary, exegesis, erudition - in order to stir the language that lay dormant within them and to make it speak at last; the existence of language preceded, as if by a mute stubbornness, what one could read in it and the words that gave it sound. From the seventeenth century, it is this massive and intriguing existence of language that is eliminated. It no longer appears hidden in the enigma of the mark; it has not yet appeared in the theory of signification. From an extreme point of view, one might say that language in the Classical era does not exist. But that it functions: its whole existence is located in its represen­tative role, is limited precisely to that role and finally exhausts it. Language has no other locus, no other value, than in representation; in the hollow it has been able to form.

In this way. Classical language discovers a certain relation with itself which had hitherto been neither possible nor conceivable. In relation to itself, the language of the sixteenth century was in a position of perpetual commentary; but this commentary can take place only if there is lan­guage - language that silently pre-exists within the discourse by which one tries to make that language speak; there can be no commentary without the absolute precondition of the text; and, inversely, if the world is a network of marks and words, how else is one to speak of them but in the form of commentary? From the Classical age, language is deployed within representation, and in that duplication of itself which hollows itself out. Henceforth, the primary Text is effaced, and with it, the entire, in­exhaustible foundation of the words whose mute being was inscribed in things; all that remains is representation, unfolding in the verbal signs that manifest it, and hence becoming discourse. For the enigma of a speech which a second language must interpret is substituted the essential discursivity of representation: the open possibility, as yet neutral and undifferentiating, but which it will be the task of discourse to fulfil and to determine. When this discourse becomes in turn an object of language, it is not questioned as if it were saying something without actually saying it, as if it were a language enclosed upon itself; one no longer attempts to uncover the great enigmatic statement that lies hidden beneath its

79

THB ORDER OF THINGS

signs; one asks how it functions: what representations it designates, what elements it cuts out and removes, how it analyses and composes, what play of substitutions enables it to accomplish its role of representation. Commentary has yielded to criticism.

This new relation that language establishes with itself is neither a simple nor a unilateral one. Criticism would appear to contrast with commentary in the same way as the analysis of a visible form with the discovery of a hidden content. But since this form is that of representation, criticism can analyse language only in terms of truth, precision, appropriateness, or expressive value. Hence the combined role of criticism and ambiguity -the former never succeeding in freeing itself from the latter. Criticism questions language as if language was a pure function, a totality of mech­anisms, a great autonomous play of signs; but, at the same time, it cannot fail to question it as to its truth or falsehood, its transparency or opacity, and therefore as to exactly how what it says is present in the words by which it represents it. It is on the basis of this double, fundamental necessity that the opposition between content and form gradually emerged and finally assumed the importance we know it to have. But no doubt this opposition was consolidated only at a relatively late date, when, in the nineteenth century, the critical relation had itself been weakened. In the Classical period, criticism was applied, without dissociation and, as it were, en bloc, to the representative role of language. It then assumed four forms, which, though distinct, were interdependent and articulated upon each other. It was deployed first, in the reflexive order, as a critique of words: the impossibility of constructing a science or a philosophy with the received vocabulary; a denunciation in general teims which confused what was distinct in representation with the abstract terms which separ­ated what should remain united; the need to build up the vocabulary of a perfectly analytic language. It was also expressed in the grammatical order as an analysis of the representative values of syntax, word order, and sen­tence construction. Is a language in a higher state of perfection when it has declensions or a system of prepositions? Is it preferable for the word order to be free or strictly determined? What system of tenses best ex­presses relations of sequence? Criticism also examines the forms of rhetoric: the analysis of figures, that is, the types of discourse, with the expressive value of each, the analysis of tropes, that is, the different relations that words may have with the same representative content (designation by a part or the whole, the essential or the accessory, the event or the circum­stance, the thing itself or its analogues). Lastly, faced with existing and

80

SPEAKING

already written language, criticism sets out to define its relation with what it represents; hence the importance assumed, since the seventeenth century, by critical methods in the exegesis of religious texts; it was no longer a question, in fact, of repeating what had already been said in them, but of defining through what figures and images, by following what order, to what expressive ends, and in order to declare what truth, God or the Prophets had given a discourse the particular form in which it was communicated to us.

Such is the diversity of the critical dimension that is necessarily estab­lished when language questions itself on the basis of its function. Since the Classical age, commentary and criticism have been in profound oppo­sition. By speaking of language in terms of representations and truth, criticism judges it and profanes it. Now as language in the irruption of its being, and questioning it as to its secret, commentary halts before the precipice of the original text, and assumes the impossible and endless task of repeating its own birth within itself: it sacralizes language. These two ways by which language establishes a relation with itself were now to enter into a rivalry from which we have not yet emerged - and which may even be sharpening as time passes. This is because since Mallarme, literature, the privileged object of criticism, has drawn closer and closer to the very being of language, and requires therefore a secondary language which is no longer in the form of criticism, but of commentary. And in fact every critical language since the nineteenth century has become im­bued with exegesis, just as the exegeses of the Classical period were im­bued with critical methods. However, until the connection between language and representation is broken, or at least transcended, in our culture, all secondary languages will be imprisoned within the alternative of criticism or commentary. And in their indecision they will proliferate ad infinitum.
II  GENERAL GRAMMAR

Once the existence of language has been eliminated, all that remains is its function in representation: its nature and its virtues as discourse. For discourse is merely representation itself represented by verbal signs. But what, then, is the particularity of these signs, and this strange power that enables them, better than any others, to signalize representation, to analyse it, and to recombine it? What is the peculiar property possessed by language and not by any other system of signs?

81

THE ORDER OF THINGS

At first sight, it is possible to define words according to their arbitrari­ness or their collective character. At its primary root, language is made up, as Hobbes says, of a system of notations that individuals first chose for themselves; by means of these marks they are able to recall representations, link them together, dissociate them, and operate upon them. It is these notations that by covenant or violence were imposed upon the collect­ivity; but the meaning of the words does not pertain, in any case, to any­thing but each individual's representation, and even though it may be accepted by everyone it has no other existence than in the thought of individuals taken separately: 'That then which words are the marks of,' says Locke, 'are the ideas of the speaker: nor can any one apply them as marks, immediately, to anything else but the ideas that he himself hath'[1]. What distinguishes language from all other signs and enables it to play a decisive role in representation is, therefore, not so much that it is individual or collective, natural or arbitrary, but that it analyses represen­tation according to a necessarily successive order: the sounds, in fact, can be articulated only one by one; language cannot represent thought, instantly, in its totality; it is bound to arrange it, part by part, in a linear order. Now, such an order is foreign to representation. It is true that thoughts succeed one another in time, but each one forms a unity, whether one agrees with Condillac[2] that all the elements of a representation are given in an instant and that only reflection is able to unroll them one by one, or whether one agrees with Destutt de Tracy[3] that they succeed one another with a rapidity so great that it is not practically possible to observe or to retain their order. It is these representations, pressed in on one another in this way, that must be sorted out into linear propositions: to my gaze, 'the brightness is within the rose'; in my discourse, I cannot avoid it coming either before or after it[4]. If the mind had the power to express ideas 'as it perceives them', there can be no doubt that 'it would express them all at the same time'[5]. But that is precisely what is not possible, for, though 'thought is a simple operation', 'its expression is a successive operation'[6]. It is here that the peculiar property of language resides, that which distinguishes it both from representation (of which, in its turn, it is nevertheless the representation) and from signs (to which it belongs without any other particular privilege). It does not stand in opposition to thought as the exterior does to the interior, or expression to reflection; it does not stand in opposition to all the other signs - ges­tures, mime, translation, paintings, emblems[7] -as the arbitrary to the natural or the collective to the singular. But it does stand in relation to all

82

SPEAKING

that as the successive to the contemporaneous. It is to thought and to signs what algebra is to geometry: it replaces the simultaneous comparison of parts (or magnitudes) with an order whose degrees must be traversed one after the other. It is in this strict sense that language is an analysis of thought: not a simple patterning, but a profound-establishment of order in space.

It is here that we find that new epistemological domain that the Classical age called 'general grammar'. It would be nonsense to see this purely and simply as the application of a logic to the theory of language. But it would be equally nonsensical to attempt to interpret it as a sort of pro-figuration of a linguistics. General grammar is the study of verbal order in its relation to the simultaneity that it is its task to represent. Its proper object is therefore neither thought nor any individual language, but discourse, understood as a sequence of verbal signs. This sequence is artificial in relation to the simultaneity of representations, and in so far as this is so language must be in opposition to thought, as what is reflected upon is to what is immediate. And yet the sequence is not the same in all languages: some of them place the action in the middle of the sentence; others at the end; some name the principal object of the representation first, others the accessory circumstances; as the Encyclopedie points out, what renders foreign languages opaque to one another, and so difficult to translate, is not so much the differences between the words as the incompatibility of their sequences [8]. In relation to the evident, necessary, universal order introduced into representation by science, and by algebra in par­ticular, language is spontaneous and un-thought-out; it is, as it were, natural. It is equally, according to the point of view from which one looks at it, an already analysed representation and a reflection in the primitive state. In fact, it is the concrete link between representation and reflection. It is not so much the instrument of men's intercommunication as the path by which, necessarily, representation communicates with reflection. This is why general grammar assumed so much importance for philosophy during the eighteenth century: it was, at one and the same time, the spontaneous form of science - a kind of logic not controlled by the mind [9] - and the first reflective decomposition of thought: one of the most primitive breaks with the immediate. It constituted, as it were, a philosophy in­herent in the mind - metaphysics, Adam Smith pointed out, was an essential ingredient in the formation of even the least of adjectives [10] -and one that any philosophy had to work through if it was to rediscover, among so many diverse choices, the necessary and evident order of

83

THE ORDER OF THINGS

representation. Language is the original form of all reflection, the primary theme of any critique. It is this ambiguous thing, as broad as knowledge, yet always interior to representation, that general grammar takes as its object. But a certain number of consequences must at once be drawn here.

1. The first is that it is easy to see how the sciences of language are divided up in the Classical period: on the one hand, rhetoric, which deals with figures and tropes, that is, with the manner in which language is spatialized in verbal signs; on the other, grammar, which deals with articulation and order, that is, with the manner in which the analysis of representation is arranged in accordance with a sequential series. Rhetoric defines the spatiality of representation as it comes into being with language; grammar defines in the case of each individual language the order that distributes that spatiality in time. This is why, as we shall see, grammar presupposes languages, even the most primitive and spontaneous ones, to be rhetorical in nature.

2. On the other hand, grammar, as reflection upon language in general, expresses the relation maintained by the latter with universality. This re­lation can take two forms, according to whether one takes into considera­tion the possibility of a universal language or that of a universal discourse. In the Classical period, what was denoted by the term universal language was not the primitive, pure, and unimpaired speech that would be able, if it were rediscovered beyond the punishment of oblivion, to restore the understanding that reigned before Babel. It refers to a tongue that would have the ability to provide every representation, and every element of every representation, with the sign by which it could be marked in a univocal manner; it would also be capable of indicating in what manner the elements in a representation are composed and how they are linked to one another; and since it would possess the necessary instruments with which to indicate all the possible relationships between the various seg­ments of representation, this language would also, by that very fact, be able to accommodate itself to all possible orders. At once characteristic and combinative, the universal language does not re-establish the order of days gone by: it invents signs, a syntax, and a grammar, in which all conceivable order must find its place. As for universal discourse, that too is by no means the unique text that preserves in the cipher of its secret the key to unlock all knowledge; it is rather the possibility of defining the natural and necessary progress of the mind from the simplest repre­sentations to the most refined analyses or the most complex combina­tions: this discourse is knowledge arranged in accordance with the unique

84

SPEAKING

order laid down for it by its origin. It traverses the whole field of know­ledge, though as it were in a subterranean manner, in order to reveal, on the basis of representation, the possibility of that knowledge, to reveal its origin, and its natural, linear, and universal link. This common de­nominator, this foundation underlying all knowledge, this origin expressed in a continuous discourse is Ideology, a language that duplicates the spon­taneous thread of knowledge along the whole of its length:

Man, by his nature, always tends towards the nearest and most pressing result. He thinks first of his needs, then of his pleasures. He occupies himself with agriculture, with medicine, with war, with practical politics, then with poetry and the arts, before turning his thoughts to philosophy; and when he turns back upon himself and begins to reflect, he prescribes rules for his judgement, which is logic, for his discourse, which is grammar, for his desires, which is ethics. He then believes himself to have reached the summit of theory...;

but he perceives that all these operations have 'a common source' and that 'this sole centre of all truths is the knowledge of his intellectual faculties'[11].

The universal characteristic and ideology stand in the same opposition to one another as do the universality of language in general (which arranges all possible orders in the simultaneity of a single fundamental table) and the universality of an exhaustive discourse (which reconstitutes the single genesis, common to the whole sequence of all possible branches of knowledge). But their aim and their common possibility reside in a power that the Classical age attributes to language: that of providing adequate signs for all representations, whatever they may be, and of establishing possible links between them. In so far as language can repre­sent all representations it is with good reason the element of the universal. There must exist within it at least the possibility of a language that will gather into itself, between its words, the totality of the world, and, inversely, the world, as the totality of what is representable, must be able to become, in its totality, an Encyclopaedia. And Charles Bonnet's great dream merges at this point with what language is in its connection and kinship with representation:

I delight in envisaging the innumerable multitude of Worlds as so many books which, when collected together, compose the immense Library of the Universe or the true Universal Encyclopaedia. I con­ceive that the marvellous gradation that exists between these different

85

THE ORDER OF THINGS

worlds facilitates in superior intelligences, to whom it has been given to traverse or rather to read them, the acquisition of truths of every kind, which it encompasses, and instils in their understanding that order and that concatenation which are its principal beauty. But these celestial Encyclopaedists do not all possess the Encyclopaedia of the Universe to the same degree; some possess only a few branches of it, others possess a greater number, others grasp even more still; but all. have eternity in which to increase and perfect their learning and develop all their faculties [12].

Against this background of an absolute Encyclopaedia, human beings constitute intermediary forms of a composite and limited universality: alphabetical encyclopaedias, which accommodate the greatest possible quantity of learning in the arbitrary order provided by letters; pasigraphies, which make it possible to transcribe all the languages of the world by means of a single system of figures [13]; polyvalent lexicons, which establish synonymies between a greater or lesser number of lan­guages; and, finally, rational encyclopaedias, which claim to 'exhibit as far as is possible the order and concatenation of human learning' by examining 'their genealogy and their filiation, the causes that must have given rise to them and the characteristics that distinguish them'[14]. Whatever the partial character of these projects, whatever the empirical circumstances of such undertakings, the foundation of their possibility in the Classical episteme is that, though language had been entirely reduced to its function within representation, representation, on the other hand, had no relation with the universal except through the intermediary of language.

3. Knowledge and language are rigorously interwoven. They share, in representation, the same origin and the same functional principle; they support one another, complement one another, and criticize one another incessantly. In their most general form, both knowing and speaking consist first of all in the simultaneous analysis of representation, in the discrimina­tion of its elements, in the establishing of the relations that combine those elements, and the possible sequences according to which they can be un­folded. It is in one and the same movement that the mind speaks and knows: 'It is by the same processes that one learns to speak and that one discovers either the principles of the world's system or those of the human mind's operations, that is, all that is sublime in our knowledge'[15]. But language is knowledge only in an unreflecting form; it imposes itself on individuals from the outside, guiding them, willy-nilly, towards notions

86

SPEAKING

that may be concrete or abstract, exact or with little foundation. Know­ledge, on the other hand, is like a language whose every word has been examined and every relation verified. To know is to speak correctly, and as the steady progress of the mind dictates; to speak is to know as far as one is able, and in accordance with the model imposed by those whose birth one shares. The sciences are well-made languages, just as languages are sciences lying fallow. All languages must therefore be renewed; in other words, explained and judged according to that analytic order which none of them now follows exactly; and readjusted if necessary so that the chain of knowledge may be made visible in all its clarity, without any shadows or lacunae. It is thus part of the very nature of grammar to be prescriptive, not by any means because it is an attempt to impose the norms of a beautiful language obedient to the rules of taste, but because it refers the radical possibility of speech to the ordering system of repre­sentation. Destutt de Tracy once observed that the best treatises on logic, in the eighteenth century, were written by grammarians: this is because the prescriptions of grammar at that time were of an analytic and not an aesthetic order.

And this link between language and knowledge opens up a whole historical field that had not existed in previous periods. Something like a history of knowledge becomes possible; because, if language is a spon­taneous science, obscure to itself and unpractised, this also means, in return, that it will be brought nearer to perfection by knowledge, which cannot lodge itself in the words it needs without leaving its imprint in them, and, as it were, the empty mould of its content. Languages, though imperfect knowledge themselves, are the faithful memory of the progress of knowledge towards perfection. They lead into error, but they record what has been learned. In their chaotic order, they give rise to false ideas; but true ideas leave in them the indelible mark of an order that chance on its own could never have created. "What civilizations and peoples leave us as the monuments of their thought is not so much their texts as their vocabularies, their syntaxes, the sounds of their languages rather than the words they spoke; not so much their discourse as the element that made it possible, the discursivity of their language.

The language of a people gives us its vocabulary, and its vocabulary is a sufficiently faithful and authoritative record of all the knowledge of that people; simply by comparing the different states of a nation's vocabulary at different times one could form an idea of its progress.

87

THE ORDER OF THINGS

Every science has its name, every notion within a science has its name too, everything known in nature is designated, as is everything in­vented in the arts, as well as phenomena, manual tasks, and tools[16].

Hence the possibility of writing a history of freedom and slavery based upon languages [17], or even a history of opinions, prejudices, super­stitions, and beliefs of all kinds, since what is written on these subjects is always of less value as evidence than are the words themselves [18]. Hence, too, the project of creating an encyclopaedia 'of the sciences and arts', which would not follow the connecting links of knowledge itself but would be accommodated in the form of the language, within the space opened up in words themselves; for that is where future ages would have to look to find what we have known or thought, since words, in their roughly hewn state, are distributed along that mid-way line that marks the adjacency of science to perception and of reflection to images. It is in them that what we imagine becomes what we know, and, on the other hand, that what we know becomes what we represent to ourselves every day. The old relation to the text, which was the Renaissance defini­tion of erudition, has now been transformed: it has become, in the Classical age, the relation to the pure element of the language.

Thus we see glowing into life the luminous clement in which language and learning, correct discourse and knowledge, universal language and analysis of thought, the history of mankind and the sciences of language freely communicate. Even when it was intended for publication, the knowledge of the Renaissance was arranged within an enclosed space. The 'Academy' was a closed circle which projected the essentially secret form of knowledge onto the surface of social configurations. For the primary task of that knowledge was to draw speech from mute signs: it had to recognize their forms, interpret them, and retranscribe them by means of other graphic signs which then had to be deciphered in their turn; so that even the discovery of the secret did not escape this array of obstacles, which had rendered it at once so difficult and yet so precious. In the Classical age, knowing and speaking are interwoven in the same fabric; in the case of both knowledge and language, it is a question of providing representation with the signs by means of which it can unfold itself in obedience to a necessary and visible order. Even when stated, knowledge in the sixteenth century was still a secret, albeit a shared one. Even when hidden, knowledge in the seventeenth and eighteenth centuries is discourse with a veil drawn over it. This is because it is of the very

88

SPEAKING

nature of science to enter into the system of verbal communications[19], and of the very nature of language to be knowledge from its very first word. Speaking, enlightening, and knowing are, in the strict sense of the term, of the same order. The interest shown by the Classical age in science, the publicity accorded to its controversies, its extremely exoteric character, its opening up to the uninitiated, Fontenelle's popularization of astronomy, Voltaire reading Newton, all this is doubtless nothing more than a sociological phenomenon. It did not provoke the slightest alteration in the history of thought, or modify the development of know­ledge one jot. It explains nothing, except of course on the doxographic level where it should be situated; but its condition of possibility is never­theless there, in that reciprocal kinship between knowledge and language. The nineteenth century was to dissolve that link, and to leave behind it, in confrontation, a knowledge closed in upon itself and a pure language that had become, in nature and function, enigmatic - something that has been called, since that time, Literature. Between the two, the intermediary languages - descendants of, or outcasts from, both knowledge and lan­guage - were to proliferate to infinity.

4. Because it had become analysis and order, language entered into relations with time unprecedented hitherto. The sixteenth century ac­cepted that languages succeeded one another in history and were capable of engendering one another. The oldest were the mother languages. The most archaic of all, since it was the tongue of the Eternal when he ad­dressed himself to men, was Hebrew, and Hebrew was thought to have given rise to Syriac and Arabic; then came Greek, from which both Coptic and Egyptian were derived; Latin was the common ancestor of Italian, Spanish, and French; lastly, 'Teutonic' had given rise to German, English, and Flemish [20]. In the seventeenth century, the relation of language to time is inverted: it is no longer time that allots languages their places, one by one, in world history; it is languages that unfold representations and words in a sequence of which they themselves define the laws. It is by means of this internal order, and the positions it allots to its words, that each language defines its specificity, and no longer by means of its place in a historical series. For language, time is its interior mode of ana­lysis, not its place of birth. Hence the paucity of interest shown by the Classical age in chronological filiation, to the point of denying, contrary to all the 'evidence' - our evidence, that is - the kinship of Italian or French with Latin[21]. The kinds of series that existed in the sixteenth century, and were to reappear in the nineteenth, were replaced by

89

THE ORDER OF THINGS

typologies, typologies of order. There is the group of languages that places the subject being dealt with first; next the action undertaken or under­gone by that subject; and last the object upon which it is exercised: as witness, French, English, Spanish. Opposed to these is the group of languages that places 'sometimes the action, sometimes the object, some­times the modification or circumstance first': for example Latin, or 'Slavonian', in which the function of words is indicated, not by their positions, but by their inflections. Finally, there is the third group made up of mixed languages (such as Greek or Teutonic), 'which have something of both the other groups, possessing an article as well as cases'[22]. But it must be understood that it is not the presence or absence of inflections that defines the possible or necessary order of the words in each language. It is order as analysis and a sequential alignment of representations that constitutes the preliminary form and prescribes the use of declensions or articles. Those languages that follow the order 'of imagination and interest' do not determine any constant position for words: they are obliged to emphasize them by means of inflections (these are the 'transpositive' languages). If, on the other hand, they follow the uniform order of re­flection, they need only indicate the number and gender of substantives by means of an article; position in the analytic ordering of the sentence has a functional value in itself: these are the 'analogical' languages [23]. Languages are related to and distinguished from one another according to a table of possible types of word order. The table shows them all simultaneously, but suggests which were the most ancient languages; it may be admitted, in fact, that the most spontaneous order (that of images and passions) must have preceded the most considered (that of logic); external dating is determined by the internal forms of analysis and order. Time has become interior to language.

The history of the various languages is no longer anything more than a question of erosion or accident, introduction, meetings, and the mingling of various elements; it has no law, no progress, no necessity proper to it. How, for instance, was the Greek language formed?

It was Phoenician merchants, adventurers from Phrygia, from Mace­donia and Illyria, Galatians, Scythians, and bands of exiles or fugitives who loaded the first stratum of the Greek language with so many kinds of innumerable particles and so many dialects [24].

French is made up of Latin and Gothic nouns, Gallic constructions, Arabic articles and numerals, words borrowed from the English and the

90

SPEAKING

Italians - as journeys, wars, or trade agreements dictated [25]. This is be­cause languages evolve in accordance with the effects of migrations, victories and defeats, fashions, and commerce; but not under the impul­sion of any historicity possessed by the languages themselves. They do not obey any internal principle of development; they simply unfold repre­sentations and their elements in a linear sequence. If there does exist a time for languages that is positive, then it must not be looked for outside them, in the sphere of history, but in the ordering of their words, in the form left by discourse.

It is now possible to circumscribe the epistemological field of general grammar, which appeared during the second half of the seventeenth century and faded away again during the last years of the following century. General grammar is not at all the same as comparative grammar: the comparisons it makes between different languages are not its object; they are merely employed as a method. This is because its generality does not consist in the discovery of peculiarly grammatical laws, common to all linguistic domains, which could then be used to display the structure of any possible language in an ideal and constricting unity; if it is indeed general, then it is so to the extent that it attempts to make visible, below the level of grammatical rules, but at the same level as their foundation, the representative function of discourse - whether it be the vertical function, which designates what is represented, or the horizontal function, which links what is represented to the same mode as thought. Since it makes language visible as a representation that is the articulation of another representation, it is indisputably 'general'; what it treats of is the interior duplication existing within representation. But since that articulation can be accomplished in many different ways, there must be, paradoxically, various general grammars: French, English, Latin, German, etc. [26]. General grammar does not attempt to define the laws of all languages, but to examine each particular language, in turn, as a mode of the articulation of thought upon itself. In every language, taken in isola­tion, representation provides itself with 'characters'. General grammar is intended to define the system of identities and differences that these spontaneous characters presuppose and employ. It must establish the taxonomy of each language. In other words, the basis, in each of them, for the possibility of discourse.

Hence the two directions that it necessarily takes. Since discourse links its parts together in the same way as representation does its elements, general grammar must study the representative function of words in

91

THE ORDER OF THINGS

relation to each other; which presupposes in the first place an analysis of the links that connect words together (theory of the proposition and b particular of the verb), then an analysis of the various types of words and of the way in which they pattern the representation and are distin­guished from each other (theory of articulation). However, since discourse is not simply a representative whole, but a duplicated representation that denotes another representation - the one that it is in fact representing -general grammar must also study the way in which words designate what they say, first of all in their primitive value (theory of origins and of the root), then in their permanent capacity for displacement, extension, and reorganization (theory of rhetoric and of derivation).
III THE THEORY OF THE VERB

The proposition is to language what representation is to thought, at once its most general and most elementary form, since as soon as it is broken down we no longer encounter the discourse but only its elements, in the form of so much scattered raw material. Below the proposition we do indeed find words, but it is not in them that language is created. It is true that in the beginning man emitted only simple cries, but these did not begin to be language until they contained - if only within their mono­syllable - a relation that was of the order of a proposition. The yell of the primitive man in a struggle becomes a true word only when it is no longer the lateral expression of his pain, and when it has validity as a judgement or as a statement of the type 'I am choking'[27]. What con­stitutes a word as a word and raises it above the level of cries and noises is the proposition concealed within it. If the wild man of Aveyron did not attain to speech, it was because words remained for him merely the vocal marks of things and of the impressions that those things made upon his mind; they had acquired no propositional value. He could, it is true, pronounce the word 'milk' when a bowl of milk was put in front of him; but that was merely 'the confused expression of that alimentary liquid, of the vessel containing it, and of the desire produced by it' [28]; the word never became a sign representing the thing, for at no point did he ever wish to say that the milk was hot, or ready, or expected. It is in fact the proposition that detaches the vocal sign from its immediate expressive values and establishes its supreme linguistic possibility. For Classical thought, language begins not with expression, but with discourse. When one says 'no', one is not translating one's refusal into a mere cry; one is

92

SPEAKING

contracting into the form of a single word 'an entire proposition: ... I do not feel that, or I do not believe that'[29].

'Let us go directly to the proposition, the essential object of gram­mar'[30]. In the proposition, all the functions of language are led back to the three elements that alone are indispensable to the formation of a proposition: the subject, the predicate, and the link between them. Even then, the subject and predicate arc of the same nature, since the pro­position affirms that the one is identical to or akin to the other; it is there­fore possible for them, under certain conditions, to exchange functions. The only difference, though it is a decisive one, is that manifested by the irreducibility of the verb: as Hobbes[31] says:

In every proposition three things are to be considered, viz. the two names, which are the subject and the predicate, and their copulation; both which names raise in our mind the thought of one and the same thing; but the copulation makes us think of the cause for which those names were imposed on that thing.

The verb is the indispensable condition for all discourse; and wherever it does not exist, at least by implication, it is not possible to say that there is language. All nominal propositions conceal the invisible presence of a verb, and Adam Smithy] thinks that, in its primitive form, language was com­posed only of impersonal verbs (such as 'it is raining' or 'it is thundering'), and that all the other parts of discourse became detached from this original verbal core as so many derived and secondary details. The threshold of language lies at the point where the verb first appears. This verb must therefore be treated as a composite entity, at the same time a word among other words, subjected to the same rules of case and agreement as other words, and yet set apart from all other words, in a region which is not that of the spoken, but rather that from which one speaks. It is on the fringe of discourse, at the connection between what is said and what is saying itself, exactly at that point where signs are in the process of becoming language.

It is this function that we must now examine - by stripping the verb of all that has constantly overlaid and obscured it. We must not stop, as Aristotle did, at the fact that the verb signifies tenses (there are many other words, adverbs, adjectives, nouns, that can carry temporal sig­nifications). Nor must we stop, as Scaliger did, at the fact that it expresses actions or passions, whereas nouns denote things - and permanent things (for there is precisely the very noun 'action' to be considered). Nor must we attach importance, as Buxtorf did, to the different persons of the verb,

93

THB ORDER OF THINGS

for these can also be designated by certain pronouns. What we must do before all else is to reveal, in all clarity, the essential function of the verb: the verb affirms, it indicates 'that the discourse in which this word is employed is the discourse of a man who does not merely conceive of nouns, but judges them'[33]. A proposition exists-and discourse too-when we affirm the existence of an attributive link between two things, when we say that this is that [34]. The entire species of the verb may be reduced to the single verb that signifies to be. All the others secretly make use of this unique function, but they have hidden it beneath a layer of determinations: attributes have been added to it, and instead of saying 'I am singing', we say "I sing'[35]; indications of time have been added, and instead of saying 'before now I am singing', we say 'I sang'; lastly, certain languages have integrated the subject itself into their verbs, and thus we find the Romans saying, not ego vivit, but vivo. All of this is merely accretion and sedimentation around and over a very slight yet essential verbal function, 'there is only the verb to be . . . that has re­mained in this state of simplicity'[36]. The entire essence of language is concentrated in that singular word. Without it, everything would have remained silent, and though men, like certain animals, would have been able to make use of their voices well enough, yet not one of those cries hurled through the jungle would ever have proved to be the first link in the great chain of language.

In the Classical period, language in its raw state - that mass of signs impressed upon the world in order to exercise our powers of inter­rogation - vanished from sight, but language itself entered into new rela­tions with being, ones more difficult to grasp, since it is by means of a word that language expresses being and is united to it; it affirms being from within itself; and yet it could not exist as language if that word, on its own, were not, in advance, sustaining all possibility of discourse. With­out a way of designating being, there would be no language at all; but without language, there would be no verb to be, which is only one part of language. This simple word is the representation of being in language; but it is equally the representative being of language - that which, by enabling language to affirm what it says, renders it susceptible of truth or error. In this respect it is different from all the signs that may or may not be consistent with, faithful to, or well adapted to, what they designate, but that are never true or false. Language is, wholly and entirely, dis­course; and it is so by virtue of this singular power of a word to leap across the system of signs towards the being of that which is signified.

94

SPEAKING

But from where does this power derive? And what is this meaning, which, by overflowing the words containing it, forms the basis of the proposition? The grammarians of Port-Royal said that the meaning of the verb to be was affirmation - which indicated well enough in what region of language its absolute privilege lay, but not at all in what it consisted. We must not imagine that the verb to be contains the idea of affirmation, for the word affirmation itself, and also the word yes, contain it equally well [37]; what the verb to be provides is rather the affirmation of the idea. But is the affirmation of an idea also the expression of its existence? This is in fact what Bauzee thinks, and he also takes it to be one reason why variations of time have been concentrated into the form of the verb: for the essence of things does not change, it is only their existence that appears and disappears, it is only their existence that has a past and a future[38]. To which Condillac can observe in reply that if existence can be withdrawn from things, this must mean that it is no more than an attribute, and that the verb can affirm death as well as existence. The only thing that the verb affirms is the coexistence of two representations: for example, those of a tree and greenness, or of man and existence or death; this is why the tenses of verbs do not indicate the time when things existed in the absolute, but a relative system of anteriority or simultaneity between different things [39]. Coexistence is not, in fact, an attribute of the thing itself; it is no more than a form of the representation: to say that the greenness and the tree coexist is to say that they are linked together in all, or most of, the impressions I receive.

So that the essential function of the verb to be is to relate all language to the representation that it designates. The being towards which it spills over its signs is neither more nor less than the being of thought. Com­paring language to a picture, one late-eighteenth-century grammarian defines nouns as forms, adjectives as colours, and the verb as the canvas itself, upon which the colours are visible. An invisible canvas, entirely overlaid by the brightness and design of the words, but one that provides language with the site on which to display its painting. What the verb designates, then, is the representative character of language, the fact that it has its place in thought, and that the only word capable of crossing the frontier of signs and providing them with a foundation in truth never attains to anything other than representation itself. So that the function of the verb is found to be identified with the mode of existence of language, which it traverses throughout its length: to speak is at the same time to represent by means of signs and to give signs a synthetic form governed

95

THE ORDER OF THINGS

by the verb. As Destutt says, the verb is attribution, the sustaining power, and the form of all attributes:

The verb to he is found in all propositions, because we cannot say that a thing is in such and such a way without at the same time saying that it is... But this word is which is in all propositions is always a part of the attribute [predicate] in those propositions, it is always the beginning and the basis of the attribute, it is the general and common attribute [40].

It will be seen how the function of the verb, once it had reached this point of generality, had no other course but to become dissociated, as soon as the unitary domain of general grammar itself disappeared. When the dimension of the purely grammatical was opened up, the proposition was to become no more than a syntactical unit. The verb was merely to figure in it along with all the other words, with its own system of agreement, inflections, and cases. And at the other extreme, the power of mani­festation of language was to reappear in an autonomous question, more archaic than grammar. And throughout the nineteenth century, language was to be examined in its enigmatic nature as verb: in that region where it is nearest to being, most capable of naming it, of transmitting or giving effulgence to its fundamental meaning, of rendering it absolutely mani­fest. From Hegel to Mallarme, this astonishment in the face of the relations of being and language was to counterbalance the reintroduction of the verb into the homogeneous order of grammatical functions.
IV ARTICULATION

The verb to be, a mixture of attribution and affirmation, the junction of discourse with the primary and radical possibility of speech, defines the first constant of the proposition, and also the most fundamental. Beside it, on either side, are elements: parts of discourse or 'oration'. These sites are still neutral, and determined solely by the slender, almost imperceptible, yet central figure designating being; they function, on either side of this mdicator' as the thing to be judged - the judicandum - and the thing judged- the judicatum[41]. How can this pure design of the proposition be transformed into distinct sentences? How can discourse express the whole content of a representation?

Because it is made up of words that name, part by part, what is given to representation.

96

SPEAKING

The word designates, that is, in its very nature it is a noun or name. A proper noun, since it is directed always towards a particular representation, and towards no other. So, in contrast to the uniformity of the verb, which is never more than the universal expression of attribution, nouns proliferate in endless differentiation. There ought to be as many of them as there are things to name. But each name would then be so strongly attached to the single representation it designated that one could never formulate even the slightest attribution; and language would fall back to a lower level:

If we had no other substantives but proper nouns, it would be necessary to create an infinite multiplicity of them. These words, whose great number would overburden our memories, would produce no order in the objects of our learning, nor, consequently, in our ideas, and all our discourse would be in the greatest state of confusion [42].

Nouns cannot function in a sentence and permit attribution unless one of the two (the attribute at least) designates some element common to several representations. The generality of the noun is as necessary to the parts of discourse as is the designation of being to the form of the pro­position.

This generality may be acquired in two ways. Either by a horizontal articulation, grouping together individuals that have certain identities in common and separating those that are different; such an articulation then forms a sequential generalization of groups growing gradually larger and larger (and less and less numerous); it may also subdivide them almost to infinity by means of fresh distinctions, and thus return to the proper noun from which it began [43]; the entire order of the resulting coordinations and subordinations is covered by a grid of language, and each one of these points will be found upon it together with its name: from the individual to the species, then from the species to the genus and on to the class, language is articulated precisely upon the dimension of increasing generali­ties; this taxonomic function is manifested in language by the substantives: we say an animal, a quadruped, a dog, a spaniel [44]. Or else by a vertical articulation, linked to the first, for each is indispensable to the other; this second articulation distinguishes the things that subsist by themselves from those - modifications, features, accidents, or characteristics - that one can never meet in an independent state: deep down, substances; on the surface, qualities; this division - this metaphysic, as Adam Smith called it-is manifested in discourse by the presence of adjectives, which designate

97

THE ORDER OF THINGS

everything in representation that cannot subsist by itself. The primary articulation of language (if we leave aside the verb to be, which is as much a condition of discourse as it is a part of it) is thus aligned along two orthogonal axes: one proceeding from the individual unit to the general; the other proceeding from the substance to the quality. At their point of intersection stands the common noun; at one extremity the proper noun, at the other the adjective.

But these two types of representation can distinguish words from one another only to precisely that degree to which representation is analysed according to this same model. As the authors of Port-Royal put it: words 'that signify things are called substantival nouns, such as earth, sun. Those that signify manners, while at the same time indicating the subject with which the manners agree, are called adjectival nouns, such as good, just, round'[45]. However, there does exist a certain amount of play between the articulation of language and that of representation. When we speak of 'whiteness', we are certainly designating a quality, but we are desig­nating it by means of a substantive; when we speak of 'humans' we are employing an adjective to designate individuals that subsist by themselves. This displacement is not an indication that language obeys other laws than those of representation, but, on the contrary, that it has relations, with itself and in its own density, that are identical with those of repre­sentation. For is it not, in fact, a duplicated kind of representation, and thus able to combine with the elements of its representation another representation distinct from the first, even though the only function and meaning of the second representation is the representation of the first? If discourse seizes upon the adjective designating a modification and gives it within the sentence the value of the very substance of the proposition, then that adjective becomes substantival; the noun, on the other hand, which behaves within the sentence like an accident, becomes adjectival, even though it is designating substances, as hitherto.

Because substance is that which subsists of itself, the term substantive has been given to all those words that subsist by themselves in dis­course, even though they may signify accidents. And, on the other hand, the term adjective has been given to those words that signify substances when, in their manner of signifying, they must be joined in discourse to other nouns [46].

The relations between the elements of the proposition are identical with those of representation; but this identity is not carefully arranged point

98

SPEAKING

by point, so that every substance is designated by a substantive and every accident by an adjective. The identity here is total and a matter of nature:

the proposition is a representation; it is articulated according to the same modes as representation; but it possesses the power to articulate the repre­sentation it transforms into discourse in more than one way. It is, in itself, a representation providing the articulation for another, with a possi­bility of displacement that constitutes at the same time the freedom of discourse and the differences between languages.

Such is the first stratum of articulation - the most superficial or in any case the most apparent. Once this has been established, everything can become discourse; but in the form of a still rather undifferentiated lan­guage: we still have nothing but the monotony of the verb to be and its attributive function to link our nouns together. Now, the elements of representation are articulated according to a whole network of complex relations (succession, subordination, consequence) that must be brought over into language if it is to become truly representative. Hence all the words, syllables, even letters, which, circulating among the nouns and the verbs, are given the task of designating those ideas that in Port-Royal were termed 'accessory'[47]; there must be prepositions and conjunctions; there must be syntactical signs indicating the relations of identity or agree­ment, and those of dependence or case[48]: marks of plurality and gender, declension endings; and, finally, there must be words relating common nouns to the individuals they designate - the articles or demons­tratives that Lemercier called 'concretizers' or 'disabstractors'[49]. Such a scattering of words constitutes an articulation inferior to the unity of the name (whether substantival or adjectival) as required by the naked form of the proposition: none of them possesses in its own right, and in an isolated state, a fixed and determinate representative content; they cannot cover an idea - even an accessory one - until they have been linked together with other words; whereas nouns and verbs are 'absolute significants', these words, on the other hand, have no power of signification except in a relative mode. It is true that they are addressed to repre­sentation; they exist only in so far as the latter, in the process of analysing itself, makes the interior network of these relationships visible; but they themselves have value only through the grammatical whole of which they are a part. They establish a new articulation in language, one of a com­posite nature, at once representative and grammatical, though without either of these two orders being able to fit exactly over the other.

At this stage, then, the sentence is peopled with syntactical elements cut

99

THB ORDER OF THINGS

out according to much more delicate patterns than the broad figures of the proposition. This new and more complicated patterning presents general grammar with a necessary choice: either to pursue its analysis at a lower level than nominal unity, and to bring into prominence, before signification, the insignificant elements of which it is constructed, or to reduce that nominal unity by means of a regressive process, to recognize its existence within more restricted units, and to find its efficacity as representation below the level of whole words, in particles, in syllables, and even in single letters themselves. These possibilities are presented -indeed, they are prescribed - as soon as the theory of languages takes as its object discourse and the analysis of its representative values. They define the point of heresy that splits all eighteenth-century grammar.

Shall we suppose, Harris asks, that all signification is, like the body, divisible into an infinity of other significations, themselves divisible to infinity? That would be an absurdity; we must therefore necessarily admit that there are significant sounds of which no part can possess signification of itself [50]. Signification disappears as soon as the representative values of words are dissociated or suspended: instead, there appear, in their inde­pendence, raw materials that are not articulated upon thought and whose links cannot be reduced to those of discourse. There is a 'mechanics' proper to agreements, to cases, to inflections, to syllables, and to sounds, and no representative value can provide us with an account of that mechanics. Language must be treated like a mechanical construction sus­ceptible of gradual improvement [51]: in its simplest form, the sentence is composed only of a subject, a verb, and a predicate; and every addition of meaning requires a fresh and entire proposition; in the same way, the most rudimentary machines presuppose principles of movement that differ for each of their organs. But as they are perfected, so they sub­ordinate all their organs to one and the same principle, of which the organs are then only the intermediaries, the means of transformation, the points of application; similarly, as languages perfect themselves, they transmit the sense of a proposition by means of grammatical organs that do not in themselves possess any representative value, but perform the tasks of making it more specific, of linking its elements together, of in­dicating its actual determinations. In a single continuous sentence it is possible to indicate relations of time, of consequence, of possession, and of localization, all of which certainly enter into the subject-verb-predicate series, but cannot be pinned down by so broad a distinction. Hence the importance accorded since Bauzee[52] to the theories of the complement,

100

SPEAKING

of subordination. Hence, too, the growing role of syntax; at the time of Port-Royal, syntax was identified with the construction and ordering of words, and thus with the interior development of the proposition [5 3]; with Sicard it became independent: it is syntax 'that determines the proper form of each word'[54]. These were the preliminary sketches for the grammatical autonomy to be defined later, at the very end of the century, by Sylvestre de Saci, when he became the first - together with Sicard - to distinguish between the logical analysis of the proposition and the gram­matical analysis of the sentence [55].

It is understandable why analyses of this kind should have remained in suspense as long as discourse remained the object of grammar; as soon as a stratum of articulation was reached where representative values crumbled away, there was a movement from the other side of grammar, where grammar no longer had any power, into the domain of usage and history - syntax, in the eighteenth century, was thought of as the locus of the arbitrary in which the habits of each people were deployed according to whim [56].

In any case, such analyses could not, in the eighteenth century, be any­thing more than abstract possibilities; not prefigurations of what was to be philology, but the non-privileged branch of a choice. Opposite, and with the same point of heresy as its starting-point, we see developing a reflec­tion, which, for us and the science of language we have constructed since the nineteenth century, is void of all value, but which at that time enabled all analysis of verbal signs to be retained within discourse itself. And which, by means of this exact overlaying, came to be included in the positive figures of knowledge. There was a search for the obscure nominal function that was thought to be invested and concealed in those words, in those syllables, in those inflections, in those letters that the over-generalized analysis of the proposition was allowing to pass through its net. Because, after all, as the authors of Port-Royal pointed out, all con­nective particles must have a certain content, since they represent the manner in which objects are linked together, and in which they are con­nected in our representations[57]. May one not suppose that they have been names like all the others? But that instead of substituting themselves for objects they have taken the place of those gestures by which men indicated them or simulated their connections and their succession [58]? It is these words that have either gradually lost their own particular meaning (which was not always visible, in any case, since it was linked to the ges­tures, the body, and the situation of the speaker) or incorporated themselves

101

THE ORDER OF THINGS

into other words, in which they found a stable support, and to which they gave in return a whole system of modifications [59]. So that all words, of whatever kind, arc dormant names: verbs have joined adjectival names to the verb to be; conjunctions and prepositions are the names of gestures now frozen into immobility; declensions and conjuga­tions are no more than names that have been absorbed. Words, now, can open up and restore their freedom of flight to all the names that have been lodged within them. As Le Bel said, stating it as a fundamental principle of analysis, 'there is no group of which the parts have not existed separately before being grouped together'[60]; this enabled him to reduce all words to syllabic elements in which the old forgotten names at last made their reappearance - the only vocables that possessed the possibility of existing side by side with the verb to be: Romulus, for example[61], comes from Roma and moliri (to build); and Roma comes from ro, which denoted strength (robur) and ma, which denoted magni­tude (magnus). In the same way, Thiebault discovers three latent sig­nifications in abandonner: a, which 'presents the idea of the tendency or destination of one thing towards another'; ban, which 'gives the idea of the totality of the social body', and do, which indicates 'the act whereby one relinquishes something'[62].

And if one is forced to descend below the level of individual syllables to the very letters of the words, one can still find the values of a rudi­mentary form of nomination. A task to which, to his greater - though even more perishable - glory. Court de Gebelin really applied himself: 'the labial contact, the easiest to bring into play, the gentlest, the most gracious, served to designate the first beings man comes to know, those who surround him and to whom he owes everything' (papa, mama). On the other hand, 'the teeth are as firm as the lips are mobile and flexible; the intonations that proceed from them are strong, sonorous, noisy ...' It is by means of dental contact that one expresses the ideas that lie behind such verbs as tonner (to thunder), retentir (to resound), etonner (to astonish); it is by this means too that one denotes tambours (drums), timbales (tim­pani), and trompettes (trumpets). Vowels, too, in isolation, are able to unfold the secret of the age-old names that usage has buried within them: A for possession (avoir, to have), E for existence, I for puissance (power), O for etonnement (astonishment, eyes opened wide), U for humidite (humidity) and therefore for humeur (mood) [63]. And perhaps, in the very oldest stratum of our history, consonants and vowels, differentiated only as two still vague groups, formed as it were the two sole names upon

102

SPEAKING

which human speech is ultimately articulated: the singing vowels speaking our passions; the rough consonants our needs [64]. It is still possible to distinguish the rocky tongues of the North - a forest of gutturals, of hunger and cold - from the Southern tongues that are all vowels, born of early morning encounters between shepherds when 'the first fires of love were bursting from the pure crystal of the springs'.

Throughout its density, even down to the most archaic of those sounds that first rescued it from its state as pure cry, language preserves its repre­sentative function; in each one of its articulations, from the depths of time, it has always named. It is nothing in itself but an immense rustling of denominations that are overlying one another, contracting into one another, hiding one another, and yet preserving themselves in existence in order to permit the analysis or the composition of the most complex representations. Within sentences, in that very depth where signification seems to be relying upon the mute support of insignificant syllables, there is always a dormant nomination, a form that holds imprisoned within its vocal walls the reflection of an invisible and yet indelible representation. For nineteenth-century philology, such analyses remained, in the literal sense of the word, 'a dead letter'. But not so for a whole way of experi'encing language - at first esoteric and mystic at the time of Saint-Marc, Reveroni, Fabre d'Olivet, Oegger, then literary when the enigma of the word re-emerged in all its density of being, with Mallarme, Roussel, Leiris, or Ponge. The idea that, when we destroy words, what is left is neither mere noise nor arbitrary, pure elements, but other words, which, when pulverized in turn, will set free still other words - this idea is at once the negative of all the modem science of languages and the myth in which we now transcribe the most obscure and the most real powers of language. It is probably because it is arbitrary, and because one can define the condition upon which it attains its power of signification, that lan­guage can become the object of a science. But it is because it has never ceased to speak within itself, because it is penetrated as far as we can reach within it by inexhaustible values, that we can speak within it in that end­less murmur in which literature is born. But in the Classical period the relation was not at all the same; the two figures fitted over each other exactly: in order that language could be entirely comprised within the general form of the proposition, each word, down to the least of its molecules, had to be a meticulous form of nomination.

103

THE ORDER OF THINGS

V DESIGNATION

And yet, the theory of 'generalized nomination' reveals at the extremity of language a certain relation to things that is of an entirely different nature from that of the prepositional form. If, fundamentally, the function of language is to name, that is, to raise up a representation or point it out, as though with a finger, then it is indication and not judgement. It is linked to things by a mark, a notation, an associated figure, a gesture of designation: nothing that could be reduced to a relation of predication. The principle of primal nomination, of the origin of words, is balanced by the formal primacy of judgement. As though, on either side of language, unfolded in all its articulations, there lay its being, in its verbal role as attribution, and its origin, in its role as primary designation. The latter permits the substitution of a sign for that which is indicated, the former makes possible the linking of one content to another. And thus we encounter once again, in their opposition yet also in their affinity, the two functions of connection and substitution that have been allotted to the sign in general with its power of analysing representation.

To bring the origin of language back into the light of day means also to rediscover the primitive moment in which it was pure designation. And one ought, by this means, to provide at the same time an explanation for its arbitrariness (since that which designates can be as different from that which it indicates as a gesture from the object towards which it is directed), and for its profound relation with that which it names (since a particular syllable or word has always been chosen to designate a parti­cular thing). The first of these requirements is fulfilled by the analysis of the language of action, the second by the study of roots. But these two things are not in opposition to one another in the same way as, in the Cratylus, are explanation in terms of 'nature' and explanation in terms of 'law'; on the contrary, they are absolutely indispensable to one another, since the first gives an account of the substitution of the sign for the thing designated and the second justifies the permanent power of designation possessed by that sign.

The language of action is spoken by the body; and yet, it is not some­thing given from the very first. All that nature permits is that man, in the various situations in which he finds himself, should be able to make gestures; his face is agitated by movements; he emits inarticulate cries - in other words, cries that are 'coined neither by the tongue nor by the lips'[65]. All this is not yet either language or even sign, but the effect and

104

speaking

consequence of our animality. This manifest agitation nevertheless has the virtue of being universal, since it depends solely upon the conformation of our organs. Hence the possibility for man to observe that it is identical in himself and his companions. He is therefore able to associate the cry he hears from another's mouth, the grimace he sees upon that other's face, with the same representations that have, on several occasions, accom­panied his own cries and movements. He is able to accept this mimesis as the mark and substitute of the other's thought. As a sign. Comprehension is beginning. He can also, in return, employ this mimesis that has become a sign in order to excite in his companions the idea that he himself is experiencing, the sensations, the needs, the difficulties that are ordinarily associated with certain gestures and certain sounds: a cry expressly directed in another's presence and towards an object, a pure interjection [66]. With this concerted use of the sign (which is already expression), something like a language is in the process of being born.

It is evident, from these analyses common to Condillac and Destutt, that the language of action does indeed link language to nature by means of a genesis - but in order to detach it from nature rather than to give it roots there, to emphasize its indelible difference from the cry and to pro­vide a basis for that which constitutes its artifice. As long as it is a simple extension of the body, action has no power to speak: it is not language. It becomes language, but only at the end of definite and complex opera­tions: the notation of an analogy of relations (the other's cry is to what he is experiencing - that which is unknown - what my cry is to my appetite or my fear); inversion of time and voluntary use of the sign before the representation it designates (before experiencing a sensation of hunger strong enough to make me cry out, I emit the cry that is associated with it); lastly, the purpose of arousing in the other the representation corre­sponding to the cry or gesture (but with this particularity, that, by emit­ting a cry, I do not arouse, and do not intend to arouse, the sensation of hunger, but the representation of the relation between this sign and my own desire to eat). Language is possible only upon the basis of this entanglement. It rests not upon a natural movement of comprehension or expression, but upon the reversible and analysable relations of signs and representations. Language does not come into being when representation is exteriorized, but only when, in a concerted fashion, it detaches a sign from itself and causes itself to be represented by that sign. It is not, therefore, because he functions as a speaking subject, or from within a language already made, that man discovers, all around him, signs that might be taken as so

105

THE ORDER OF THINGS

many mute words to be deciphered and rendered audible again; it is because representation provides itself with signs that words can come into being, and with them a whole language that is no more than the ulterior organiza­tion of vocal signs. Despite its name, the 'language of action' calls into exist­ence the irreducible network of signs that separates language from action.

And in this way it bases its artifice in nature. For the elements of which this language of action is composed (sounds, gestures, grimaces) are sug­gested successively by nature, and yet they have no identity of content -for the most part - with what they designate, but above all relations of simultaneity or succession. The cry does not resemble fear, nor the out­stretched hand the sensation of hunger. Once they have become con­certed, these signs will remain without 'fantasy and without caprice'[67], since they have been established once and for all by nature; but they will not express the nature of what they designate, for they are in no way its image. And from this starting-point men will be able to establish a lan­guage of convention: they now have at their disposal enough signs as marks for things to enable them to invent further signs that will analyse and combine the primary ones. In his Discours sur I'origine de I'indgalite[68], Rousseau made the point that no language can have an agreement between men as its basis, since such an agreement presupposes that some estab­lished, recognized, and practised language already exists; we would there­fore have to imagine it as having been received by men, not built by them. In fact, the language of action confirms this necessity and renders this hypothesis futile. Man receives from nature the material to make signs, and those signs serve him first of all as a means of reaching agree­ment with other men as to the choice of those that shall be retained, the values that they shall be recognized as possessing, and the rules for employing them; after that, they serve him as a means of forming new signs on the model of the primary ones. The first form of agreement con­sists in selecting the vocal signs (which are easier to recognize from a distance and the only ones that can be used when it is dark), the second in composing, in order to designate representations still left without signs, sounds close to those indicating neighbouring representations. It is in this way that language, properly speaking, is constituted, by a series of analogies that are a lateral extension of the language of action or at least of its vocal element: language resembles this vocal element, and 'it is this resemblance that facilitates the understanding of it. We term it analogy . . . You observe that analogy, which gives us law, does not permit us to choose signs at random or arbitrarily.'[69]

106

SPEAKING

The genesis of language in the language of action entirely avoids the alternatives of natural imitation and arbitrary convention. In that which is natural - in the signs that arise spontaneously through the medium of our bodies - there is no resemblance; and where there is employment of resemblances it is after a voluntary agreement has been reached between men. Nature juxtaposes the differences and binds them together by force;

reflection discovers the resemblances, and analyses and develops them. The first phase makes artifice possible, but with material imposed upon all men in identical fashion; the second excludes arbitrary choice but opens up channels for analysis that will not be exactly superimposable in the case of all men and all peoples. The law of nature is constituted by the difference between words and things - the vertical division between lan­guage and that lying beneath it which it is the task of language to desig­nate; the rule prescribed by conventions is the resemblance that exists between words, the great horizontal network that forms words from other words and propagates them ad infinitum.

It now becomes comprehensible why the theory of roots in no way contradicts the analysis of the language of action, but is to be found within it. Roots are those rudimentary words that are to be found, always identical, in a great number of languages - perhaps in all; they have been imposed upon language by nature in the form of involuntary cries spon­taneously employed by the language of action. It was there that men sought them out in order to give them a place in their conventional languages. And if all peoples, in all climates, chose these same elementary sounds from among the raw material of the language of action, that is because they discerned in them, though in a secondary and reflective manner, a resemblance with the object they designated, or the possibility of applying it to an analogous object. The resemblance of the root to what it names assumes its value as a verbal sign only through the agency of the convention that brought men together and regulated their language of action so as to create a language. In this way, from within representa­tion, signs are united with the very nature of what they designate, and the primitive treasury of vocables is imposed, in identical fashion, on all languages.

Roots may be formed in several ways. By onomatopoeia, of course, which is not a spontaneous expression, but the deliberate articulation of a sign that is also a resemblance: 'to make the same sound with one's voice as the object that one wishes to name'[70]. By employing a resemblance experienced in one's sensations: 'the impression made by the colour red,

107

THE ORDER OF THINGS

which is vivid, rapid, harsh to the eye, will be very well rendered by the sound R, which makes an analogous impression upon the ear'[71]. By imposing movements upon the organs of the voice analogous to those one wishes to signify: 'so that the sound resulting from the form and natural movement of the organ when placed in this state becomes the name of the object'; the throat rasps to designate the rubbing of one body against another, it hollows itself inside to indicate a concave surface [72]. Finally, by employing the sounds an organ naturally produces to desig­nate that organ: the glottal stop determined the name of the throat in which it occurs, and the dentals (d and t) are used to designate the teeth [73]. Using these conventional articulations of resemblance, every language is able to provide itself with its pack of primitive roots. The pack is a small one, since the roots are almost all monosyllabic and exist only in very small numbers - two hundred for Hebrew, according to Bergier's estimate [74]; and even smaller when one remembers that (because of the relations of resemblance that they establish) they are common to almost all of our languages: de Brosses thinks that all of them together, from all the dialects of Europe and the Orient, would not fill 'a single sheet of writing paper'. But it is on the basis of them that each language develops its own particularity: "their development is prodigious. Just as one elm seed produces a great tree, which by growing new shoots from each root produces in the end an entire forest'[75].

Language can now reveal its genealogy, the genealogy that de Brosses attempted to display in a dimension of continuous filiation that he called the 'Universal Archaeologist'[76]. At the top of this space, one would write the roots - very few in number - employed in all European and Oriental languages; below each root one could place the more complicated words derived from it, but taking care to place first those that are nearest to the roots, and to follow them in a sequence sufficiently tight for there to be as small a distance as possible-between each word in the series. In this way one would be able to constitute a number of perfect and ex­haustive series, of absolutely continuous chains in which the breaks, if there were any, would indicate the place of a word, a dialect, or a lan­guage no longer in existence[77]. Once this vast, seamless expanse had been constituted, one would have a two-dimensional space that one could cross either on abscissae or on ordinates: vertically, one would have the complete filiation of each root; horizontally, one would have the words employed in any given language; the further away one moved from the primitive roots, the more complicated - and no doubt more recent –

108

SPEAKING

would the languages defined by any transversal line become, but, at the same time, the more subtle and efficacious would the words be as instru­ments for the analysis of representations. And thus superimposed, the historical space and the grid of thought would be exactly coincidental.

This quest for the roots of language may well appear to be a return to the historical hypothesis and to the theory of mother-languages that Classicism seemed, for a time, to have suspended. In reality, an analysis of its roots does not replace language in a history that is, as it were, the environment into which it was born and in which it developed. Rather, it makes history a journey, accomplished in successive stages, across the simultaneous patterning of representation and words. In the Classical period, language is not a fragment of history authorizing at any given moment a definite mode of thought and reflection; it is an area of analysis upon which time and human knowledge pursue their journey. And the fact that language does not become - or become once again - through the agency of the root theory a historical entity is proved quite easily by the way in which etymologies were sought for in the eighteenth century. The guiding thread used for such investigations was not the material transformations undergone by the word, but the constancy of its significa­tions.

This search had two aspects: definition of the root, and isolation of the inflectional endings and prefixes. To define the root was to discover an etymology. It was an art with codified rules [78]; one had to strip the word of all the subsequent traces that might have been left upon it by combinations and inflections; arrive at a monosyllabic element; follow that element through the entire past of the language, through all the ancient 'charts and glossaries'; then follow it back into other and more primitive languages. And it must also be accepted that at any point along this backward journey the monosyllable may change: all the vowels may replace one another in the history of a root, for the vowels are the voice itself, which knows no discontinuity or rupture; the consonants, on the other hand, are modified according to certain privileged channels: gut­turals, linguals, palatals, dentals, labials, and nasals all make up families of homophonous consonants within which changes of pronunciation are made for preference, though without any obligation[79]. The only in­delible constant guaranteeing the continuity of the root throughout its history is the unity of meaning: the representative area that persists in­definitely. This is because 'nothing perhaps can limit inductions and everything can serve as a basis for them, from total resemblance to the

109

THE ORDER OF THINGS

very slightest of resemblances': the meaning of words is 'the surest source of enlightenment we can consult'[80].
VI DERIVATION

How is it that words, which in their primary essence are names and designations, and which are articulated just as representation itself is analysed, can move irresistibly away from their original signification and acquire either a broader or more limited adjacent meaning? How can they change not only their forms but their field of application? How can they acquire new sounds, and also new contents, to such an extent that various languages, equipped in the first place with a number of probably identical roots, have formed different sounds, to say nothing of words whose meanings are lost to us?

The modifications of form obey no rule, are more or less endless, and never stable. All their causes are external: ease of pronunciation, fashions, habits, climate - cold weather encourages 'unvoiced labials', hot weather 'guttural aspirates'[81 ]. The alterations of meaning, on the other hand-since they are so limited as to justify an etymologlcal science, which, if not absolutely exact, is at least 'probable'[82] - do obey fixed principles. These principles, which foment the internal history of languages, are all of a spatial order. Some concern the visible resemblance or adjacency between things; others concern the area in which language and the form it uses to preserve itself coexist. Figures and writing.

We know of two broad types of writing: that which retraces the mean­ing of words, and that which analyses and reconstitutes their sounds. Between these two there is a strict dividing-line, whether one accepts that the second took over from the first among certain peoples as the result of a veritable 'stroke of genius'[8 3], or whether one accepts - so different are they from one another - that they both appeared more or less simul­taneously, the first among graphically oriented peoples, the second among song-oriented peoples[84]. To represent the meaning of words graphically is originally to make an exact drawing of the thing to be designated. In fact, it is scarcely writing at all - at the very most a pictorial reproduction with the aid of which one can scarcely transcribe anything more than the most concrete form of narrative. According to Warburton, the Mexicans scarcely knew of any other method[85]. True writing began when the attempt was made to represent, no longer the thing itself, but one of its constituent elements, or one of the circumstances that habitually attend

110

SPEAKING

it, or again some other thing that it resembles. These three methods pro­duced three techniques: the curiological writing of the Egyptians - the crudest of the three - which employs 'the principal circumstance of a sub­ject in lieu of the whole' (a bow for a battle, a ladder for a siege); then the 'tropal' hieroglyphics - somewhat more perfected - which employ some notable circumstance (since God is all-powerful he knows every­thing and sees all that men do: he is therefore represented by an eye); finally, symbolic writing, which makes use of more or less concealed resemblances (the rising sun is expressed by the head of a crocodile whose round eyes are just level with the surface of the water) [86]. We can recognize here the three great figures of rhetoric: synecdoche, metonymy, catachresis. And it is by following the nervure laid down by these figures that those languages paralleled with a symbolic form of writing will be able to evolve. They become endowed, little by little, with poetic powers; their primary nominations become the starting-points for long metaphors; these metaphors become progressively more complicated, and are soon so far from their points of origin that it is difficult to recall them. This is how superstitions arise whereby people believe that the sun is a crocodile, or that God is a great eye keeping watch on the world; it is also how esoteric forms of knowledge arise among those (the priests) who pass on the metaphors to their successors from generation to generation; and it is how allegorical discourse (so frequent in the most ancient literatures) comes into being, as well as the illusion that knowledge consists in understanding resemblances.

But the history of a language endowed with a figurative writing soon comes to a halt. For it is hardly possible to achieve much progress in such a language. Its signs do not multiply with the meticulous analysis of representations but with the most distant analogies; so that it is the imagination of the peoples using them that is encouraged rather than their powers of reflection, their credulity rather than science. Moreover, know­ledge necessitates two kinds of apprenticeship: first in words (as with all languages), then with written signs that have no bearing upon the pro­nunciation of the words; a human life-span is not too long for this double education; and if one has had, in addition, the leisure to make some discovery, one has no signs at one's disposal to hand it on. Inversely, since it bears no intrinsic relation to the word it represents, a transmitted sign always remains dubious: from one age to the next one can never be sure that the same sound resides in the same figure. Innovations are therefore impossible, and traditions compromised. With the result that the only

111

THE ORDER OF THINGS

concern of the learned is to maintain 'a superstitious respect' for the learn­ing handed down by their ancestors and for the institutions preserving that heritage: 'they feel that any change in manners will bring change in the language, and that any change in the language will confound and annul all their knowledge'[87]. When a people possesses nothing but a figurative form of writing, its politics must exclude history, or at least all history other than pure and simple conservation. It is here, according to Volney[88], in this relation of space to language, that the essential difference between East and West is situated. As though the spatial arrangement of the language prescribed the law of time; as though their particular language did not come to men via history, but that, inversely, their only means of access to history was via their system of signs. It is in this nexus of representation, words, and space (the words representing the space of the representation, and in turn representing themselves in time) that the destiny of peoples is silently formed.

With alphabetic writing, in fact, the history of men is entirely changed. They transcribe in space, not their ideas but sounds, and from those sounds they extract the common elements in order to form a small number of unique signs whose combination will enable them to form all possible syllables and words. Whereas symbolic writing, in attempting to spatialize representations themselves, obeys the confused law of simili­tudes, and causes language to slip out of the forms of reflective thought, alphabetical writing, by abandoning the attempt to draw the repre­sentation, transposes into its analysis of sounds the rules that are valid for reason itself. So that it does not matter that letters do not represent ideas, since they can be combined together in the same way as ideas, and ideas can be linked together and disjoined just like the letters of the alphabet[89]. The disruption of the exact parallelism between representation and graphic signs makes it possible to bring language, even written language, as a totality, into the general domain of analysis, thus allowing the pro­gress of writing and that of thought to provide each other with mutual support[90]. The same graphic signs can break down all new words, and hand on each new discovery, as soon as it is made, without fear of its being forgotten; the same alphabet can be used to transcribe different languages, and thus to convey the ideas of one people to another. Since it is very easy to learn this alphabet, because of its very small number of elements, everyone is able to devote to reflection and to the analysis of ideas the time that the hieroglyphic peoples wasted in learning how to write. And so it is within language itself, exactly in that fold of words

112

SPEAKING

where analysis and space meet, that the first but endless possibility of progress arises. In its root, progress, as defined in the eighteenth century, is not a movement within history, but the result of a fundamental relation between space and language:

The arbitrary signs of language and writing provide men with the means of ensuring the possession of their ideas and of communicating them to others in the manner of an inheritance, constantly augmented with the new discoveries of each age; and the human race, considered from its origin, appears to the eyes of the philosopher as an immense whole that itself possesses, like every individual, its childhood and its progress[91].

Language gives the perpetual disruption of time the continuity of space, and it is to the degree that it analyses, articulates, and patterns repre­sentation that it has the power to link our knowledge of things together across the dimension of time. With the advent of language, the chaotic monotony of space is fragmented, while at the same time the diversity of temporal successions is unified.

There remains one last problem, however. For though writing is in­deed the buttress and ever-watchful guardian of these progressively more refined analyses, it is neither their principle nor even their initial move­ment. This latter is a slipping movement common to attention, to signs, and to words. In any representation, the mind can attach itself, and attach a verbal sign, to one element of that representation, to a circumstance attending it, to some other, absent, thing that is similar to it and is recalled to memory on account of it [92]. There is no doubt that this is how lan­guage developed and gradually drifted away from primary designations. Originally, everything had a name - a proper or peculiar name. Then the name became attached to a single element of the thing, and became applicable to all the other individual things that also contained that cle­ment: it is no longer a particular oak that is called tree, but anything that includes at least a trunk and branches. The name also became attached to a conspicuous circumstance: night came to designate, not the end of this particular day, but the period of darkness separating all sunsets from all dawns. Finally, it attached itself to analogies: everything was called a leaf that was as thin and flexible as the leaf of a tree [93]. The progressive analysis and more advanced articulation of language, which enable us to give a single name to several things, were developed along the lines of these three fundamental figures so well known to rhetoric: synecdoche,

113

THE  ORDER OF THINGS

metonymy, and catachresis (or metaphor, if the analogy is less imme­diately perceptible). For these things are not the effect of a refinement of style; on the contrary, they reveal the mobility peculiar to all language whenever it is spontaneous: 'La Halle produces more figures of speech in one market day than our academic assemblies do in a week'[94]. It is very probable that this mobility was even greater in the beginnings of language than it is now: today, the analysis is so detailed, the grid so fine, the rela­tions of coordination and subordination are so firmly established, that words scarcely have any opportunity to move from their places. But at the beginning of human history, when words were few, when repre­sentations were still confused and not well analysed, when the passions both modified them and provided them with a basis, words had greater mobility. One might even say that words were figurative- before being proper: in other words, that they had scarcely attained their status as particular names before they were being scattered over representations by the force of spontaneous rhetoric. As Rousseau says, we probably talked about giants before designating men [95]. Boats were originally desig­nated by their sails, and the soul, the 'psyche', was initially given the figurative form of the moth [96].

So that at the base of spoken language, as with writing, what we dis­cover is the rhetorical dimension of words: that freedom of the sign to alight, according to the analysis of representation, upon some internal element, upon some adjacent point, upon some analogous figure. And if languages possess the diversity we observe in them; if from the starting-point of their primitive designations, which were doubtless common to them all owing to the universality of human nature, they have not ceased to develop according to the dictates of differing forms; if they have all had their own history, fashions, customs, and periods of oblivion; this is because words have their locus, not in time, but in a space in which they are able to find their original site, change their positions, turn back upon themselves, and slowly unfold a whole developing curve: a tropological space. And in this way one returns once more to what had served as a starting-point for reflection upon language. Language was of all signs the one having the property of being sequential: not because it was itself part of a chronology, but because it drew out into sequential sounds the simultaneity of representation. But this succession, which analyses dis­continuous elements and brings them into view one after the other, traverses the space offered by representation to the mind's eye. So that language merely arranges into a linear order the scattered fragments

114

SPEAKING

represented. The proposition unfolds and makes audible the figure that rhetoric makes visible. Without this tropological space, language would not be formed of all those common names that make it possible to estab­lish a predicative relation. And without this analysis of the words, the figures would have remained mute and momentary; and since they would have been perceived only in the incandescence of the instant, they would have fallen forthwith into a darkness in which there is not even any time.

From the theory of the proposition to that of derivation, all Classical reflection upon language-all that was called 'general grammar'-is merely a detailed commentary upon the simple phrase: 'language analyses'. It was upon this point, in the seventeenth century, that the whole Western experience of language foundered - the experience that had always led men to believe, until then, that language spoke.
VII THE QUADRILATERAL OF LANGUAGE

A few concluding remarks. The four theories - of the proposition, of articulation, of designation, and of derivation - form, as it were, the seg­ments of a quadrilateral. They confront each other in pairs and reinforce each other in pairs. Articulation gives content to the pure and still empty verbal form of the proposition; it fills that form, yet is in opposition to it, as a nomination that differentiates things is in opposition to the pre­dication that links them together. The theory of designation reveals the point of attachment of all the nominal forms cut out by articulation; but they are in opposition to articulation, just as the instantaneous, gestural, perpendicular designation is in opposition to patterns based on generali­ties. The theory of derivation indicates the continuous movement of words from their source of origin, but the slipping that occurs on the surface of representation is in opposition to the single stable bond that links one root to one representation. Finally, derivation leads back to the proposition, since without it all designation would remain folded in on itself and could never acquire the generality that alone can authorize a predicating link; yet derivation is created by means of a spatial figure, whereas the proposition unfolds in obedience to a sequential and linear order.

It should be noted that there also exist diagonal relations, as it were, between the opposing comers of this rectangle. First of all, between articulation and derivation: if the existence of an articulated language is

115

THE ORDER OF THINGS

possible, with words in juxtaposition, interlocking, or arranging them­selves in relation to one another, then it is so only in so far as the words of that language - starting from their original values and from the simple act of designation that was their basis - have never ceased to move further and further away, by a process of derivation, thus acquiring a variable extension; hence an axis that cuts across the whole quadrilateral of lan­guage; and it is along this line that the state of a language is marked off: its articulative capacities are determined by the distance it has moved along the line of derivation; such a reading defines both its historical posture and its power of discrimination. The other diagonal runs from the proposition back to the origin, that is, from the affirmation at the heart of every act of judgement to the designation implied by any act of nomination; it is along this axis that the relation of words-to what they represent is established: here it becomes apparent that words never speak anything other than the being of representation, but that they always name something represented. The first diagonal marks the progress of a language from the point of view of its specification; the second the end­less interleaving of language and representation - the duplicating process which is the reason why the verbal sign is always representing a repre­sentation. On this latter line, the word functions as a substitute (with its power to represent); on the former, as an element (with its power to make combinations and break them down).

At the point where these two diagonals intersect, at the centre of the quandrilateral, where the duplicating process of representation is revealed as analysis, where the substitute has the power of distribution, and where, in consequence, there resides the possibility and the principle of a general taxonomy of representation, there is the name. To name is at the same time to give the verbal representation of a representation, and to place it in a general table. The entire Classical theory of language is organized around this central and privileged entity. All the various functions of language intersect within it, since it is by nomination that representations are enabled to enter as figures into a proposition. It is therefore also through nomination that discourse is articulated upon knowledge. Only the judge­ment, of course, can be true or false. But if all names were exact, if the analysis upon which they are based had been perfectly thought out, if the language in question had been 'well made', there would be no difficulty in pronouncing true judgements, and error, should it occur, would be as easy to uncover and as evident as in a calculation in algebra. But the imperfection of analysis, and all the slight shifts caused by derivation, have

116

SPEAKING

caused names to be attached to analyses, abstractions, and combinations that are in fact illegitimate. There would be no disadvantage in this (any more than in giving names to fabulous monsters) if words did not posit themselves as being representations of representations: with the result that we cannot think of a word - however abstract, general, and empty it may be - without affirming the possibility of what it represents. This is why, in the middle of the quadrilateral of language, the name appears both as die point upon which all the structures of a language converge (for the name is its most secret, most closely guarded figure, the pure internal result of all its conventions, rules, and history), and as the point from which all language in general can enter into a relation with the truth according to which it will be judged.

This is the nexus of the entire Classical experience of language: the reversible character of grammatical analysis, which is at one and the same time science and prescription, a study of words and a rule for constructing them, employing them, and remoulding them into their representative function; the fundamental nominalism of philosophy from Hobbes to Ideology, a nominalism that is inseparable from a critique of language and from all that mistrust with regard to general and abstract words that we find in Malebranche, Berkeley, Condillac, and Hume; the great Utopia of a perfectly transparent language in which things themselves could be named without any penumbra of confusion, cither by a totally arbitrary but precisely thought-out system (artificial language), or by a language so natural that it would translate thought like a face expressing a passion (it was this language of immediate sign that Rousseau dreamed of in the first of his Dialogues). One might say that it is the Name that organizes all Classical discourse; to speak or to write is not to say things or to express oneself, it is not a matter of playing with language, it is to make one's way towards the sovereign act of nomination, to move, through lan­guage, towards the place where things and words are conjoined in their common essence, and which makes it possible to give them a name. But once that name has been spoken, all the language that has led up to it, or that has been crossed in order to reach it, is reabsorbed into it and dis­appears. So that Classical discourse, in its profound essence, tends always towards this boundary; but, in surviving it, pushes the boundary further away. It continues on its way in the perpetually maintained suspension of the Name. This is why, in its very possibility, it is linked with rhetoric, that is, with all the space that surrounds the name, causes it to oscillate around what it represents, and reveals the elements, or the adjacency, or

117

THE ORDER OF THINGS

the analogies of what it names. The figures through which discourse passes act as a deterrent to the name, which then arrives at the last moment to fulfil and abolish them. The name is the end of discourse. And possibly all Classical literature resides in this space, in this striving to reach a name that remains always formidable because it exhausts, and thereby kills, the possibility of speech. It is this striving movement that carried the ex­perience of language onwards from the restrained confession of La Princesse de Cleves to the immediate violence of Juliette. In the latter, nomination is at last posited in its starkest nudity, and the rhetorical figures, which until then had been holding it in suspense, collapse and become the endless figures of desire - and the same names, constantly repeated, exhaust themselves in their effort to cross those figures, without ever being able to reach their end.

All Classical literature resides in the movement that proceeds from the figure of the name to the name itself, passing from the task of naming the same thing yet again by means of new figures (which is preciosity) to that of finding words that will at last name accurately that which has never been named before or that which has remained dormant in the enveloping folds of words too far removed from it: of this latter kind are those secrets of the soul, those impressions born at the frontier of things and the body for which the language of the Cinquieme Reverie made itself spon­taneously transparent. Later, Romanticism was to believe that it had broken with the previous age because it had learned to name things by their name. In fact all Classicism tended towards this end: Hugo was the fulfilment of Voiture's promise. But, by this very fact, the name ceases to be the reward of language; it becomes instead its enigmatic raw material. The only moment - an intolerable one, for long buried in secrecy - at which the name was at the same time the fulfilment and the substance of language, its promise and its raw material, was when, with Sade, it was traversed throughout its whole expanse by desire, of which it was at once the place of occurrence, the satisfaction, and the perpetual recurrence. Hence the fact that Sade's works play the role of an incessant primordial murmur in our culture. With this violence of the name being uttered at last for its own sake, language emerges in all its brute being as a thing;

the other 'parts of oration' assume in turn their autonomy, escaping from the sovereignty of the name, and ceasing to form around it an accessory circle of ornaments. And since there is no longer any particular beauty in 'retaining' language around the frontiers of the name, in making it show what it does not say, the result will be a non-discursive discourse

118

SPEAKING

whose role will be to manifest language in its brute being. This proper being of language is what the nineteenth century was to call the Word (le Verbe), as opposed to the Classical 'verb', whose function is to pin language, discreetly but continuously, to the being of representation. And the discourse that contains this being and frees it for its own sake is literature.

Around the privileged position occupied by the name in the Classical period, the theoretical segments (proposition, articulation, designation, and derivation) constitute the frontiers of what the experience of language was at that time. Our step-by-step analysis of these segments was not undertaken in order to provide a history of grammatical conceptions in the seventeenth and eighteenth centuries, or to establish the general out­line of what men might have thought about language at that time. The intention was to determine in what conditions language could become the object of a period's knowledge, and between what limits this epistemological domain developed. Not to calculate the common denominator of men's opinions, but to define what made it possible for opinions about language - whatever the opinions may have been - to exist at all. This is why our rectangle defines a periphery rather than provides an interior figure, and it shows how language intertwines with what is exterior and indispensable to it. We have seen that language existed only by virtue of the proposition: without at least the implicit presence of the verb to be, and of the predicative relation for which it provides authority, it would not be language that we were dealing with at all, but a collection of signs like any others. The prepositional form posits as a condition of language the affirmation of a relation of identity or difference: we can speak only in so far as this relation is possible. But the other three theoretical segments enclose a quite different requirement: if it is to be possible to derive words from their first source, if an original kinship is to be already in existence between a root and its signification, if there is to be an articulated pattern­ing of representations, there must be a murmur of analogies rising from things, perceptible even in the most immediate experience; there must be resemblances that posit themselves from the very start. If everything were absolute diversity, thought would be doomed to singularity, and like Condillac's statue before it began to remember and make comparisons, it would be doomed also to absolute dispersion and absolute monotony. Neither memory nor imagination, nor, therefore, reflection, would be possible. And it would be impossible to compare things with each other, to define their identical characteristics, and to establish a common name 119

THE ORDER OF THINGS

for them. There would be no language. If language exists, it is because below the level of identities and differences there is the foundation pro­vided by continuities, resemblances, repetitions, and natural criss-crossings. Resemblance, excluded from knowledge since the early seventeenth cen­tury, still constitutes the outer edge of language: the ring surrounding the domain of that which can be analysed, reduced to order, and known. Discourse dissipates the murmur, but without it it could not speak.

It is now possible to grasp how solid and tightly knit the unity of language is in the Classical experience. It is this unity that, through the play of an articulated designation, enables resemblance to enter the pro-positional relation, that is, a system of identities and differences as based upon the verb to be and manifested by the network of names. The funda­mental task of Classical 'discourse' is to ascribe a name to things, and in that name to name their being. For two centuries. Western discourse was the locus of ontology. When it named the being of all representation in general, it was philosophy: theory of knowledge and analysis of ideas. When it ascribed to each thing represented the name that was fitted to it, and laid out the grid of a well-made language across the whole field of representation, then it was science - nomenclature and taxonomy.

NOTES

[1] John Locke, An essay concerning human understanding (1690, book III, chap. II, section 2).

[2] Condillac, Grammaire (CEuvres, t. V, pp. 39-40).

[3] Destutt de Tracy, Elements d'Ideologie, t.1.

[4] U. Domergue, Grammaire generale analytique (Paris, year VII, t. I, pp. 10-11).

[5] Condillac, Grammaire (CEuvres, t. V, p. 336).

[6] Abbe Sicard, Elements de grammaire generate (3rd edn., Paris, 1808, t. II, p. 113).

[7] Destutt de Tracy, Elements d'ldeologie, t.1, pp. 261-6.

[8] Encyclopedie, article on 'Langue'.

[9] Condillac, Grammaire (CEuvres, t. V, pp. 4-5 and 67-73).

[10] Adam Smith, Considerations concerning the formation of languages.

[11] Destutt de Tracy, Elements d'ldeologie, preface, 1.1, p. 2.

[12] C. Bonnet, Contemplation de la nature (CEuvres completes, t. IV, p. 136 note).

[13] Destutt de Tracy, Memoires de I'Academic des Sciences morales et politiques, t. III, p. 535.

120

SPEAKING

[14] D'Alembert, Discours preliminaire de 'I'Encyclopedic'.

[15] Destutt de Tracy, Elements d'ldeologie, 1.1, p. 24.

[16] Diderot, article on 'Encyclopedie' in the Encyclopedie, t. V, p. 637.

[17] Rousseau, Essai sur I'origine des langues (CEuvres, Paris, 1826 edn., t. XIII, pp. 220-1).

[18] Cf. Michaclis, De I'influence des opinions sur le langage (1759, Fr. trans. Paris, 1762): we know that the Greeks identified both fame and public opinion by the word doxa, and that the Teutons believed in the fertilizing virtues of ' storms from their expression das Hebe Cewitter (pp. 24 and 40).

[19] It is thought (cf., for instance, W. Warburton, The divine legation of Moses (1737-41, book IV, sections II-VI)) that the knowledge of the Ancients, and, above all, that of the Egyptians, was not first of all secret then subsequently made public, but that, having first been constructed communally, it was later confiscated, masked, and travestied by the priests. Esoterism, far from being the first form of knowledge, is only a perversion of it.

[20] E. Guichard, Harmonic etymologique (1606). Cf. other classifications of the same type in Scaliger, Diatribe de Europaeorum linguis, or Wilkins, An essay towards real character (London, 1668, p. 3 et sea.').

[21] Le Blan, Theorie nouveclle de la parole (Paris, 1750), according to which Latin bequeathed nothing to Italian, Spanish, or French other than 'the heritage of a few words'.

[22] Abbe Girard, Les Vrais Principes de la langue francaise (Paris, 1747, t. I, pp. 22-5).

[23] On this problem and the discussions it has raised, cf. Bauzee, Grammaire generale (Paris, 1767); Abbe Batteux, Nouvel examen du prejuge de I'inversion (Paris, 1767); and Abbe d'Olivet, Remarques sur la langue francaise (Paris, 1771).

[24] Abbe Pluche, La Mecaniqne des langues (reissued 1811, p. 26).

[25] Ibid., p. 23.

[26] Cf. for example. Burner, Grammaire francaise (Paris, new edn., 1723). This is why, at the end of the eighteenth century, the expression 'philosophical grammar' came to be preferred to that of'general grammar', which 'would be that of all languages'; D. Thiebault, Grammaire philosophique (Paris, 1802, t. I, pp. 6 and 7).

[27] Destutt de Tracy, Elements d'ldeologie, t. II, p. 87.

[28] J. Itard, Rapport sur les nouveaux developpements de Victor de 1'Aveyron, 1964 edn., p. 209.

[29] Destutt de Tracy, Elements d'ldeologie, t. II, p. 60.

[30] U. Domergue, Grammaire generate analytique, p. 34.

[31] Hobbes, Logic, chap. Ill, section 3.

[32] Adam Smith, Considerations concerning the formation of languages.

[33] Logique de Port-Royal, pp. 106-7.

[34] Condillac, Craimmaire, p. 115.

121

THB ORDBK OF THINGS

[35] In the French this phrase reads:'. . . au lieu de dire "je suis chantant", on dit "je chante"'. The significance of the author's remark is lost on the English reader since he can indeed say 'I am singing' whereas the Frenchman cannot say 'je suis chantant'. This form, often known as the 'progressive', is not to be found in French, or in most other languages. [Translator's note.]

[36] Logique de Port-Royal, p. 107. Cf. Condillac, Grammaire, pp. 132-4. In his L'Origine des connaissances, the history of the verb is analysed in a somewhat different fashion, but not its function. D. Thiebault, Grammaire philosophique, t.1, p. 216.

[37] Cf. Logique de Port-Royal, p. 107, and Abbe Girard, Les Vrais Prindpes de la langue francaise, p. 56.

[38] Bauzee, Grammaire generale, I, p. 426 et seq.

[39] Condillac, Grammaire, pp. 185-6.

[40] Destutt de Tracy, elements d'ldeologie, t. II, p. 64.

[41] U. Domergue, Grammaire generale analytique, p. n.

[42] Condillac, Grammaire, p. 152.

[43] Ibid.. p. 155.

[44] Ibid., p. 153. Cf. also A. Smith, Considerations concerning the formation of languages, pp. 408-10.

[45] Logique de Port-Royal, p. 101.

[46] Ibid., pp. 59-60.

[47] Ibid., p. lor.

[48] Duclos, Commentaire a la 'Crammaire de Port-Royal' (Paris, 1754, p. 213).

[49] J.-B. Lemercier. Lettre sur la possibility de faire de la grammaire un Art-Science (Paris, 1806, pp. 63-5).

[50] James Harris, Hermes.

[51] A. Smith, Considerations concerning the formation of languages, pp. 430-1.

[52] Bauzee (Grammaire generale) was the first to employ the term 'com­plement'.

[53] Logique de Port-Royal, p. 117 et seq.

[54] Abbe Sicard, elements de grammaire generale, t. II, p. 2.

[55] Sylvestre de Saci, Prindpes de grammaire generale (1799). Cf. also U. Domergue, Grammaire generale analytique, pp. 29-30.

[56] Cf., for example, Abbe Girard, Les Vrais Prindpes de la langue francaise, pp. 82-3.

[57] Logique de Port-Royal, p. 59.

[58] Batteux, Nouvel examen du prejuge de I'inversion, pp. 23-4.

[59] Ibid., pp. 24-8.

[60] Le Bel, Anatomie de la langue latine (Paris, 1764, p. 24).

[61] Ibid., p. 8.

[62] D. Thiebault, Grammaire philosophique, pp. 172-3.

[63] Court de Gebelin, Histoire naturelle de la parole (1816 edn., pp. 98-104).

122

SPEAKING

[64] Rousseau, Essai sur 1'origine des langues (CEuvres, 1826 edn., t. XIII, pp. 144-51 and 188-92).

[65] Condillac, Crammaire, p. 8.

[66] All the parts of discourse are, therefore, merely the fragments, broken down and recombined, of that initial interjection (Destutt de Tracy, elements d'ldeologie, t. II, p. 75).

[67] Condillac, Crammaire, p. 10.

[68] Rousseau, Discours sur 1'origine de I'inegalite (cf. Condillac, Grammaire, p. 27, note 1).

[69] Condillac, Grammaire, pp. 11-12.

[70] De Brosses, Traite de la formation mecanique des langues (Paris, 1765, t.1, P. 9).

[71] Abbe Copineau, Essai synthetique sur 1'origine et la formation des langues (Paris, 1774, pp. 34-5).

[72] De Brosses, Traite de la formation mecanique des langues, t.1, pp. 16-18.

[73] Ibid., t.1, p. 14.

[74] Bergier, Les Elements primitifs des langues (Paris, 1764, pp. 7-8).

[75] De Brosses, Traite de la formation mecanique des langues, t.1, p. 18.

[76] Ibid., t. II, pp. 490-9.

[77] Ibid., t.1, preface, p. 1.

[78] Cf, especially, Turgot's article on 'Etymologie' in the Encyclopedie.

[79] These, together with a few accessory variants, are the only laws of phonetic variation recognized by de Brosses (Traite de la formation mecanique des langues, pp. 108-23), Bergier (Elements primitifs des langues, pp. 45-62), Court de Gebelin (Histoire naturelle de la parole, pp. 59-64), and Turgot ('Etymologie' in the Encyclopedie).

[80] Turgot, article 'Etymologic' in the Encyclopedic. Cf. de Brosses, op. cit., p. 420.

[81] De Brosses, Traite de la formation mecanique des langues, t.1, pp. 66-7.

[82] Turgot, article 'Etymologie' in the Encyclopedic.

[83] Duclos, Remarques sur la grammaire generale, pp. 43-4.

[84] Destutt de Tracy, Elements d'ldeologie, II, pp. 307-12.

[85] Warburton, The divine legation of Moses.

[86] Warbuton, op. cit.

[87] Destutt de Tracy, Elements d'ldeologie, t. II, pp. 284-300.

[88] Volney, Les Ruines (Paris, 1791, chap. XIV).

[89] Condillac, Grammaire, chap. 2.

[90] Adam Smith, Considerations concerning the formation of languages.

[91] Turgot, Tableau des progres successifs de l'esprit humain (17 50; CEuvres, ed. Schelle, p. 215).

[92] Condillac, Essai sur 1'origine des connaissances humaines (CEuvres, t. I, PP. 75-87).

123

THE ORDER OF THINGS

[93]     Du Marsais, Traite des tropes (1811 edn., pp. 150-1).

[94]     Ibid., p. 2.

[95]   Rousseau, Essai sur l'origine des Iangues, pp. 152-3.

[96]   De Brosses, Traite de la prononciation mecaniqe, p. 267.

124

CHAPTER 5
Classifying
I WHAT THE HISTORIANS SAY

Histories of ideas or of the sciences - by which is meant here an average cross-section of them - credit the seventeenth century, and especially the eighteenth, with a new curiosity: the curiosity that caused them, if not to discover the sciences of life, at least to give them a hitherto unsuspected scope and precision. A certain number of causes and several essential manifestations are traditionally attributed to this phenomenon.

On the side of origins or motives, we place the new privileges accorded to observation: the powers attributed to it since Bacon and the technical improvements introduced in it by the invention of the microscope. Alongside these is set the then recently attained prestige of the physical sciences, which provided a model of rationality; since it had proved pos­sible, by means of experimentation and theory, to analyse the laws of movement or those governing the reflection of light beams, was it not normal to seek, by means of experiments, observations, or calculations, the laws that might govern the more complex but adjacent realm of living beings? Cartesian mechanism, which subsequently proved an obstacle, was used at first, the historians tell us, as a sort of instrument of transference, and led, rather in spite of itself, from mechanical rationality to the discovery of that other rationality which is that of the living being. Still on the side of causes, and in a somewhat pell-mell fashion, the his­torians of ideas place a variety of new interests: the economic attitude towards agriculture - the Physiocrats' beliefs were evidence of this, but so too were the first efforts to create an agronomy; then, half-way between husbandry and theory, a curiosity with regard to exotic plants and animals, which attempts were made to acclimatize, and of which the great voyages of inquiry or exploration - that of Toumefort to the Middle East, for example, or that of Adanson to Senegal - brought back

125

THE ORDER OF THINGS

descriptions, engravings, and specimens; and then, above all, the ethical valorization of nature, together with the whole of that movement, ambiguous in its principle, by means of which - whether one was an aristocrat or a bourgeois - one 'invested' money and feeling into a land that earlier periods had for so long left fallow. Rousseau, at the heart of the eighteenth century, was a student of botany.

In their list of manifestations, the historians then include the varied forms that were taken by these new sciences of life, and the 'spirit', as they put it, that directed them. Apparently, under the influence of Des­cartes, they were mechanistic to begin with, and continued to be so to the end of the seventeenth century; then the first efforts of an infant chemistry made its imprint upon them, but throughout the eighteenth century the vitalist themes are thought to have attained or returned to their privileged status, finally coalescing to form a unitary doctrine - that 'vitalism' which in slightly differing forms was professed by Bordeu and Barthez in Montpellier, by Blumenbach in Germany, and by Diderot then Bichat in Paris. Under these different theoretical regimens, questions were asked that were almost always the same but were given each time a different solution: the possibility of classifying living beings-some, like Linnaeus, holding that all of nature can be accommodated within a taxonomy, others, like Buffon, holding that it is too rich and various to be fitted within so rigid a framework; the generative process, with the more mechanistically minded in favour of preformation, and others believing in the specific development of germs; analysis of functions (circulation after Harvey, sensation, motivity, and, towards the end of the century, respiration).

After examining these problems and the discussions they give rise to, it is simple enough for the historians to reconstruct the great controversies that are said to have divided men's opinions and passions, as well as their reasoning. By these means they believe that they can discover the traces of a major conflict between a theology that sees the providence of God and the simplicity, mystery, and foresight of his ways residing beneath each form and in all its movements, and a science that is already attempt­ing to define the autonomy of nature. They also recognize the contra­diction between a science still too attached to the old pre-eminence of astronomy, mechanics, and optics, and another science that already sus­pects all the irreducible and specific contents there may be in the realms of life. Lastly, the historians see the emergence, as though before their very eyes, of an opposition between those who believe in the immobility

126

CLASSIFYING

of nature - in the manner of Tournefort, and above all Linnaeus - and those who, with Bonnet, Benoit de Maillet, and Diderot, already have a presentiment of life's creative powers, of its inexhaustible power of transformation, of its plasticity, and of that movement by means of which it envelops all its productions, ourselves included, in a time of which no one is master. Long before Darwin and long before Lamarck, the great debate on evolution would appear to have been opened by the Telliamed, the Pahngencsie and the Reve de d'Alembert. Mechanism and theology, supporting one another or ceaselessly conflicting with one another, tended to keep the Classical age as close as possible to its origin - on the side of Descartes and Malebranche; whereas, opposite them, irreligion and a whole confused intuition of life, conflicting in turn (as in Bonnet) or acting as accomplices (as with Diderot), are said to be drawing it towards its imminent future - towards the nineteenth century, which is supposed to have provided the still obscure and fettered endeavours of the eighteenth with their positive and rational fulfilment in a science of life which did not need to sacrifice rationality in order to preserve in the very quick of its consciousness the specificity of living things, and that somewhat subterranean warmth which circulates between them - the object of our knowledge - and us, who are here to know them.

It would be pointless to go back over the presuppositions inherent in such a method. Let it suffice here to point out its consequences: the diffi­culty of apprehending the network that is able to link together such diverse investigations as attempts to establish a taxonomy and micro­scopic observations; the necessity of recording as observed facts the con­flicts between those who were fixists and those who were not, or between the experimentalists and the partisans of the system; the obligation to divide knowledge into two interwoven fabrics when in fact they were alien to one another - the first being defined by what was known already and from elsewhere (the Aristotelian or scholastic inheritance, the weight of Cartesianism, the prestige of Newton), the second by what still remained to be known (evolution, the specificity of life, the notion of organism); and above all the application of categories that are strictly anachronistic in relation to this knowledge. Obviously, the most impor­tant of all these refers to life. Historians want to write histories of biology in the eighteenth century; but they do not realize that biology did not exist then, and that the pattern of knowledge that has been familiar to us for a hundred and fifty years is not valid for a previous period. And that, if biology was unknown, there was a very simple reason for it: that

127

THE ORDER OF THINGS

life itself did not exist. All that existed was living beings, which were viewed through a grid of knowledge constituted by natural history.
II NATURAL HISTORY

How was the Classical age able to define this realm of 'natural history', the proofs and even the unity of which now appear to us so distant, and as though already blurred? What is this field in which nature appeared sufficiently close to itself for the individual beings it contained to be classified, and yet so far removed from itself that they had to be so by the medium of analysis and reflection?

One has the impression - and it is often expressed - that the history of nature must have appeared as Cartesian mechanism ebbed. When it had at last become clear that it was impossible to fit the entire world into the laws of rectilinear movement, when the complexity of the vegetable and animal kingdoms had sufficiently resisted the simple forms of extended substance, then it became necessary for nature to manifest itself in all its strange richness; and the meticulous observation of living beings was thus born upon the empty strand from which Cartesianism had just with­drawn. Unfortunately, things do not happen as simply as that. It is quite possible - though it would be a matter requiring careful scrutiny - that one science can arise out of another; but no science can be generated by the absence of another, or from another's failure, or even from some obstacle another has encountered. In fact, the possibility of natural history, with Ray, Jonston, Christophorus Knauth, is contemporaneous with Cartesianism itself, and not with its failure. Mechanism from Descartes to d'Alembert and natural history from Tournefort to Daubenton were authorized by the same episteme.

For natural history to appear, it was not necessary for nature to become denser and more obscure, to multiply its mechanisms to the point of acquiring the opaque weight of a history that can only be retraced and described, without any possibility of measuring it, calculating it, or explaining it; it was necessary - and this is entirely the opposite-for History to become Natural. In the sixteenth century, and right up to the middle of the seventeenth, all that existed was histories: Belon had written a History of the nature of birds; Duret, an Admirable history of plants; Aldrovandi, a History of serpents and dragons. In 1657, Jonston published a Natural history of quadrupeds. This date of birth is not, of course, absolutely definitivel[1]; it is there only to symbolize a landmark, and to indicate,

128

CLASSIFYING

from afar, the apparent enigma of an event. This event is the sudden separation, in the realm of Historia, of two orders of knowledge hence­forward to be considered different. Until the time of Aldrovandi, History was the inextricable and completely unitary fabric of all that was visible of things and of the signs that had been discovered or lodged in them: to write the history of a plant or an animal was as much a matter of des­cribing its elements or organs as of describing the resemblances that could be found in it, the virtues that it was thought to possess, the legends and stories with which it had been involved, its place in heraldry, the medica­ments that were concocted from its substance, the foods it provided, what the ancients recorded of it, and what travellers might have said of it. The history of a living being was that being itself, within the whole semantic network that connected it to the world. The division, so evident to us, between what we see, what others have observed and handed down, and what others imagine or naively believe, the great tripartition, apparently so simple and so immediate, into Observation, Document, and Fable, did not exist. And this was not because science was hesitating between a rational vocation and the vast weight of naive tradition, but for the much more precise and much more constraining reason that signs were then part of things themselves, whereas in the seventeenth century they become modes of representation.

When Jonston wrote his Natural history of quadrupeds, did he know any more about them than Aldrovandi did, a half-century earlier? Not a great deal more, the historians assure us. But that is not the question. Or, if we must pose it in these terms, then we must reply that Jonston knew a great deal less than Aldrovandi. The latter, in the case of each animal he examined, offered the reader, and on the same level, a description of its anatomy and of the methods of capturing it; its allegorical uses and mode of generation; its habitat and legendary mansions; its food and the best ways of cooking its flesh. Jonston subdivides his chapter on the horse under twelve headings: name, anatomical parts, habitat, ages, generation, voice, movements, sympathy and antipathy, uses, medicinal uses [2]. None of this was omitted by Aldrovandi, and he gives us a great deal more besides. The essential difference lies in what is missing in Jonston. The whole of animal semantics has disappeared, like a dead and useless limb. The words that had been interwoven in the very being of the beast have been unravelled and removed: and the living being, in its anatomy, its form, its habits, its birth and death, appears as though stripped naked. Natural, history finds its locus in the gap that is now opened up between

129

THE ORDER OF THINGS

things and words - a silent gap, pure of all verbal sedimentation, and yet articulated according to the elements of representation, those same elements that can now without let or hindrance be named. Things touch against the banks of discourse because they appear in the hollow space of representation. It is not therefore at the moment when one gives up cal­culation that one finally begins to observe. We must not see the constitu­tion of natural history, with the empirical climate in which it develops, as an experiment forcing entry, willy-nilly, into a knowledge that was keeping watch on the truth of nature elsewhere; natural history - and this is why it appeared at precisely this moment - is the space opened up in representation by an analysis which is anticipating the possibility of naming; it is the possibility of seeing what one will be able to say, but what one could not say subsequently, or see at a distance, if things and words, distinct from one another, did not, from the very first, communicate in a representation. The descriptive order proposed for natural history by Linnaeus, long after Jonston, is very characteristic. According to this order, every chapter dealing with a given animal should follow the following plan: name, theory, kind, species, attributes, use, and, to con­clude, Litteraria. All the language deposited upon things by time is pushed back into the very last category, like a sort of supplement in which discourse is allowed to recount itself and record discoveries, tra­ditions, beliefs, and poetical figures. Before this language of language, it is the thing itself that appears, in its own characters, but within the reality that has been patterned from the very outset by the name. The constitu­tion of a natural science in the classical age is not the effect, either direct or indirect, of the transference of a rationality formed elsewhere (for geometrical or mechanical purposes). It is a separate formation, one that has its own archaeology, even though it is linked (though in a correlative and simultaneous mode) to the general theory of signs and to the project for a universal mathesis.

Thus the old word 'history' changes its value, and perhaps rediscovers one of its archaic significations. In any case, though it is true that the historian, for the Greeks, was indeed the individual who sees and who recounts from the starting-point of his sight, it has not always been so in our culture. Indeed, it was at a relatively late date, on the threshold of the Classical age, that he assumed - or resumed - this role. Until the mid-seventeenth century, the historian's task was to establish the great com­pilation of documents and signs - of everything, throughout the world, that might form a mark, as it were. It was the historian's responsibility to 130

CLASSIFYING

restore to language all the words that had been buried. His existence was defined not so much by what he saw as by what he retold, by a secondary speech which pronounced afresh so many words that had been muffled.

The Classical age gives history a quite different meaning: that of under­taking a meticulous examination of things themselves for the first time, and then of transcribing what it has gathered in smooth, neutralized, and uithful words. It is understandable that the first form of history consti­tuted in this period of 'purification' should have been the history of nature. For its construction requires only words applied, without inter­mediary, to things themselves. The documents of this new history are not other words, texts or records, but unencumbered spaces in which things are juxtaposed: herbariums, collections, gardens; the locus of this history is a non-temporal rectangle in which, stripped of all commentary, of all enveloping language, creatures present themselves one beside another, their surfaces visible, grouped according to their common features, and thus already virtually analysed, and bearers of nothing but

their own individual names. It is often said that the establishment of botanical gardens and zoological collections expressed a new curiosity about exotic plants and animals, In fact, these had already claimed men's interest for a long while. What had changed was the space in which it was possible to see them and from which it was possible to describe them. To the Renaissance, the strangeness of animals was a spectacle: it was featured in fairs, in tournaments, in fictitious or real combats, in reconstitutions of legends in which the bestiary displayed its ageless fables. The natural history room and the garden, as created in the Classical period, replace the circular procession of the 'show' with the arrangement of things in a 'table'. What came surreptitiously into being between the age of the theatre and that of the catalogue was not the desire for knowledge, but a new way of connecting things both to the eye and to discourse. A new way of making history.

We also know what methodological importance these 'natural' alloca­tions assumed, at the end of the eighteenth century, in the classification of words, languages, roots, documents, records-in short, in the con­stitution of a whole environment of history (in the now familiar sense of the word) in which the nineteenth century was to rediscover, after this pure tabulation of things, the renewed possibility of talking about words. And of talking about them, not in the style of commentary, but in a mode that was to be considered as positive, as objective, as that of natural history.

131

THE ORDER OF THINGS

The ever more complete preservation of what was written, the estab­lishment of archives, then of filing systems for them, the reorganization of libraries, the drawing up of catalogues, indexes, and inventories, all these things represent, at the end of the Classical age, not so much a new sensitivity to time, to its past, to the density of history, as a way of intro­ducing into the language already imprinted on things, and into the traces it has left, an order of the same type as that which was being established between living creatures. And it is in this classified time, in this squared and spatialized development, that the historians of the nineteenth century were to undertake the creation of a history that could at last be 'true' - in other words, liberated from Classical rationality, from its ordering and theodicy:a history restored to the irruptive violence of time.
III  Structure

Thus arranged and understood, natural history has as a condition of its possibility the common affinity of things and language with representa­tion; but it exists as a task only in so far as things and language happen to be separate. It must therefore reduce this distance between them so as to bring language as close as possible to the observing gaze, and the things observed as close as possible to words. Natural history is nothing more than the nomination of the visible. Hence its apparent simplicity, and that air of naivete it has from a distance, so simple does it appear and so obviously imposed by things themselves. One has the impression that with Toumefort, with Linnaeus or Buffon, someone has at last taken on the task of stating something that had been visible from the beginning of time, but had remained mute before a sort of invincible distraction of men's eyes. In fact, it was not an age-old inattentiveness being suddenly dissipated, but a new field of visibility being constituted in all its density.

Natural history did not become possible because men looked harder and more closely. One might say, strictly speaking, that the Classical age used its ingenuity, if not to see as little as possible, at least to restrict deliberately the area of its experience. Observation, from the seventeenth century onward, is a perceptible knowledge furnished with a series of systematically negative conditions. Hearsay is excluded, that goes without saying; but so are taste and smell, because their lack of certainty and their variability render impossible any analysis into distinct elements that could be universally acceptable. The sense of touch is very narrowly limited to the designation of a few fairly evident distinctions (such as that

132

CLASSIFYING

between smooth and rough); which leaves sight with an almost exclusive privilege, being the sense by which we perceive extent and establish proof, and, in consequence, the means to an analysis partes extra partes acceptable to everyone: the blind man in the eighteenth century can per­fectly well be a geometrician, but he cannot be a naturalist [3]. And, even then, everything that presents itself to our gaze is not utilizable: colours especially can scarcely serve as a foundation for useful comparisons. The area of visibility in which observation is able to assume its powers is thus only what is left after these exclusions: a visibility freed from all other sensory burdens and restricted, moreover, to black and white. This area, much more than the receptivity and attention at last being granted to things themselves, defines natural history's condition of possibility, and the appearance of its screened objects: lines, surfaces, forms, reliefs.

It may perhaps be claimed that the use of the microscope compensates for these restrictions; and that though sensory experience was being restricted in the direction of its more doubtful frontiers, it was neverthe­less being extended towards the new objects of a technically controlled form of observation. In fact, it was the same complex of negative con­ditions that limited the realm of experience and made the use of optical instruments possible. To attempt to improve one's power of observation by looking through a lens, one must renounce the attempt to achieve knowledge by means of the other senses or from hearsay. A change of scale in the visual sphere must have more value than the correlations between the various kinds of evidence that may be provided by one's impressions, one's reading, or learned compilations. Though indefinite confinement of the visible within its own extent is made more easily perceptible to the eye by a microscope, it is nevertheless not freed from it. And the best proof of this is probably that optical instruments were used above all as a means of resolving problems of generation. In other words, as a means of discovering how the forms, arrangements, and characteristic proportions of individual adults, and of their species, could be handed on down the centuries while preserving their strictly defined identity. The microscope was called upon not to go beyond the frontiers of the fundamental domain of visibility, but to resolve one of the problems it posed: the maintenance of specific visible forms from generation to generation. The use of the microscope was based upon a non-instru­mental relation between things and the human eye - a relation that defines natural history. It was Linnaeus, after all, who said that Naturalia - as opposed to Coelestia and Elementa - were intended to be transmitted

133

THE ORDER OF THINGS

directly to the senses [4]. And Tournefort thought that, in order to gain a knowledge of plants, 'rather than scrutinize each of their variations with a religious scruple', it was better to analyse them 'as they fall beneath the gaze'[5].

To observe, then, is to be content with seeing - with seeing a few things systematically. With seeing what, in the rather confused wealth of representation, can be analysed, recognized by all, and thus given a name that everyone will be able to understand: 'All obscure similitudes,' said Linnaeus, 'are introduced only to the shame of art' [6]. Displayed in them­selves, emptied of all resemblances, cleansed even of their colours, visual representations will now at last be able to provide natural history with what constitutes its proper object, with precisely what it will convey in the well-made language it intends to construct. This object is the extension of which all natural beings are constituted - an extension that may be affected by four variables. And by four variables only: the form of the elements, the quantity of those elements, the manner in which they are distributed in space in relation to each other, and the relative magnitude of each element. As Linnaeus said, in a passage of capital importance, 'every note should be a product of number, of form, of proportion, of situation'[7]. For example, when one studies the reproductive organs of a plant, it is sufficient, but indispensable, to enumerate the stamens and pistil (or to record their absence, according to the case), to define the form they assume, according to what geometrical figure they are dis­tributed in the flower (circle, hexagon, triangle), and what their size is in relation to the other organs. These four variables, which can be applied in the same way to the five parts of the plant - roots, stem, leaves, flowers, fruits - specify the extension available to representation well enough for us to articulate it into a description acceptable to everyone: confronted with the same individual entity, everyone will be able to give the same description; and, inversely, given such a description everyone will be able to recognize the individual entities that correspond to it. In this fundamental articulation of the visible, the first confrontation of language and things can now be established in a manner that excludes all uncer­tainty.

Each visibly distinct part of a plant or an animal is thus describable in so far as four series of values are applicable to it. These four values affect­ing, and determining, any given element or organ are what botanists term its structure. 'By the structure of a plant's parts we mean the com­position and arrangement of the pieces that make up its body.' [8] Struc-

134

CLASSIFYING

ture also makes possible the description of what one sees, and this in two ways which are neither contradictory nor mutually exclusive. Number and magnitude can always be assigned by means of a count or a measure; they can therefore be expressed in quantitative terms. Forms and arrange­ments, on the other hand, must be described by other methods: either by identification with geometrical figures, or by analogies that must all be 'of the utmost clarity'[9]. In this way it becomes possible to describe certain fairly complex forms on the basis of their very visible resemblance to the human body, which serves as a sort of reservoir for models of visibility, and acts as a spontaneous link between what one can see and what one can say[l0].

By limiting and filtering the visible, structure enables it to be tran­scribed into language. It permits the visibility of the animal or plant to pass over in its entirety into the discourse that receives it. And ultimately, perhaps, it may manage to reconstitute itself in visible form by means of words, as with the botanical calligrams dreamed of by Linnaeus[11]. His wish was that the order of the description, its division into paragraphs, and even its typographical modules, should reproduce the form of the plant itself. That the printed text, in its variables of form, arrangement, and quantity, should have a vegetable structure. 'It is beautiful to follow nature: to pass from the Root to the Stems, to the Petioles, to the Leaves, to the Peduncles, to the Flowers.' The description would have to be divided into the same number of paragraphs as there are parts in the plant, everything concerning its principal parts being printed in large type, and the analysis of the 'parts of parts' being conveyed in small type. One would then add what one knew of the plant from other sources in the same way as an artist completes his sketch by introducing the interplay of light and shade: 'the Adumbration would exactly contain the whole history of the plant, such as its names, its structure, its external assemblage, its nature, its use.' The plant is thus engraved in the material of the language into which it has been transposed, and recomposes its pure form before the reader's very eyes. The book becomes the herbarium of living structures. And let no one reply that this is merely the reverie of a systematizer and does not represent the whole of natural history. Buffon was a constant adversary of Linnaeus, yet the same structure exists in his work and plays the same role: 'The method of examination will be directed towards form, magnitude, the different parts, their number, their position, and the very substance of the thing'[12]. Buffon and Linnaeus employ the same grid; their gaze occupies the same surface of contact upon things;

135

THE ORDER OF THINGS

there arc the same black squares left to accommodate the invisible; the same open and distinct spaces to accommodate words.

By means of structure, what representation provides in a confused and simultaneous form is analysed and thereby rendered suitable to the linear unwinding of language. In effect, description is to the object one looks at what the proposition is to the representation it expresses: its arrangement in a series, elements succeeding elements. But it will be remembered that language in its empirical form implied a theory of the proposition and a theory of articulation. In itself, the proposition remained empty;

and the ability of articulation to give form to authentic discourse was conditional upon its being linked together by the patent or secret function of the verb to be. Natural history is a science, that is, a language, but a securcly based and well-constructed one: its propositional unfolding is indisputably an articulation; the arrangement of its elements into a linear series patterns representation according to an evident and universal mode. Whereas one and the same representation can give rise to a con­siderable number of propositions, since the names that embody it articu­late it according to different modes, one and the same animal, or one and the same plant, will be described in the same way, in so far as their struc­ture governs their passage from representation into language. The theory of structure, which runs right through natural history in the Classical age, superimposes the roles played in language by the proposition and articula­tion in such a way that they perform one and the same function.

And it is by this means that structure links the possibility of a natural history to the mathesis. In fact, it reduces the whole area of the visible to a system of variables all of whose values can be designated, if not by a quantity, at least by a perfectly clear and always finite description. It is therefore possible to establish the system of identities and the order of differences existing between natural entities. Adanson was of the opinion that one day it would be possible to treat botany as a rigorously mathe­matical science, and that it would prove permissible to pose botanical problems in the same way as one does algebraic or geometrical ones: 'find the most obvious point that establishes the line of separation or discussion between the scabious family and the honeysuckle family'; or again, find a known genus of plants (whether natural or artificial is unimportant) that stands exactly half-way between Dog's-bane and Borage[13]. By virtue of structure, the great proliferation of beings occupying the surface of the globe is able to enter both into the sequence of a descriptive lan­guage and into the field of a mathesis that would also be a general science 136

CLASSIFYING

of order. And this constituent relation, complex as it is, is established within the apparent simplicity of a description of the visible.

All this is of great importance for the definition of natural history in terms of its object. The latter is provided by surfaces and lines, not by functions or invisible tissues. The plant and the animal are seen not so much in their organic unity as by the visible patterning of their organs. They are paws and hoofs, flowers and fruits, before being respiratory systems or internal liquids. Natural history traverses an area of visible, simultaneous, concomitant variables, without any internal relation of subordination or organization. In the seventeenth and eighteenth cen­turies anatomy lost the leading role that it had played during the Renais­sance and that it was to resume in Cuvier's day; it was not that curiosity had diminished in the meantime, or that knowledge had regressed, but rather that the fundamental arrangement of the visible and the expressible no longer passed through the thickness of the body. Hence the epistemological precedence enjoyed by botany: the area common to words and things constituted a much more accommodating, a much less 'black' grid for plants than for animals; in so far as there are a great many constituent organs visible in a plant that are not so in animals, taxonomic knowledge based upon immediately perceptible variables was richer and more coher­ent in the botanical order than in the zoological. We must therefore reverse what is usually said on this subject: it is not because there was a great interest in botany during the seventeenth and eighteenth centuries that so much investigation was undertaken into methods of classification. But because it was possible to know and to say only within a taxonomic area of visibility, the knowledge of plants was bound to prove more extensive than that of animals.

At the institutional level, the inevitable correlatives of this patterning were botanical gardens and natural history collections. And their import­ance, for Classical culture, does not lie essentially in what they make it possible to see, but in what they hide and in what, by this process of obliteration, they allow to emerge: they screen off anatomy and function, they conceal the organism, in order to raise up before the eyes of those who await the truth the visible relief of forms, with their elements, their mode of distribution, and their measurements. They are books furnished with structures, the space in which characteristics combine, and in which classifications are physically displayed. One day, towards the end of the eighteenth century, Cuvier was to topple the glass jars of the Museum, smash them open and dissect all the forms of animal visibility that the

137

THE ORDER OF THINGS

Classical age had preserved in them. This iconoclastic gesture, which Lamarck could never bring himself to make, does not reveal a new curiosity directed towards a secret that no one had the interest or courage to uncover, or the possibility of uncovering, before. It is rather, and much more seriously, a mutation in the natural dimension of Western culture:

the end of history in the sense in which it was understood by Tournefort, Linnaeus, Buffon, and Adanson - and in the sense in which it was under­stood by Boissier de Sauvages also, when he opposed historical knowledge of the visible to philosophical knowledge of the invisible, of what is hidden and of causes [14]. And it was also to be the beginning of what, by substituting anatomy for classification, organism for structure, internal subordination for visible character, the series for tabulation, was to make possible the precipitation into the old flat world of animals and plants, engraved in black on white, a whole profound mass of time to which men were to give the renewed name of history.
IV CHARACTER

Structure is that designation of the visible which, by means of a kind of pre-linguistic sifting, enables it to be transcribed into language. But the description thus obtained is nothing more than a sort of proper noun: it leaves each being its strict individuality and expresses neither the table to which it belongs, nor the area surrounding it, nor the site it occupies. It is designation pure and simple. And for natural history to become lan­guage, the description must become a 'common noun'. It has been seen how, in spontaneous language, the primary designations, which con­cerned only individual representations, after having originated in the lan­guage of action and the resultant primitive roots, had little by little, through the momentum of derivation, acquired more general values. But natural history is a well-constructed language: it should not accept the constraint imposed by derivation and its forms; it should not lend credit to any etymology[i5]. It should unite in one and the same operation what everyday language keeps separate: not only must it designate all natural entities very precisely, but it must also situate them within the system of identities and differences that unites them to and distinguishes them from all the others. Natural history must provide, simultaneously, a certain designation and a controlled derivation. And just as the theory of struc­ture superimposed articulation and the proposition so that they became one and the same, so the theory of character must identify the values

138

CLASSIFYING

that designate and the area in which they are derived. Toumefort says:

To know plants is to know with precision the names that have been given to them in relation to the structure of some of their parts. . . The idea of the character that essentially distinguishes plants from one another ought invariably to be one with the name of each plant[16].

Establishing character is at the same time easy and difficult. Easy, because natural history does not have to establish a system of names based upon representations that are difficult to analyse, but only to derive it from a language that has already been unfolded in the process of descrip­tion. The process of naming will be based, not upon what one sees, but upon elements that have already been introduced into discourse by struc­ture. It is a matter of constructing a secondary language based upon that primary, but certain and universal, language. But a major difficulty appears immediately. In order to establish the identities and differences existing between all natural entities, it would be necessary to take into account every feature that might have been listed in a given description. Such an endless task would push the advent of natural history back into an inac­cessible never-never land, unless there existed techniques that would avoid this difficulty and limit the labour of making so many comparisons. It is possible, a priori, to state that these techniques are of two types. Either that of making total comparisons, but only within empirically con­stituted groups in which the number of resemblances is manifestly so high that the enumeration of the differences will not take long to complete;

and in this way, step by step, the establishment of all identities and dis­tinctions can be guaranteed. Or that of selecting a finite and relatively limited group of characteristics, whose variations and constants may be studied in any individual entity that presents itself. This last procedure was termed the System, the first the Method. They are usually contrasted, in the same way as Linnaeus is contrasted with Buffon, Adanson, or Antoine-Laurent de Jussieu - or as a rigid and simple conception of nature is contrasted with the detailed and immediate perception of its relations, or as the idea of a motionless nature is contrasted with that of a teeming continuity of beings all communicating with one another, mingling with one another, and perhaps being transformed into one another.. .. And yet the essential does not lie in this conflict between the great intuitions of nature. It lies rather in the network of necessity which at this point rendered the choice between two ways of constituting

139

THE ORDER OF THINGS

natural history as a language both possible and indispensable. The rest is merely a logical and inevitable consequence.

From the elements that the System juxtaposes in great detail by means of description, it selects a particular few. These define the privileged and, in fact, exclusive structure in relation to which identities or differences as a whole are to be examined. Any difference not related to one of these elements will be considered irrelevant. If, like Linnaeus, one selects as the characteristic elements 'all the different parts related to fructification'[17], then a difference of leaf or stem or root or petiole must be systematically ignored. Similarly, any identity not occurring in one of these selected elements will have no value in the definition of the character. On the other hand, when these elements are similar in two individuals they receive a common denomination. The structure selected to be the locus of pertinent identities and differences is what is termed the character. According to Linnaeus, the character should be composed of 'the most careful description of the fructification of the first species. All the other species of the genus are compared with the first, all discordant notes being eliminated; finally, after this process, the character emerges'[18].

The system is arbitrary in its basis, since it deliberately ignores all differences and all identities not related to the selected structure. But there is no law that says that it will not be possible to arrive one day, through a use of this technique, at the discovery of a natural system - one in which all the differences in the character would correspond to differences of the same value in the plant's general structure; and in which, inversely, all the individuals or all the species grouped together under a common character would in fact have the same relation of resemblance in all and each of their parts. But one cannot find the way to this natural system unless one has first established with certainty an artificial system, at least in certain of the vegetable or animal domains. This is why Linnaeus does not seek to establish a natural system immediately, 'before a complete knowledge has been attained of everything that is relevant'[19] to his system. It is true that the natural method constitutes 'the first and last wish of botanists', and that all its 'fragments should be searched for with the greatest care'[20], as Linnaeus himself searches for them in his Classes Plantarum; but until this natural method appears in its certain and finished form, 'artificial systems are absolutely necessary'[21].

Moreover, the system is relative: it is able to function according to a desired degree of precision. If the selected character is composed of a large structure, having a large number of variables, then as soon as one

140

CLASSIFYING

passes from one individual to another, even if it is immediately adjacent, the differences will appear at once: the character in this case is very close to pure description [22]. If, on the other hand, the selected structure is limited in extent, and its variables few, then the differences will be rare and the individuals grouped in compact masses. The character is chosen according to the degree of detail required in the classification. In order to establish genera, Tournefort chose the combination of flower and fruit as his character. Not, as with Cesalpino, because these were the most useful parts of the plant, but because they permitted a numerically satisfying combinability: the elements that would be taken from the other three parts (roots, stems, and leaves) were, in effect, either too numerous if treated together or too few if taken separately [23]. Linnaeus calculated that the thirty-eight organs of reproduction, each comprising the four variables of number, form, situation, and proportion, would produce 5,776 configurations, or sufficient to define the general]. If one wishes to obtain groups more numerous than genera, then one must make use of more limited characters ('factitious characters agreed upon between botanists'), as, for example, the stamens alone, or the pistil alone. In this way one would be able to distinguish classes or orders[25].

In this way, a grid can be laid out over the entire vegetable or animal kingdom. Each group can be given a name. With the result that any species, without having to be described, can be designated with the greatest accuracy by means of the names of the different groups in which it is included. Its complete name will cross the entire network of char­acters that one has established, right up to the largest classifications of all. But for convenience, as Linnaeus points out, part of this name should remain 'silent' (one does not name the class and order), while the other part should be 'sounded' (one must name the genus, the species, and the variety [26]. The plant thus recognized in its essential character and desig­nated upon that basis will express at the same time that which accurately designates it and the relation linking it to those plants that resemble it and belong to the same genus (and thus to the same family and the same order). It will have been given at the same time its proper name and the whole series of common names (manifest or hidden) in which it resides. 'The generic name is, as it were, the official currency of our botanical republic'[27]. Natural history will have accomplished its fundamental task, which is that of 'arrangement and designation'[28].

The Method is another technique for resolving the same problem. Instead of selecting, from the totality described, the elements - whether

141

THE ORDER OF THINGS

few or numerous - that are to be used as characters, the method consists in deducing them stage by stage. Deduction is to be taken here in the sense of subtraction. One begins - as Adanson did in his examination of the plants of Senegal [29]-with a species either arbitrarily chosen or encountered by chance. One describes it in its entirety, leaving out none of its parts and determining all the values that the variables have derived from it. This process is repeated with the next species, also given by the arbitrary nature of representation; the description should be as total as in the first instance, but with the one difference that nothing that has been mentioned in the first description should be repeated in the second. Only the differences are listed. And similarly with the third species in relation to the first two, and so on indefinitely. So that, at the very end, all the different features of all the plants have been listed once, but never more than once. And by arranging the later and progressively more sparse des­criptions around the earlier ones, we shall be able to perceive, through the original chaos, the emergence of the general table of relations. The char­acter that distinguishes each species or each genus is the only feature picked out from the background of tacit identities. Indeed, such a tech­nique would probably be the most reliable, only the number of existing species is so great that it would be impossible to deal with them all. Nevertheless, the examination of such species as we do meet with reveals the existence of great 'families', of very broad groups in which the species and the genera have a considerable number of identities. So considerable, indeed, that they signalize themselves by a very large number of char­acteristics, even to the least analytic eye; the resemblance between all the species of Ranunculus, or between all the species of Aconite, is immedi­ately apparent to the senses. At this point, in order to prevent the task becoming infinite, one is obliged to reverse the process. One admits the existence of the great families that are manifestly recognizable, and whose general features have been defined, as it were blindfold, by the first descriptions of them. These are the common features that we now estab­lish in a positive way; then, whenever we meet with a genus or species that is manifestly contained by them, it will suffice to indicate what difference distinguishes it from the others that serve it as a sort of natural entourage. A knowledge of each species can be acquired easily upon the basis of this general characterization: 'We shall divide each of the three kingdoms into several families which will group together all those beings that are strikingly related, and we shall review all the general and par­ticular characters of the beings contained within those families'; in this way

142

CLASSIFYING

we shall be assured of relating all these beings to their natural families; and thus, beginning with the ferret and the wolf, the dog and the bear, we shall come to know sufficient about the lion, the tiger, and the hyena, which are animals of the same family [30].

It is immediately apparent in what way the method and the system are opposed. There can be only one method; but one can invent and apply a considerable number of systems: Adanson alone set out sixty-five [31]. The system is arbitrary throughout its development, but once the system of variables - the character - has been defined at the outset, it is no longer possible to modify it, to add or subtract even one element. The method is imposed from without, by the total resemblances that relate things together; it immediately transcribes perception into discourse; it remains, in its point of departure, very close to description; but it is always possible to apply to the general character it has defined empirically such modifica­tions as may be imposed: a feature one had thought essential to a whole group of plants or animals may very well prove to be no more than a particularity of a few of them, if one discovers others that, without possessing that feature, belong quite obviously to the same family; the method must always be ready to rectify itself. As Adanson says, the system is like 'the trial and error method in mathematics': it is the result of a decision, but it must be absolutely coherent; the method, on the other hand, is

a given arrangement of objects or facts grouped together according to certain given conventions or resemblances, which one expresses by a general notion applicable to all those objects, without, however, regarding that fundamental notion or principle as absolute or invariable, or as so general that it cannot suffer any exception . . . The method differs from the system only in the idea that the author attaches to his principles, regarding them as variables in the method and as absolutes in the system [32].

Moreover, the system can recognize only relations of coordination between animal or vegetable structures. Since the character is selected, not on account of its functional importance but on account of its com­binative efficacity, there is no proof that in the internal hierarchy of any individual plant such and such a form of pistil or arrangement of stamens necessarily entails such and such a structure: if the germ of the Adoxa is placed between the calyx and the corolla, or if, in the arum, the stamens 143

THE ORDER OF THINGS

are arranged between the pistils, these are nothing more or less than 'singular Structures'[33]; their slight importance is a product of their rarity alone, whereas the equal division of calyx and corolla derives its value only from its frequency[34]. The method, on the other hand, because it proceeds from identities and differences of the most general kind to those that are less so, is capable of bringing out vertical relations of subordination. It enables us, in fact, to see which characters are important enough never to be negated within a given family. In relation to the system, the reversal is very important: the most essential characters make it possible to distinguish the largest and most visibly distinct families, whereas, for Toumefort or Linnaeus, the essential character defined the genus; and it was sufficient for the naturalists' 'agreement' to select a factitious character that would distinguish between classes or orders. In the method, general organization and its internal dependencies are more important than the lateral application of a constant apparatus of variables.

Despite these differences, both system and method rest upon the same epistemological base. It can be defined briefly by saying that, in Classical terms, a knowledge of empirical individuals can be acquired only from the continuous, ordered, and universal tabulation of all possible differ­ences. In the sixteenth century, the identity of plants or animals was assured by the positive mark (sometimes hidden, often visible) which they all bore: what distinguished the various species of birds, for instance, was not the differences that existed between them but the fact that this one hunted its food at night, that another lived on the water, that yet another fed on living flesh [35]. Every being bore a mark, and the species was measured by the extent of a common emblem. So that each species identified itself by itself, expressed its individuality independently of all the others: it would have been perfectly possible for all those others not to exist, since the criteria of definition would not thereby have been modified for those that remained visible. But, from the seventeenth cen­tury, there can no longer be any signs except in the analysis of representa­tions according to identities and differences. That is, all designation must be accomplished by means of a certain relation to all other possible designations. To know what properly appertains to one individual is to have before one the classification - or the possibility of classifying - all others. Identity and what marks it are defined by the differences that remain. An animal or a plant is not what is indicated - or betrayed - by the stigma that is to be found imprinted upon it; it is what the others are not; it exists in itself only in so far as it is bounded by what is distinguish-

144

CLASSIFYING

able from it. Method and system are simply two ways of defining iden­tities by means of the general grid of differences. Later on, beginning with Cuvier, the identity of species was to be determined in the same way by a set of differences, but the differences were in this case to emerge from the background of the great organic unities possessing their own internal systems of dependencies (skeleton, respiration, circulation); the invertebrates were to be defined, not only by their lack of vertebrae, but also by a certain mode of respiration, by the existence of a type of circula­tion, and by a whole organic cohesiveness outlining a positive unity. The internal laws of the organism were to replace differential characters as the object of the natural sciences. Classification, as a fundamental and constituent problem of natural history, took up its position historically, and in a necessary fashion, between a theory of the mark and a theory of the organism.
V CONTINUITY AND CATASTROPHE

At the heart of this well-constructed language that natural history has become, one problem remains. It is possible after all that the transforma­tion of structure into character may never be possible, and that the common noun may never be able to emerge from the proper noun. Who can guarantee that the descriptions, once made, are not going to display elements that vary so much from one individual to the next, or from one species to the next, that any attempt to use them as the basis for a common noun would be doomed in advance? Who can be certain that each structure is not strictly isolated from every other structure, and that it will not function as an individual mark? In order that the simplest character can become apparent, it is essential that at least one element in the structure examined first should be repeated in another. For the general order of differences that makes it possible to establish the arrangement of species implies a certain number of similarities. The problem here is isomorphic with the one we have already met in relation to language[36]: for a common noun to be possible, there had to be an immediate resem­blance between things that permitted the signifying elements to move along the representations, to slide across the surface of them, to cling to their similarities and thus, finally, to form collective designations. But in order to outline this rhetorical space in which nouns gradually took on their general value, there was no need to determine the status of that resemblance, or whether it was founded upon truth; it was sufficient for

145

THE ORDER OF THINGS

it to strike the imagination with sufficient force. In natural history, however, which is a well-constructed language, these analogies of the imagination cannot have the value of guarantees; and since natural history is threatened, like all language, by the radical doubt that Hume brought to bear upon the necessity for repetition in experience, it must find a way of avoiding that threat. There must be continuity in nature.

This requirement that nature should be continuous does not take exactly the same form in the systems as it does in the methods. For the systematician, continuity consists only of the unbroken juxtaposition of the different regions that can be clearly distinguished by means of char­acters; all that is required is an uninterrupted gradation of the values that the structure selected as a character can assume in the species as a whole;

starting from this principle, it will become apparent that all these values are occupied by real beings, even though they may not yet be known. 'The system indicates the plants, even those it has not mentioned; which is something that the enumeration of a catalogue can never do'[37]. And the categories will not simply be arbitrary conventions laid out over this continuity of juxtaposition; they will correspond (if they have been properly established) to areas that have a distinct existence on this uninter­rupted surface of nature; they will be areas that are larger than individuals but just as real. In this way, according to Linnaeus, the reproductive system made it possible to establish the existence of indisputably well-founded genera: 'Know that it is not the character that constitutes the genus, but the genus that constitutes the character, that the character derives from the genus, not the genus from the character'[38]. In the methods, on the other hand, since resemblances - in their massive and clearly evident form - are posited to start with, the continuity of nature will not be this purely negative postulate (no blank spaces between dis­tinct categories), but a positive requirement: all nature forms one great fabric in which beings resemble one another from one to the next, in which adjacent individuals are infinitely similar to each other; so that any dividing-line that indicates, not the minute difference of the individual, but broader categories, is always unreal. There is a continuity produced by fusion in which all generality is nominal. Our general ideas, says Buffon,

are relative to a continuous scale of objects of which we can clearly perceive only the middle rungs and whose extremities increasingly flee from and escape our considerations. . . The more we increase the

146

CLASSIFYING

number of divisions in the productions of nature, the closer we shall approach to the true, since nothing really exists in nature except indi­viduals, and since genera, orders, and classes exist only in our imagina­tion [39].

And Bonnet, meaning much the same thing, said:

There are no leaps in nature: everything in it is graduated, shaded. If there were an empty space between any two beings, what reason would there be for proceeding from the one to the other? There is thus no being above and below which there are not other beings that are united to it by some characters and separated from it by others.

It is therefore always possible to discover 'intermediate productions', such as the polyp between the animal and the vegetable, the flying squirrel between the bird and the quadruped, the monkey between the quadruped and man. Consequently, our divisions into species and classes 'are purely nominal'; they represent no more than 'means relative to our needs and to the limitations of our knowledge'[40].

In the eighteenth century, the continuity of nature is a requirement of all natural history, that is, of any effort to establish an order in nature and to discover general categories within it, whether they be real and pre­scribed by obvious distinctions or a matter of convenience and quite simply a pattern produced by our imagination. Only continuity can guarantee that nature repeats itself and that structure can, in consequence, become character. But this requirement immediately becomes a double one. For if it were given to experience, in its uninterrupted momentum, to traverse exactly, step by step, the great continuity comprising indi­viduals, varieties, species, genera, and classes, there would be no need to constitute a science; descriptive designations would attain to generality quite freely, and the language of things would be constituted as scientific discourse by its own spontaneous momentum. The identities of nature would be presented to the imagination as though spelled out letter by letter, and the spontaneous shift of words within their rhetorical space would reproduce, with perfect exactitude, the identity of beings with their increasing generality. Natural history would become useless, or rather it would already have been written by man's everyday language;

general grammar would at the same time be the universal taxonomy of beings. But if a natural history perfectly distinct from the analysis of words is indispensable, that is because experience does not reveal the 147

continuity of nature as such, but gives it to us both broken up - since there are a great many gaps in the series of values effectively occupied by the variables (there are possible creatures whose place in the grid one can note without ever having had the opportunity to observe them) - and blurred, since the real, geographic and terrestrial space in which we find ourselves confronts us with creatures that are interwoven with one another, in an order which, in relation to the great network of taxonomies, is nothing more than chance, disorder, or turbulence. Linnaeus pointed out that, by associating the hydra (which is an animal) and the conferva (which is an alga), or the sponge and the coral, in the same localities, nature is not, as the order of our classifications would have it, linking together 'the most perfect plants with the animals termed very imperfect, but combining imperfect animals with imperfect plants'[41]. And Adanson remarked that nature is

a confused mingling of beings that seem to have been brought together by chance: here, gold is mixed with another metal, with stone, with earth; there, the violet grows side by side with an oak. Among these plants, too, wander the quadruped, the reptile, and the insect; the fishes are confused, one might say, with the aqueous element in which they swim, and with the plants that grow in the depths of the waters . . . This mixture is indeed so general and so multifarious that it appears to be one of nature's laws [42].

Now, this great mixture is the result of a chronological series of events. And these events have their point of origin and their primary locus of application, not in the living species themselves, but in the space in which those species reside. They are produced in the relation of the Earth to the Sun, in climatic conditions, in the movements of the earth's crust; what they affect first are the oceans and the continents, the surface of the globe;

living beings are affected only indirectly and in a secondary way: they are attracted or driven away by heat; volcanoes destroy them; they disappear with the land that crumbles away beneath them. It is possible, as Buffon, for example, supposed [43], that the earth was originally incandescent, before gradually growing colder; the animals, accustomed to living in very high temperatures, then regrouped themselves in the only region that still remains torrid, whereas the temperate or cold lands were peopled by species that had not had the opportunity to appear until that time. With the revolutions in the history of the earth, the taxonomic area (in which adjacencies are of the order of character and not of modus vivendi) was

148

CLASSIFYING

divided up into a concrete and geographical area that jumbled it all up. Moreover, it was probably broken up into fragments, and many species, adjacent to those we know or intermediary between taxonomic squares familiar to us, must have disappeared, leaving nothing behind them but traces difficult to decipher. In any case, this historical series of events is an addition to the expanse of beings: it does not properly appertain to it; its development lies in the real dimension of the world, not the analytic one of classifications; what it calls into question is the world as a locus for beings, not the beings themselves in so far as they have the property of being alive. There is a historicity, symbolized by the biblical accounts, which affects our astronomic system directly and the taxonomic grid of species indirectly; and apart from Genesis and the Flood, it is very possible that

our globe underwent other revolutions that have not been revealed to us. It is connected to the whole astronomic system, and the links that join this globe to the other celestial bodies, in particular to the Sun and the comets, could have been the source of many revolutions that have left no traces perceptible to us, but of which the inhabitants of neigh­bouring worlds may perhaps have some knowledge[44].

To be able to exist as a science, natural history must, then, presuppose two groupings. One of them is constituted by the continuous network of beings; this continuity may take various spatial forms; Charles Bonnet thinks of it sometimes as a great linear scale of which one extremity is very simple, the other very complicated, with a narrow intermediary region - the only one that is visible to us - in the centre; sometimes as a central trunk from which there is a branch forking out on one side (that of the shellfish, with the crabs and crayfish as supplementary ramifica­tions) and the series of insects on the other, branching out to include the frogs [45]; Buffon defines this same continuity 'as a wide woven strip, or rather a bundle which every so often puts out side branches that join it up with the bundles of another order' [46]; Pallas sees it as a polyhedric figure [47]; Hermann wished to constitute a three-dimensional model composed of threads all starting from a common point of origin, separa­ting from one another, 'spreading out through a very great number of lateral branches', then coming together again[48]. The series of events, however, is quite distinct from these spatial configurations, each of which describes the taxonomic continuity in its own way; the series of events is discontinuous, and different in each of its episodes; but, as a whole, it can

149

THE ORDER OF THINGS

be drawn only as a simple line, which is that of time itself (and which can be conceived as straight, broken, or circular). In its concrete form, and in the depth that is proper to it, nature resides wholly between the fabric of the taxinomia and the line of revolutions. The tabulations that it forms in the eyes of men, and that it is the task of the discourse of science to tra­verse, are the fragments of the great surface of living species that are apparent according to the way it has been patterned, burst open, and frozen, between two temporal revolutions.

It will be seen how superficial it is to oppose, as two different opinions confronting one another in their fundamental options, a 'fixism' that is content to classify the beings of nature in a permanent tabulation, and a sort of 'evolutionism' that is supposed to believe in an immemorial his­tory of nature and in a deep-rooted, onward urge of all beings throughout its continuity. The solidity, without gaps, of a network of species and genera, and the series of events that have blurred that network, both belong, at the same level, to the epistemological foundation that made a body of knowledge like natural history possible in the Classical age. They are not two ways of perceiving nature, radically opposed because deeply rooted in philosophical choices older and more fundamental than any science; they are two simultaneous requirements in the archaeological network that defines the knowledge of nature in the Classical age. But these two requirements are complementary, and therefore irreducible. The temporal series cannot be integrated into the gradation of beings. The eras of nature do not prescribe the internal time of beings and their continuity; they dictate the intemperate interruptions that have constantly dispersed them, destroyed them, mingled them, separated them, and interwoven them. There is not and cannot be even the suspicion of an evolutionism or a transformism in Classical thought; for time is never conceived as a principle of development for living beings in their internal organization; it is perceived only as the possible bearer of a revolution in the external space in which they live.
VI MONSTERS AND FOSSILS

It will be objected that, long before Lamarck, there already existed a whole body of thought of the evolutionist type. That its importance was considerable in the middle of the eighteenth century, and up to the sudden halt marked by the work of Cuvier. That Bonnet, Maupertuis, Diderot, Robinet, and Benoit de Maillet all very clearly articulated the idea that

150

CLASSIFYING

living forms may pass from one into another, that the present species are no doubt the result of former transformations, and that the whole of the living world is perhaps in motion towards a future point, so that one cannot guarantee of any living form that it has been definitively acquired and is now stabilized forever. In fact, such analyses are incompatible with what we understand today by evolutionary thought. They are concerned, in fact, with linking the table of identities and differences to the series of successive events. And in order to conceive of the unity of that table and that series they have only two means at their disposal.

The first consists in integrating the series of successions with the con­tinuity of the beings and their distribution over the table. All the creatures that taxonomy has arranged in an uninterrupted simultaneity are then subjected to time. Not in the sense that the temporal series would give rise to a multiplicity of species that a horizontally oriented eye could then arrange according to the requirements of a classifying grid, but in the sense that all the points of the taxonomy are affected by a temporal index, with the result that 'evolution' is nothing more than the interdependent and general displacement of the whole scale from the first of its elements to the last. This system is that of Charles Bonnet. He implies in the first place that the chain of being, stretching up through an innumerable series of links towards the perfection of God, does not at present attain to it [49]; that the distance between God and the least defective of his creatures is still infinite; and that across this, perhaps unbridgeable, distance the whole uninterrupted fabric of beings is ceaselessly advancing towards a greater perfection. He implies further that this 'evolution' keeps intact the relation that exists between the different species: if one of them, in the process of perfecting itself, should attain the degree of complexity pos­sessed beforehand by the species one step higher, this does not mean that the latter has thereby been overtaken, because, carried onward by the same momentum, it cannot avoid perfecting itself to an equivalent degree:

There will be a continual and more or less slow progress of all the species towards a superior perfection, with the result that all the degrees of the scale will be continually variable within a determined and con­stant relation . . . Man, once transported to an abode more suited to the eminence of his faculties, will leave to the monkey and the elephant that foremost place that he occupied before among the animals of our planet. . . There will be Newtons among the monkeys and Vaubans

151

THE ORDER OF THINGS

among the beavers. The oysters and the polyps will stand in the same relation to the species at the top of the scale as the birds and the quad­rupeds do now to man [50].

This 'evolutionism' is not a way of conceiving of the emergence of beings as a process of one giving rise to another; in reality, it is a way of general­izing the principle of continuity and the law that requires that all beings form an uninterrupted expanse. It adds, in a Leibnizian style [51], the continuity of time to the continuity of space, and the infiniteness of the progress of beings towards perfection to their infinite multiplicity. It is not a matter of progressive hierarchization, but of the constant and total force exerted by an already established hierarchy. In the end this pre­supposes that time, far from being a principle of taxinomia, is merely one of its factors, and that it is pre-established, like all the other values assumed by all the other variables. Bonnet must, therefore, be a preformationist -and as far removed as possible from what we understand, since the nine­teenth century, by 'evolutionism'; he must suppose that the upheavals or catastrophes of the globe were arranged in advance as so many oppor­tunities for the infinite chain of being to continue its progress in the direction of infinite amelioration: 'These evolutions were foreseen and inscribed in the germs of animals upon the very first day of creation. For these evolutions are linked with revolutions in the whole solar system that were arranged by God in advance.' The universe in its entirety has been a larva; now it is a chrysalis; one day it will, no doubt, become a butterfly[52]. And every species will be caught up in the same way in that great mutation. Such a system, it is clear, is not an evolutionism beginning to overthrow the old dogma of fixism; it is a taxinomia that includes time in addition - a generalized classification.

The other form of'evolutionism' consists of giving time a completely opposite role to play. It is used no longer to move the classifying table as a whole along the finite or infinite line leading to perfection, but to reveal, one after the other, the squares that, when viewed together, will form the continuous network of the species. It causes the variables of the living world to assume all possible values successively: it is the immediacy of a characterization that is accomplished little by little and, as it were, element after element. The partial identities or resemblances that make a taxinomia possible would then be the marks, revealed in the present, of one and the same living being, persisting through all the upheavals of nature and thereby filling all the vacant possibilities offered by the 152

CLASSIFYING

taxonomic table. If birds have wings in the way that fishes have fins, Benolt de Maillet points out, it is because they were once, at the time when the original waters of the earth were ebbing, dehydrated giltheads or dolphins that passed over, once for all, into an aerial home.

The seed of these fishes, carried into swamps, may perhaps have pro­duced the first transmigration of the species from its marine to its ter­restrial home. Even though a hundred millions may have perished without having been able to grow accustomed to it, it was sufficient for two of them to arrive at that point to give rise to the species [53].

Changes in the conditions of life of living beings seem here, as in certain forms of evolutionism, to be the necessary cause of the appearance of new species. But the mode in which the air, the water, the climate, or the earth acts upon animals is not that of an environment upon a function and upon the organs in which that function takes place; here, the exterior elements intervene only in so far as they occasion the emergence of a character. And that emergence, though it may be chronologically deter­mined by such and such a global event, is rendered possible a priori by the general table of variables that defines all the possible forms of the living world. The quasi-evolutionism of the eighteenth century seems to presage equally well the spontaneous variation of character, as it was later to be found in Darwin, and the positive action of the environment, as it was to be described by Lamarck. But this is an illusion of hindsight:

for this form of thought, in fact, the sequence of time can never be any­thing but the line along which all the possible values of the pre-established variables succeed one another. Consequently, a principle of modification must be defined within the living being, enabling it to take on a new character when a natural revolution occurs.

We are presented, then, with another choice: either to presuppose a spontaneous aptitude in living beings to change their forms (or at least to acquire - with succeeding generations - a slightly different character from that originally given, so that it will change gradually from one to the next and finally become unrecognizable), or to attribute to them some obscure urge towards a terminal species that will possess the char­acters of all those that have preceded it, but in a higher degree of com­plexity and perfection.

The first system is that of errors to infinity - as it is to be in Maupertuis. According to this system, the table of species that it is possible for natural history to establish has been built up piecemeal by the balance, constantly 153

THE ORDER OF THINGS

present in nature, between a memory that guarantees its continuity (main­tenance of the species in time and their resemblance to one another) and a tendency towards deviation that simultaneously guarantees the existence of history, differences, and dispersion. Maupertuis supposes that the par­ticles of matter are endowed with activity and memory. When attracted to one another, the least active form mineral substances; the most active form the more complex bodies of animals. These forms, which are the result of attraction and chance, disappear if they are unable to survive. Those that do remain in existence give rise to new individuals in which the characters of the parent couple are preserved by memory. And this process continues until a deviation of the particles - a chance happening -brings into being a new species, which the stubborn force of memory maintains in existence in turn: 'By dint of repeated deviations, the infinite diversity of the animals came to pass* [5 4]. Thus, progressing from one to the next, living beings acquired by successive variations all the characters we now recognize in them, and, when one considers them in the dimen­sion of time, the coherent, solid expanse they form is merely the frag­mentary result of a much more tightly knit, much finer, continuity: a continuity that has been woven from an incalculable number of tiny, forgotten, or miscarried differences. The visible species that now present themselves for our analysis have been separated out from the ceaseless background of monstrosities that appear, glimmer, sink into the abyss, and occasionally survive. And this is the fundamental point: nature has a history only in so far as it is susceptible of continuity. It is because it takes on all possible characters in turn (each value of all the variables) that it is presented in the form of a succession.

The same can be said for the inverse system of the prototype and the terminal species. In this case it is necessary to suppose, with J-B. Robinet, that continuity is assured, not by memory, but by a project - the project of a complex being towards which nature makes its way from the starting-point of simple elements which it gradually combines and arranges: 'First of all, the elements combine. A small number of simple principles serves as a basis for all bodies'; these are the ones that govern exclusively the organization of minerals; then 'the magnificence of nature' continues to increase without a break 'up to the level of the beings that move upon the surface of the globe'; 'the variation of the organs in number, in size, in refinement, in internal texture, and in external form, produces species which are divided and subdivided to infinity by new arrangements'[55]. And so on, until we reach the most complex arrangement we know of. 154

CLASSIFYING

So that the entire continuity of nature resides between an absolutely archaic prototype, buried deeper than any history, and the extreme complication of this model as it is now possible to observe it, at least on this earthly globe, in the person of the human being [56]. Between these two extremes there lie all the possible degrees of complexity and com­bination - like an immense series of experiments, of which some have persisted in the form of continuing species and some have sunk into oblivion. Monsters are not of a different 'nature' from the species them­selves:

We should believe that the most apparently bizarre forms... belong necessarily and essentially to the universal plan of being; that they are metamorphoses of the prototype just as natural as the others, even though they present us with different phenomena; that they serve as means of passing to adjacent forms; that they prepare and bring about the combinations that follow them, just as they themselves were brought about by those that preceded them; that far from disturbing the order of things, they contribute to it. It is only, perhaps, by dint of producing monstrous beings that nature succeeds in producing beings of greater regularity and with a more symmetrical structure[57].

In Robinet, as in Maupertuis, succession and history are for nature merely means of traversing the infinite fabric of variations of which it is capable. It is not, then, that time or duration ensures the continuity and specifica­tion of living beings throughout the diversity of successive environments, but that against the continuous background of all the possible variations time traces out an itinerary upon which climates and geography pick out only certain privileged regions destined to survive. Continuity is not the visible wake of a fundamental history in which one same living principle struggles with a variable environment. For continuity precedes time. It is its condition. And history can play no more than a negative role in relation to it: it either picks out an entity and allows it to survive, or ignores it and allows it to disappear.

This has two consequences. First, the necessity of introducing monsters into the scheme - forming the background noise, as it were, the endless murmur of nature. Indeed, if it is necessary for time, which is limited, to run through-or perhaps to have already run through-the whole continuity of nature, one is forced to admit that a considerable number of possible variations have been encountered and then erased; just as the geological catastrophe was necessary to enable us to work back from the

155

THE ORDER OF THINGS

taxonomic table to the continuum, through a blurred, chaotic, and frag­mented experience, so the proliferation of monsters without a future is necessary to enable us to work down again from the continuum, through a temporal series, to the table. In other words, what must be construed, as we move in one direction, as a drama of the earth and waters must be construed, in the other direction, as an obvious aberration of forms. The monster ensures in time, and for our theoretical knowledge, a continuity that, for our everyday experience, floods, volcanoes, and subsiding continents confuse in space. The other consequence is that the signs of continuity throughout such a history can no longer be of any order other than that of resemblance. Since this history is not defined by any relation of organism to environment[58], the living forms will be subjected in it to all possible metamorphoses and leave behind them no trace of the path they have followed other than the reference points represented by similitudes. How, for example, are we to recognize that nature, starting from a primitive prototype, has never ceased to work towards the provisionally terminal form that is man? By the fact that it has abandoned on the way thousands of forms that provide us with a picture of the rudimentary model. How many fossils are there, for man's car, or skull, or sexual parts, like so many plaster statues, fashioned one day and dropped the next in favour of a more perfected form?

The species that resembles the human heart, and for that reason is named Anthropocardite ... is worthy of particular attention. Its sub­stance is flint inside. The form of a heart is imitated as perfectly as pos­sible. One can distinguish in it the stump of the vena cava, together with a portion of its two cross-sections. One can also see the stump of the great artery emerging from the left ventricle, together with its lower or descending branch [59].

The fossil, with its mixed animal and mineral nature, is the privileged locus of a resemblance required by the historian of the continuum, whereas the space of the taxinomia decomposed it with rigour.

The monster and the fossil both play a very precise role in this con­figuration. On the basis of the power of the continuum held by nature, the monster ensures the emergence of difference. This difference is still without law and without any well-defined structure; the monster is the root-stock of specification, but it is only a sub-species itself in the stub­bornly slow stream of history. The fossil is what permits resemblances to subsist throughout all the deviations traversed by nature; it functions as 156

CLASSIFYING

a distant and approximative form of identity; it marks a quasi-character in the shift of time. And this is because the monster and the fossil are merely the backward projection of those differences and those identities that provide taxinomia first with structure, then with character. Between table and continuum they form. a shady, mobile, wavering region in which what analysis is to define as identity is still only mute analogy; and what it will define as assignable and constant difference is still only free and random variation. But, in truth, it is so impossible for natural history to conceive of the history of nature, the epistemological arrangement delineated by the table and the continuum is so fundamental, that becom­ing can occupy nothing but an intermediary place measured out for it solely by the requirements of the whole. This is why it occurs only in order to bring about the necessary passage from one to the other - either as a totality of destructive events alien to living beings and occurring only from outside them, or as a movement ceaselessly being outlined, then halted as soon as sketched, and perceptible only on the fringes of the table, in its unconsidered margins. Thus, against the background of the continuum, the monster provides an account, as though in caricature, of the genesis of differences, and the fossil recalls, in the uncertainty of its resemblances, the first buddings of identity.

VII THE DISCOURSE OF NATURE

The theory of natural history cannot be dissociated from that of language. And yet it is not a question of a transference of method, from one to the other; nor of a communication of concepts; nor of the prestige of a model which, because it has 'succeeded' in one field, has been tried out in the one next to it. Nor is it a question of a more general rationality imposing identical forms upon grammatical thinking and upon taxinomia. Rather, it concerns a fundamental arrangement of knowledge, which orders the knowledge of beings so as to make it possible to represent them in a system of names. There were doubtless, in this region we now term life, many inquiries other than attempts at classification, many kinds of analysis other than that of identities and differences. But they all rested upon a sort of historical a priori, which authorized them in their dispersion and in their singular and divergent projects, and rendered equally possible all the differences of opinion of which they were the source. This a priori does not consist of a set of constant problems uninterruptedly presented to men's curiosity by concrete phenomena as so many enigmas; nor is it 157

THE ORDER OF THINGS

made up of a certain state of acquired knowledge laid down in the course of the preceding ages and providing a ground for the more or less irregu­lar, more or less rapid, progress of rationality; it is doubtless not even determined by what is called the mentality or the 'framework of thought' of any given period, if we are to understand by that the historical outline of the speculative interests, beliefs, or broad theoretical options of the time. This a priori is what, in a given period, delimits in the totality of experience a field of knowledge, defines the mode of being of the objects that appear in that field, provides man's everyday perception with theoretical powers, and defines the conditions in which he can sustain a discourse about things that is recognized to be true. In the eighteenth century, the historical a priori that provided the basis for inquiry into or controversy about the existence of genera, the stability of species, and the transmission of characters from generation to generation, was the exist­ence of a natural history: the organization of a certain visible existence as a domain of knowledge, the definition of the four variables of description, the constitution of an area of adjacencies in which any individual being whatever can find its place. Natural history in the Classical age is not merely the discovery of a new object of curiosity; it covers a series of complex operations that introduce the possibility of a constant order into a totality of representations. It constitutes a whole domain of empiricity as at the same time describable and orderahle. What makes it akin to theories of language also distinguishes it from what we have understood, since the nineteenth century, by biology, and causes it to play a certain critical role in Classical thought.

Natural history is contemporaneous with language: it is on the same level as the spontaneous play that analyses representations in the memory, determines their common elements, establishes signs upon the basis of those elements, and finally imposes names. Classification and speech have their place of origin in the same space that representation opens up within itself because it is consecrated to time, to memory, to reflection, to con­tinuity. But natural history cannot and should not exist as a language independent of all other languages unless it is a well-constructed language -and a universally valid one. In spontaneous and 'badly constructed' language, the four elements (proposition, articulation, designation, de­rivation) leave interstices open between them: individual experiences, needs or passions, habits, prejudices, a more or less awakened concentra­tion, have established hundreds of different languages - languages that differ from one another not only in the form of their words, but above all 158

CLASSIFYING

in the way in which those words pattern representation. Natural history can be a well-constructed language only if the amount of play in it is enclosed: if its descriptive exactitude makes every proposition into an invariable pattern of reality (if one can always attribute to the representa­tion what is articulated in it) and if the designation of each being indicates clearly the place it occupies in the general arrangement of the whole. In language, the function of the verb is universal and void; it merely pre­scribes the most general form of the proposition; and it is within the latter that the names bring their system of articulation into play; natural history regroups these two functions into the unity of the structure, which articulates together all the variables that can be attributed to a being. And whereas in language the designation, in its individual functioning, is exposed to the hazard of derivations, which endow the common names with their scope and extension, the character, as established by natural history, makes it possible both to indicate the individual and to situate it in a space of generalities that fit inside one another. So that above the ordinary, everyday words (and by means of them, since it is of course necessary to use them for the initial descriptions) there is raised the edifice of a language in the second degree in which the exact Names of things finally rule:

The method, the soul of science, designates at first sight any body in nature in such a way that the body in question expresses the name that is proper to it, and that this name recalls all the knowledge that may, in the course of time, have been acquired about the body thus named: so that in the midst of extreme confusion there is revealed the sovereign order of nature [60].

But this essential nomination - this transition from the visible structure to the taxonomic character - leads back to a costly requirement. In order to fulfil and enclose the figure that proceeds from the monotonous function of the verb to be to derivation and traversal of rhetorical space, spontaneous language had no need of anything but the play of imagina­tion: that is, of immediate resemblances. For taxonomy to be possible, on the other hand, nature must be truly continuous, and in all its plenitude. Where language required the similarity of impressions, classification requires the principle of the smallest possible difference between things. Now, this continuum, which appears therefore at the very basis of nomination, in the opening left between description and arrangement, is presupposed well before language, as its condition. And not only because 159

THE ORDER OF THINGS

it can provide the basis for a well-constructed language, but because it accounts for all language in general. It is without doubt the continuity of nature that gives memory the opportunity of exercising itself, as when a representation, through some confused and ill-perceived identity, recalls another and makes it possible to apply to both the arbitrary sign of a common name. What was presented in the imagination as a blind similitude was merely the blurred and unreflected trace of the great uninterrupted fabric of identities and differences. Imagination (which, by making comparison possible, justifies language) formed, without its then being known, the ambiguous locus in which the shattered but insistent continuity of nature was united with the empty but attentive continuity of consciousness. It would not have been possible to speak, there would have been no place for even the merest name, if nature, in the very depth of things, before all representation, had not been con­tinuous. To establish the great, unflawed table of the species, genera, and classes, natural history had to employ, criticize, and finally reconstitute at new expense a language whose condition of possibility resided pre­cisely in that continuum. Things and words are very strictly interwoven:

nature is posited only through the grid of denominations, and - though without such names it would remain mute and invisible - it glimmers far off beyond them, continuously present on the far side of this grid, which nevertheless presents it to our knowledge and renders it visible only when wholly spanned by language.

This, no doubt, is why natural history, in the Classical period, cannot be established as biology. Up to the end of the eighteenth century, in fact, life does not exist: only living beings. These beings form one class, or rather several classes, in the series of all the things in the world; and if it is possible to speak of life it is only as of one character - in the taxon-omic sense of that word - in the universal distribution of beings. It is usual to divide the things in nature into three classes: minerals, which are recognized as capable of growth, but not of movement or feeling;

vegetables, which are capable of growth and susceptible to sensation; and animals, which are capable of spontaneous movement [61]. As for life and the threshold it establishes, these can be made to slide from one end of the scale to the other, according to the criteria one adopts. If, with Maupertuis, one defines life by the mobility and relations of affinity that draw elements towards one another and keep them together, then one must conceive of life as residing in the simplest particles of matter. But one must situate it much higher in the series if one defines it by means of 160

CLASSIFYING

a crowded and complex character, as Linnaeus did when he set up as his criteria birth (by seed or bud), nutrition (by intussusception), ageing, exterior movement, internal propulsion of fluids, diseases, death, and presence of vessels, glands, epiderms, and utricles [62]. Life docs not con­stitute an obvious threshold beyond which entirely new forms of know­ledge are required. It is a category of classification, relative, like all the other categories, to the criteria one adopts. And also, like them, subject to certain imprecisions as soon as the question of deciding its frontiers arises. Just as the zoophyte stands on the ambiguous frontier between animals and plants, so the fossils, as well as the metals, reside in that uncertain frontier region where one does not know whether one ought to speak of life or not. But the dividing-line between the living and the non-living is never a decisive problem [63]. As Linnaeus says, the naturalist - whom he calls Historiens naturalis - 'distinguishes the parts of natural bodies with his eyes, describes them appropriately according to their number, form, position, and proportion, and he names them'[64]. The naturalist is the man concerned with the structure of the visible world and its denomination according to characters. Not with life.

We must therefore not connect natural history, as it was manifested during the Classical period, with a philosophy of life, albeit an obscure and still faltering one. In reality, it is interwoven with a theory of words. Natural history is situated both before and after language; it decomposes the language of everyday life, but in order to recompose it and discover what has made it possible through the blind resemblances of imagination;

it criticizes language, but in order to reveal its foundation. If natural history reworks language and attempts to perfect it, this is because it also delves down into the origin of language. It leaps over the everyday vocabulary that provides it with its immediate ground, and beyond that ground it searches for that which could have constituted its raison d'etre; but, inversely, it resides in its entirety in the area of language, since it is essentially a concerted use of names and since its ultimate aim is to give things their true denomination. Between language and the theory of nature there exists therefore a relation that is of a critical type; to know nature is, in fact, to build upon the basis of language a true language, one that will reveal the conditions in which all language is possible and the limits within which it can have a domain of validity. The critical question did exist in the eighteenth century, but linked to the form of a deter­minate knowledge. For this reason it could not acquire either autonomy or the value of radical questioning: it prowled endlessly through a region 161

THE ORDER OF THINGS

where what mattered was resemblance, the strength of the imagina­tion, nature and human nature, and the value of general and abstract ideas - in short, the relations between the perception of similitude and the validity of the concept. In the Classical age - Locke and Linnaeus, Buffon and Hume are our evidence of this - the critical question concerned the basis for resemblance and the existence of the genus.

In the late eighteenth century, a new configuration was to appear that would definitively blur the old space of natural history for modem eyes. On the one hand, we see criticism displacing itself and detaching itself from the ground where it had first arisen. Whereas Hume made the problem of causality one case in the general interrogation of resem­blances [65], Kant, by isolating causality, reverses the question; whereas before it was a question of establishing relations of identity or difference against the continuous background of similitudes, Kant brings into prominence the inverse problem of the synthesis of the diverse. This simultaneously transfers the critical question from the concept to the judgement, from the existence of the genus (obtained by the analysis of representations) to the possibility of linking representations together, from the right to name to the basis for attribution, from nominal articula­tion to the proposition itself, and to the verb to be that establishes it. Whereupon it becomes absolutely generalized. Instead of having validity solely when applied to the relations of nature and human nature, it questions the very possibility of all knowledge.

On the other hand, however, and during the same period, life assumes its autonomy in relation to the concepts of classification. It escapes from that critical relation which, in the eighteenth century, was constitutive of the knowledge of nature. It escapes - which means two things: life becomes one object of knowledge among others, and is answerable, in this respect, to all criticism in general; but it also resists this critical juris­diction, which it takes over on its own account and brings to bear, in its own name, on all possible knowledge. So that throughout the nine­teenth century, from Kant to Dilthey and to Bergson, critical forms of thought and philosophies of life find themselves in a position of reciprocal borrowing and contestation.

NOTES

[1] J. Ray published a Historia plantarum generalis as late as 1686. [2] Jonston, Historia naturalis de quadripedidus (Amsterdam, 1657, pp. I-II). [3] Diderot, Lettre sur les aveugles. Cf. Linnaeus: 'We should reject... all 162

CLASSIFYING

accidental notes that do not exist in the Plant either for the eye or for the touch' (Philosophic botanique, section 258).

[4] Linnaeus, Systema naturae, p. 214. On the limited usefulness of the microscope, cf. ibid, pp. 220-1. (We have retained throughout the author's references to the French editions of the works of Linne (Linnaeus) - translator's note.)

[5] Toumefort, Isagoge in rem herbarium (1719); Fr. trans. in Becker-Tournefort (Paris, 1956, p. 295). Buffon criticized the Linnaean method for relying upon characters so tenuous that it rendered the use of the microscope unavoidable. From one naturalist to another, reproof concerning the use of an optical instrument has value as a theoretical objection. [6] Linnaeus, Philosophic botanique, section 299. [7] Ibid., section 167; cf. also section 327. [8] Toumefort, Elements de botanique, p. 558. [9] Linnaeus, Philosophic botanique, section 299.

[10] Linnaeus (op. cit., section 331) lists the parts of the body that can be used as archetypes, whether for dimensions or, above all, for forms: hair, nails, thumbs, palms, eyes, ears, fingers, navel, penis, vulva, breasts.

[11] Ibid., sections 328-9.

[12] Buffon, Discours sur la maniere de trailer I'histoire naturelle (CEuvres com­pletes, t.I, p. 21).

[13] Adanson, Familles des plantes (Paris, 1763,1.1, preface, p. cci).

[14] Boissier de Sauvages, Nosologie methodique (Fr. trans. Lyon, 1772, t. I, pp. 91-2).

[15] Linnaeus, Philosophie botanique, section 258.

[r6] Toumefort, Elements de botanique, pp. 1-2.

[17] Linnaeus, Philosophic botanique, section 192.

[18] Ibid., section 193.

[19] Linnaeus, Systema naturae, section 12.

[20] Linnaeus, Philosophie botanique, section 77.

[21] Linnaeus, Systema naturae, section 12.

[22] 'The natural character of the species is its description' (Linnaeus, Philo­sophic botanique, section 193).

[23] Toumefort, Elements de botanique, p. 27.

[24] Linnaeus, Philosophic botanique, section 167.

[25] Linnaeus, Systeme sexuel des vegetaux (Fr. trans. Paris, year VI, p. 21).

[26] Linnaeus, Philosophie botanique, section 212.

[27] Ibid., section 284.

[28] Ibid., section 151. These two functions, which are guaranteed by the character, correspond exactly to the functions of designation and derivation performed in language by the common noun.

[29] Adanson, Histoire natmelle du Senegal (Paris, 1757).

163

THE ORDER OF THINGS

[30] Adanson, Cours d'histoire naturelle (Paris, 1772; 1845 edn., p. 17).

[31] Adanson, Families des plantcs.

[32] Ibid., t. I, preface.

[33] Linnaeus, Philosophic botaniquie, section 105.

[34] Ibid., section 94.

[35] Cf. P. Belon, Histoire de la nature des oisvaux.

[36] Cf. p. 113 above.

[37] Linnaeus, Philosophic botanique, section 156.

 [38] Ibid., section 169.

[39] Buffon, Discours sur la maniere de traiter I'histoire naturelle (CEuvres com­pletes, t.I, pp. 36 and 39).

[40] ?. Bonnet, Contemplation de la nature, lere partie (CEuvres completes,t. IV, pp. 35-0).

[41] Linnaeus, Philosophie botanique.

[42] Adanson, Cours d'histoire naturelle, 1845 edn., pp. 4-5.

[43 ] Buffon, Histoire de la terre.

[44] ?. Bonnet, Palingenesie philosophique (CEuvres completes, t. VII, p. 122).

[45] C. Bonnet, Contemplation de la nature, chap. XX, pp. 130-8.

[46] Buffon, Histoire naturelle des oiseaux (1770,t. I, p. 396).

[47] Pallas, Elenchus Zoophylorum (1786).

[48] J. Hermann, Tabulae affinltatum animaHum (Strasbourg, 1783, p. 24).

[49] C. Bonnet, Contemplation de la nature, lere partie (CEuvres completes, t. IV, p. 34 et seq.).

[50] C. Bonnet, Palingenesie philosophise (CEuvres completes, t. VII, pp. 149-150).

[51] C. Bonnet (CEuvres completes, t. Ill, p. 173) quotes a letter from Leibniz to Hermann on the chain of being.

[52] C. Bonnet, Palingenesie philosophique (CEuvres completes, t. VII, p. 193).

[53] Benolt de Maillet, Telliamed ou les entretiens d'un philosophc chinois avec un missionnaire francais (Amsterdam, 1748, p. 142).

[54] Maupcrtuis, Essai sur la formation des corps organises (Berlin, 1754, p. 41).

[55] J-B. Robinet, De la nature (yd edn., 1766, pp. 25-8).

[56] J-B. Robinet, Considerations philosophiques sur la gradation naturelte des formes de I'etre (Paris, 1768, pp. 4-5).

[57] Ibid., p. 198.

[58] On the non-existence of the biological notion of the 'environment' in the eighteenth century, cf. G. Canguilhem, La Connaissance de la vie (Paris, 2nd edn., 1965, pp. 129-54).

[59] J-B. Robinet, Considerations philosophiques sur la gradation naturelle des formes de I'etre, p. 19.

[60] Linnaeus, Systerna naturae, p. 13.

[61] Cf, for example, Linnaeus, Systema naturae, p. 215.

164

CLASSIFYING

[62] Linnaeus, Philosophie botanique, section 133. Cf. also Systeme sexuel des vegetaux, p. I.

[63] Bonnet accepted a quadripartite division in nature: unstructured brute beings, inanimate structured beings (vegetables), animate structured beings (animals), animate structured and reasoning beings (men). Cf. Contemplation de la nature, II ieme partie, chap. I.

[64] Linnaeus, Systema naturae, p. 215.

[65] Hume, A treatise of human nature (1739, book I, part III, section III, and part IV, section VI).

165

CHAPTER 6
Exchanging
I THE ANALYSIS OF WEALTH

There is no life in the Classical period, nor any science of life; nor any philology either. But there is natural history, and general grammar. In the same way, there is no political economy, because, in the order of knowledge, production does not exist. On the other hand, there does exist in the seventeenth and eighteenth centuries a notion that is still familiar to us today, though it has lost its essential precision for us. But 'notion' is not really the word we should apply to it, since it does not occur within an interplay of economic concepts that it might displace to some slight extent by taking over a little of their meaning or eating into their sphere of application. It is more a question of a general domain: a very coherent and very well-stratified layer that comprises and contains, like so many partial objects, the notions of value, price, trade, circulation, income, interest. This domain, the ground and object of 'economy' in the Classical age, is that of wealth. It is useless to apply to it questions deriving from a different type of economics - one organized around pro­duction or work, for example; useless also to analyse its various concepts (even, and above all, if their names have been perpetuated in succeeding ages with somewhat analogous meanings), without taking into account the system from which they draw their positivity. One might as well try to analyse the Linnaean genus outside the domain of natural history, or Bauzee's theory of tenses without taking into account the fact that general grammar was its historical condition of possibility.

We must therefore avoid a retrospective reading of these things that would merely endow the Classical analysis of wealth with the ulterior unity of a political economy in the tentative process of constituting itself. Yet it is in this way that historians of ideas do go about their reconstruc­tions of the enigmatic birth of this knowledge, which, according to them,

166

EXCHANGING

sprang up in Western thought, fully armed and already full of danger, at the time of Ricardo and J-B. Say. They presuppose that a scientific economics had for long been rendered impossible by a purely moral problematics of profit and income (theory of the fair price, justification or condemnation of interest), then by a systematic confusion between money and wealth, value and market price: and of this assimilation they take mercantilism to be one of the principle causes and the most striking manifestation. But then the eighteenth century is supposed to have pro­vided the essential distinctions and outlined some of the great problems that positivist economics subsequently treated with tools better adapted to the task: money is supposed to have revealed in this way its conven­tional - though not arbitrary - character (as a consequence of the long discussion between bullionists and anti-bullionists: among the first would have to be included Child, Petty, Locke, Cantillon, Galiani; among the latter, Barbon, Boisguillebert, and, above all, Law; then, to a lesser degree, after the disaster of 1720, Montesquieu and Melon); a beginning is thought to have been made, too - in the work of Cantillon - on the task of disentangling the theory of intrinsic value from that of market value; and the great 'paradox of value' was dealt with, by opposing the useless dearness of the diamond to the cheapness of the water without which we cannot live (it is possible, in fact, to find this problem rigor­ously formulated in Galiani); a start is supposed to have been made, thus prefiguring the work of Jevons and Menger, at connecting value to a general theory of utility (which we find sketched out in Galiani, in Graslin, and in Turgot); an understanding of the importance of high prices to the development of trade was supposedly reached (this is the 'Becher principle', taken up in France by Boisguillebert and Quesnay); lastly-and here we meet the Physiocrats - a start was made on the analysis of the mechanics of production. And thus, in fragments here and there, political economy is thought to have been silently bringing into position its essential themes, until the moment when, taking up the analysis of production again in another direction, Adam Smith is supposed to have brought to light the process of the increasing division of labour, Ricardo the role played by capital, and J-B. Say some of the fundamental laws of the market economy. From this moment on, political economy is supposed to have begun to exist with its own proper object and its own inner coherence.

In fact, the concepts of money, price, value, circulation, and market were not regarded, in the seventeenth and eighteenth centuries, in terms 167

THE ORDER OF THINGS

of a shadowy future, but as part of a rigorous and general epistemological arrangement. It is this arrangement that sustains the 'analysis of wealth' in its overall necessity. The analysis of wealth is to political economy what general grammar is to philology and what natural history is to biology. And just as it is not possible to understand the theory of verb and noun, the analysis of the language of action, and that of roots and their de­velopment, without referring, through the study of general grammar, to the archaeological network that makes those things possible and neces­sary; just as one cannot understand, without exploring the domain of natural history, what Classical description, characterization, and taxonomy were, any more than the opposition between system and method, or Tixism' and 'evolution'; so, in the same way, it would not be possible to discover the link of necessity that connects the analysis of money, prices, value, and trade if one did not first clarify this domain of wealth which is the locus of their simultaneity.

It is true that the analysis of wealth is not constituted according to the same curves or in obedience to the same rhythm as general grammar or natural history. This is because reflection upon money, trade, and ex­change is linked to a practice and to institutions. And though practice and pure speculation may be placed in opposition to one another, they nevertheless rest upon one and the same fundamental ground of know­ledge. A money reform, a banking custom, a trade practice can all be rationalized, can all develop, maintain themselves or disappear according to appropriate forms; they are all based upon a certain ground of know­ledge: an obscure knowledge that does not manifest itself for its own sake in a discourse, but whose necessities are exactly the same as for abstract theories or speculations without apparent relation to reality. In any given culture and at any given moment, there is always only one episteme that de­fines the conditions of possibility of all knowledge, whether expressed in a theory or silently invested in a practice. The monetary reform prescribed by the States General of 1575, mercantilist measures, or Law's experiment and its liquidation, all have the same archaeological basis as the theories of Davanzatti, Bouteroue, Petty, or Cantillon. And it is these fundamental necessities of knowledge that we must give voice to.
II MONEY AND PRICES

In the sixteenth century, economic thought is restricted, or almost so, to the problem of prices and that of the best monetary substance. The

168

EXCHANGING

question of prices concerns the absolute or relative character of the in­creasing dearness of commodities and the effect that successive devaluations or the influx of American metals may have had upon prices. The problem of monetary substance is that of the nature of the standard, of the price relation between the various metals employed, and of the distortion between the weights of coins and their nominal values. But these two scries of problems were linked, since the metal appeared only as a sign, and as a sign for measuring wealth, in so far as it was itself wealth. It possessed the power to signify because it was itself a real mark. And just as words had the same reality as what they said, just as the marks of living beings were inscribed upon their bodies in the manner of visible and positive marks, similarly, the signs that indicated wealth and measured it were bound to carry the real mark in themselves. In order to represent prices, they themselves had to be precious. They had to be rare, useful, desirable. Moreover, all these qualities had to be stable if the mark they imprinted upon things was to be an authentic and universally legible signature. Hence the correlation between the problem of prices and the nature of money, which constitutes the privileged object of all reflection upon wealth from Copernicus to Bodin and Davanzatti.

The two functions of money, as a common measure between commod­ities and as a substitute in the mechanism of exchange, are based upon its material reality. A measure is stable, recognized as valid by everyone and in all places, if it has as a standard an assignable reality that can be com­pared to the diversity of things that one wishes to measure: as is the case, Copernicus points out, with the fathom and the bushel, whose material length and volume serve as units [1]. In consequence, money does not truly measure unless its unit is a reality that really exists, to which any commodity whatever may be referred. In this sense, the sixteenth century returns to the theory accepted during at least part of the Middle Ages, which gave either the prince or popular consent the right to fix the valor impositus of money, to modify its rate, to withdraw any category of coins or any particular metal. The value of money must be determined by the quantity of metal it contains; that is, it returns to what it was before, when princes had not yet stamped their effigy, or seal upon pieces of metal; at that time 'neither copper, nor gold, nor silver were minted, but only valued according to their weight'[2]; arbitrary signs were not accorded the value of real marks; money was a fair measure because it signified nothing more than its power to standardize wealth on the basis of its own material reality as wealth.

169

THE ORDER OF THINGS

It is upon this epistemological foundation that reforms were effected in the sixteenth century, and that the controversies of the age assumed their particular dimensions. There was an attempt to bring monetary signs back to their exactitude as measures: the nominal values stamped on the coins had to be in conformity with the quantity of metal chosen as a standard and incorporated into each coin; money would then signify nothing more than its measuring value. In this sense, the anonymous author of the Compendious insists that all the money actually current should cease to be so after a certain date, since the'forcing up'of its nominal value has long since vitiated its functions of measurement; all coinage already minted should then be accepted only in accordance with the amount of metal it is estimated to contain; as for new money, that will have its own weight as its nominal value, so that henceforward only the new and the old money will be current, each in accordance with one and the same value, weight and denomination, so that all money will be re-established at its former rate and regain its former goodness [3]. It is not known whether the Compendious, which was not published before 1581, but was certainly in existence and circulating in manuscript for thirty years be­forehand, inspired England's monetary policy under Elizabeth. One thing is certain: that after a series of'forcings up' (devaluations) between 1544 and 1559, the proclamation of March 1561 'brought down' the nominal value of money and made it equal once more to the quantity of metal each coin contained. Similarly, in France, the States General of 1575 asked for and obtained the suppression of accounting units (which introduced a third definition of money, a purely arithmetical one, in addition to the definition by weight and that by nominal value: this supplementary relation concealed the sense of monetary operations from those who did not understand it); the edict of September 1577 established the gold ecu as both a real coin and an accounting unit, decreed the subordination of all other metals to gold - in particular, silver, which re­tained its legality as tender but lost its legal immutability. The coinage was thus restandardized on the basis of its metallic weight. The sign the coins bore - the valor impositus - was merely the exact and transparent mark of the measure they constituted.

But at the same time as this restandardization was being demanded, and occasionally accomplished, a certain number of phenomena came to light which are peculiar to the money-sign and perhaps definitively compromised its role as a measure. First, the fact that coinage circulates all the quicker for being less good, whereas coins with a high percentage

170

EXCHANGING

of metal are hoarded and do not take part in trade: this is what was called Gresham's law [4], and both Copernicus [5] and the author of the Compendious[6] were already aware of it. Second, and above all, there was the relation between the monetary facts and the movement of prices: it was this that revealed money as a commodity like any other - not an absolute standard for all equivalences, but a commodity whose capacity for ex­change, and consequently whose value as a substitute in exchange, are modified according to its abundance or rarity: money too has its price. Malestroit[7] had pointed out that, despite appearances, there had been no increase in prices during the sixteenth century: since commodities are always what they are, and since money, in its particular nature, is a con­stant standard, the increased dearness of commodities can be due only to the augmentation of the nominal values borne by an unchanging metallic mass: but, for the same quantity of wheat, one still gives the same weight in silver or gold. So that 'nothing has become dearer': since the golden ecu was worth twenty sols tournois in accounting money under Philippe VI, and since it is now worth fifty, it is inevitable that an ell of velvet, which formerly cost four livres, should now be worth ten. 'The increasing dearness of things does not come from having to deliver more but from receiving a lesser quantity of gold or fine silver than one was accustomed to before.' But once this identification has been established between the role of money and the mass of metal it causes to circulate, it becomes clearly apparent that it is subjected to the same variations as all other merchandise. And though Malestroit implicitly admitted that the quantity and marketable value of metals remained stable, Bodin, only a very few years later [8], observes that there has been an increase in the stock of metal imported from the New World, and in consequence a real increase in the price of commodities, since princes, now possessing ingots in larger quantity or receiving more from private persons, have been minting more and better-quality coins; for the same amount of a commodity one is therefore giving a larger quantity of metal. The rise in prices therefore has a 'principal cause, and that almost the only one that no one has touched upon hitherto': 'the abundance of gold and silver', 'the abundance of that which gives things estimation and price'.

The standard of equivalences is itself involved in the system of ex­changes, and the buying power of money signifies nothing but the marketable value of the metal. The mark that distinguishes money, determines it, renders it certain and acceptable to all, is thus reversible, and may be construed in either direction: it refers to a quantity of metal

171

THE ORDER OF THINGS

that is a constant measure (which is the construction Malestroit puts upon it); but it also refers to certain commodities, variable in quantity and price, called metals (which is Bodin's reading of the matter). We are, then, presented with an arrangement analogous to that which characterizes the general organization of signs in the sixteenth century: signs, it will be remembered, were constituted by resemblances which, in turn, neces­sitated further signs in order to be recognized. Here, the monetary sign cannot define its exchange value, and can be established as a mark only on a metallic mass which in turn defines its value in the scale of other commodities. If one admits that exchange, in the system of needs, cor­responds to similitude in the system of acquired knowledge, then one sees that knowledge of nature, and reflection or practices concerning money, were controlled during the Renaissance by one and the same configuration of the episteme.

And just as the relation of the microcosm to the macrocosm was in­dispensable in order to arrest the indefinite oscillation between resem­blance and sign, so it was necessary to lay down a certain relation between mital and merchandise which, when it came to it, made it possible to fix the total marketable value of the precious metals, and consequently to standardize the price of all commodities in a certain and definitive fashion. This relation is the one that was established by Providence when it buried gold and silver mines under the earth, and caused them to grow, just as plants grow and animals multiply on the surface of the earth. Between all the things that man may need or desire, and the glittering, hidden veins where those metals grow in darkness, there is an absolute correspon­dence. As Davanzatti says:

Nature made all terrestrial things good; the sum of these, by virtue of the agreement concluded by men, is worth all the gold that is worked; all men therefore desire everything in order to acquire all things ... In order to ascertain each day the rule and mathematical proportions that exist between things and between them and gold, we should have to be able to contemplate, from the height of heaven or some very tall observatory, all the things that exist or are done on earth, or rather their images reproduced and reflected in the sky as in a faithful mirror. We would then abandon all our calculations and we would say: there is upon earth so much gold, so many things, so many men, so many needs; and to the degree that each thing satisfies needs, its value shall be so many things, or so much gold[9].

172

EXCHANGING

This celestial and exhaustive calculation can be accomplished by none other than God: it corresponds to that other calculation that brings each and every element of the microcosm into relation with a corresponding element in the macrocosm - with this one difference, that the latter unites the terrestrial to the celestial, going from things, from animals, or from man, up to the stars; whereas the former links the earth to its caves and mines; it makes those things that are brought into being by the hands of men correspond with the treasures buried in the earth since the creation of the world. The marks of similitude, because they are a guide to know­ledge, are addressed to the perfection of heaven; the signs of exchange, because they satisfy desire, are sustained by the dark, dangerous, and accursed glitter of metal. An equivocal glitter, for it reproduces in the depths of the earth that other glitter that sings at the far end of the night: it resides there like an inverted promise of happiness, and, because metal resembles the stars, the knowledge of all these perilous treasures is at the same time knowledge of the world. And thus reflection upon wealth has its pivot in the broadest speculation upon the cosmos, just as, inversely, profound knowledge of the order of the world must lead to the secret of metals and the possession of wealth. It becomes apparent how tightly knit is the network of necessities that, in the sixteenth century, links together all the elements of knowledge: how the cosmology of signs pro­vides a duplication, and finally a foundation, for reflection upon prices and money; how it also authorizes theoretical and practical speculation upon metals; how it provides a communicating link between the promises of desire and those of knowledge, in the same way as the metals and the stars communicate with one another and are drawn together by secret affinities. On the confines of knowledge, in that region where it becomes all powerful and quasi-divine, three great functions meet - those of the Basileus, of the Philosophos, and of the Metallicos. But just as this know­ledge is given only in fragments and in the attentive lightning-flash of the divinatio, so, in the case of the singular and partial relations of things with metal, of desire with prices, divine knowledge, or, that which one might acquire from 'some very tall observatory', is not given to man. Except for brief instants, and as though at random, to those minds that know how to watch for it - in other words, to merchants. What the soothsayers were to the undefined interplay of resemblances and signs, the merchants are to the interplay, also forever open, of exchange and money.

173

??E ORDER OF THINGS

From here below, we have difficulty in perceiving the few things that surround us, and we give a price to them according to whether we perceive them to be more or less in demand in each place and at each time. The merchants are promptly and very well advised of these things, and that is why they have an admirable knowledge of the price of things[10].