Les Mots et les choses, forty years on
Ian Hacking, Collège de France
For Humanities Center, Columbia University, 6
th
October 2005
1 1966
Les Mots et les Choses, une archéologie des sciences
humaines was an instant success when it was published in April,
1966, sold out in 90 days. Everyone was talking about the
famous final paragraph about the erasure of Man, and a sentence
shocking to Parisian eyes, namely ‘Marxism swam in 19
th
century thought like a fish in the water’. L’Express, France’s
simulacrum to Time magazine, billed it as the greatest revolution
in philosophy since existentialism [23 May 1966].
For a good sense of one way that Foucault saw his book just
after it was published, look at an interview for La quinzaine
littéraire, 16 May 1966. [‘Entretien avec Madeleine Chapsal’,
Dits et écrits, 2 vol. edn., vol. 1, 541546.] He is a member, he
told the interviewer, of the generation who were not yet 20
during the war. (He himself was 13 when it began and 18 when
France was liberated.) Much as that generation admired Sartre’s
courage and generosity, his passion for life, politics, and
existence, he said, ‘we, we have discovered something else,
another passion: the passion of the concept and for what I call
the “system”.’ [p. 542.]
As far as grand and overstated themes go, we find them in
this interview: ‘Our present task is to liberate ourselves
definitively from humanism, and, in this sense, our work is
political.’ Political? Yes, for in Foucault’s view, ‘all the regimes
of East and West market their evil wares under the flag of
humanism.’ [p. 544.] Remember those were the days when
Mots et les choses, 40 years on
1
Teilhard de Chardin was a big thing, and when Foucault could
praise ‘Althusser and his courageous companions battling
against “chardinomarxism”.’ [p. 544.] But note also his
denunciation of ‘the monolingual narcissism of the French’ for
thinking that they have just discovered a new set of problems,
when in fact the field of research that so engaged young French
intellectuals had emerged in America, England and France just
after the first world war, when ideas were coming in from the
German and Slavicspeaking lands. France has been called the
Hexagon by the French ever since their boundaries became
roughly hexagonal. We have such hexagonal minds, Foucault
continued by saying, that De Gaulle passes among us for an
intellectual.
Forty years. A few months ago in Paris I organized, as an act
of local piety, what turned out to be a very lively
commemoration of a book published sixty years ago, Merleau
Ponty’s Phenomenology of Perception. That remains a great
book. Since I own up to having been a fan of JeanPaul Sartre
ever since I first read the man at the age of 18, I may be allowed
to say that the Phenomenology is far more interesting philosophy
than anything Sartre wrote. But what an amazing time span is
twenty years. The cultural and conceptual gulf between the
Phenomenology and The Order of Things is total. It was not only
this book that came on the scene. Daniel Defert observes in the
absolutely terrific 90 page Chronologie which introduces his
collection of Michel Foucault’s Dits et écrits, that ‘1966 is one
of the great vintage years (grands crus) of French human
sciences: Lacan, LéviStrauss, Benveniste, Genette, Greimas,
Doubrovsky, Todorov and Barthes published some of their most
important texts.’ [ Vol. 1, p. 37, entry for June 1966.] Some new
generation.
Mots et les choses, 40 years on
2
2 1970
The intended title of Les Mots et les choses was L’Ordre des
choses, but that could not be used because it had served as the
title of one recent and one less recent book by other authors.
Hence the English translation of 1970, The Order of Things: An
Archaeology of the Human Sciences was the right title. It had a
more mixed reception when it was published in 1970 by the
Tavistock Press, than the French book had in 1966. My own
response was unequivocal. I bought my third hardcover copy
within the year. On the flyleaf I wrote, This is my third copy
after losing 2. Please return; I don’t want to buy a 4
th
@ £4.60 a
time! So evidently my copies were being loaned around. The
book enabled me to do philosophy in what was, for me, a new
way. This does not mean that I quit doing philosophy in old
ways, but that I started also to do something different. The
Archaeology of Knowledge came out in English in 1972, and I
wrote it up immediately in the weekly Cambridge Review – so
hastily that I, the editors, and the printers all left out the ‘a’ in
the middle of ‘Archaeology’, thereby making it look like a
French word that had lost its accent.
In the early seventies, I gave as lectures what was to be
published as, Why Does Language Matter to Philosophy? There
are some signs there, but not too many, of having read Foucault.
In the spring of 1974 I gave a course of lectures about some of
Foucault’s work. A colleague is reported to have told a visitor,
‘if you wonder why the bookshops have copies of Foucault in
their front windows, it is all Hacking’s fault’. That spring, or the
previous autumn, I gave lectures on what was to become The
Emergence of Probability, published in 1975. If you were
unkind, you might call that book The Order of Things, the
Mots et les choses, 40 years on
3
footnote. But it is a footnote only to some early parts of Les Mots
et les choses, Foucault’s discussion of the radical transition from
the Renaissance doctrine of similitudes to the formal structures
of representation.
Representation, Foucault taught, is characteristic of what in
English we call, or used to call, the Age of Reason, and what he
called for his French readers, the Classical age, namely the
Cartesian era when French thought and the French language
became world thought and a world language. Emergence said
nothing relevant to the second great mutation described by
Foucault, the transition from representation to history. Maybe
Foucault himself had begun with the conception that I got from
the book, for he said he was writing ‘his book about signs’. That
is according to Daniel Defert, [D&é, I, p. 34], who says that the
first composition of the book about signs was finished in Tunisia
over Christmas 1964. But in a letter of 13 February, 1965,
Foucault realized that his book had changed: ‘I’ve not been
talking about signs but about order’.
I myself did not even want to think about what most
impressed people about his book, namely its final marvellous
paragraph announcing the incipient erasure of Man as an object
of knowledge or a topic of discourse. I did not want to think
about it because it seemed to me to be a mistake.
My talk today should perhaps be called not, ‘Les Mots et les
choses, forty years on’, but ‘The Order of Things, thirtyfive
years later’. Today, 35 years after I read the book and reacted
negatively to the endofMan thesis, I shall try to say what the
mistake was. It is not a very exciting mistake, and I am sure
others have pointed to the problem long ago. But first I would
like to say a little about the preceding, lesser, blockbuster, the
book on madness. My last personal remark in this overly
Mots et les choses, 40 years on
4
personal introduction is that I was given the English
abridgement, Madness and Civilization: A History of Insanity in
the Age of Reason, while I was working in Uganda, probably
early 1968, but maybe late 1967, the year in which it was first
published in London by the Tavistock Press. I was bowled over
by it, and thus was primed for the larger blockbuster,
archaeology in full spate. By the way, the Tavistock Press, which
I have now mentioned twice, was the publishing arm of
Tavistock House, a venerable London institution for
psychotherapy, hospitable to new ideas, and which was then a
base for the antipsychiatry movement.
I had intended to talk today in some detail about the structure
of the last few chapters of The Order of Things, but gradually my
plan changed. I shall say more about the earlier big book about
madness, published in 1961.
3 The timing of Folie et Déraison
The big book, as I shall continue to call it, was published in
1961. The French wars of the 1950s in Vietnam and Algeria had
been lost, and the Republic had put them behind it. The sixties
were the most fertile decade of French intellectual life in the
twentieth century. In addition to the list of names cited by
Defert, we recall that during that decade Gilles Deleuze
published seven of his books, and Jacques Derrida established
himself. Foucault’s work on madness was one of the first
flowerings of this unique decade.
1961 also marks the critical rethinking of madness in the
English speaking world. There was the polemical assault on
psychiatry by Thomas Szasz, The Myth of Mental Illness. Erving
Goffman, one of the greatest of sociologists, published Asylums,
Mots et les choses, 40 years on
5
an analysis of what actually happens in institutions for the
insane. In 1960, R. D. Laing’s The Divided Self: Sanity and
Mental Illness had made radical proposals that schizophrenia
was a product of the family situation, not of the demented
patient. None of these authors knew of each other at the time.
Laing’s collaborator David Cooper coined the name ‘anti
psychiatry’, and later helped get the abridgement of Foucault’s
work on madness published in English.
The book was, then, launched in France as part of a gigantic
wave of new thinking, which coincided with brilliant anglophone
voices challenging established psychiatry. All over the world
Foucault was read as a critic of psychiatry. That always happens
when an author establishes that something, that we think of as
inevitable, is the product of a series of historical events.
Foucault’s next book was The Birth of the Clinic: An
Archaeology of the Medical Gaze. He finished it at the end of
1961, published in 1962. He thought of it as the remains of, or
fallout from, the book on madness – it’s chute. He tried to assert
in the first pages that in writing about madness, and now in
writing about medicine, he was not intending to favour one
system over another, but to say what made all of them possible at
various times.
The 1961 Folie et Déraison did indeed turn out to be, as its
first preface promised, the beginning of a long series of
investigations. Innumerable themes found here were to be
reworked over the next few years, not least among them being
that of something to be called ‘archaeology’, a neologism
announced in the same preface. Other themes: Exclusion;
Conditions of possibility; The coming into being of a sense of
history; Time, yes, but also the spatial character of all Foucault’
s analyses is prefigured. ‘Discursive formation’, which became a
Mots et les choses, 40 years on
6
vogue phrase in American, and still clutters up undergraduate
essays, had not yet surfaced.
One thought was to be decisively dropped. It was the
idea of there being some underlying truth about madness.
The idea that there is in madness an inaccessible primitive
purity (F&D, p. 000, in italics). Those words are from the
original preface. Part of that preface, including those
words, was suppressed by 1964. But I believe they are
true to the thought that went into the book. The romantic
fantasy lurks here, the purity of the possessed, of those
who not only speak the truth in paradox, like the fools in
Shakespeare, but who are themselves the truth.
Today we know all about Foucault’s complete
rejection of the idea of an essential self. So it is strange to
read David Copper’s preface to Madness and Civilization,
published in 1965. [Have to check it is in the US 1965 1
st
edition, or is it only in the Tavistock edition of 1967?] He
said that antipsychiatry – by implication Laing’s and his
own – can do better than psychoanalysis. At great cost to
a patient, analysis may ‘achieve a workable conformism –
defined as normality, maturity, developedness’. Instead,
said Cooper: ‘The truer goal, however, must be in terms
of a recognizable synthesis of the field of social
practicality with its secret antithesis – the autonomous
assertion of pure, spontaneous Self.’ (M&C xi.) That
capital S ‘Self’ is not, I believe, wholly false to Folie et
Déraison as originally conceived.
Everybody knows that Foucault withdrew any
suggestion that there might be some pure, ahistorical,
Mots et les choses, 40 years on
7
madness, but I think it is worth looking again at what he
withdrew.
4 Two books about madness?
Foucault worked on madness for a long time. The big book
came out in 1961. He also submitted it for his doctoral degree,
along with a shorter thesis on Kant’s Anthropology from a
Practical Point of View. He was then 35 years old. The second
edition of 1972 differs only in preface and appendices. In
between 1961and 1972 there was an abridgement and there were
translations. In the splendidly brief, and seemingly transparent
preface, for the 1972 edition, Foucault spoke, as was then his
wont, of a book being an event, which if it lives, lives through
repeated doublings and simulacra. It has its own life, free of the
author.
Doublings: I suggest that there are two distinct books, 1961
and 1972. The main text of each is the same. It is all too easy to
compare them to the two Don Quixotes invented by Borges, the
one written by Cervantes, the other, identical in words, written
much later by an imagined Pierre Menard. Despite the words
being the same, so much has happened that the meaning is
different.
1
I shall explain this in several stages. First, the text remains,
but not the title. Second, something called unreason – déraison
– is highlighted in the first book but not the second. Third,
Foucault’s concept of archaeology may change its role. All three
evolutions are, I would like to say, connected with the way in
which the author liberated himself from an obsession with
Mots et les choses, 40 years on
8
1
‘Pierre Menard, Author of Don Quixote’, in J. L. Borges,
Ficciones, Grove Press, 1962, 4555 (original Spanish of 1939).
madness and inaccessible primitive purity. I say ‘author’
deliberately: this is not intended as a remark about the
psychohistory of Michel Foucault, but about the topics of the
texts.
2
5 Titles
The exact title in 1961 was Folie et Déraison. Histoire de la
folie à l’âge classique. Daniel Defert, Foucault’s long time
intellectual colleague, companion, and posthumous editor, laid
emphasis on the exact title of the original. That first title is a bit
like Alice’s Cheshire Cat, of which nothing is left but the grin,
namely the subtitle. The unabridged book will appear for the first
time in English later this year. It will be called History of
Madness. Even the classical age, or the age of reason, is to
disappear.
A trip through the successive titles is a gradual vanishing act,
which points us away from déraison. At the beginning,
‘Unreason’ was right up there alongside ‘Madness’, and it was
capitalized in the original French title, as it sometimes is in the
book itself. On the cover of the 1964 paperback abridgement we
see only Histoire de la folie. On the title page the full 1961 title
appears in block letters, but with the title Folie et Déraison in
smaller print than the subtitle Histoire de la folie. Fading, like
the cat. Parts of the preface were eliminated, including the
reference to inaccessible primitive purity. This version was
Mots et les choses, 40 years on
9
2
Each of the authors who discusses Historie de la folie in Michel
Foucault. Lire l’œuvre (ed. Luce Giard, Grenoble : Jérôme
Millin 1992) discusses the differences between the two prefaces;
but I hope to convince you that there are many more differences.
translated into many languages, while only an Italian publisher
did the unabridged book.
For the 1965 English version, Madness and Civilisation,
Foucault restored some material that he had cut from the 1964
French abridgement. For a moment, as we shall see, part of the
cat’s face flickered back. The English slightlylessabridged
abridgement reinserted a chapter that is all about unreason.
As an aside, let me say that Madness and Civilisation was
widely criticized by the best of scholars writing in English –
Lawrence Stone, Roy Porter, and others – for being, in a word,
too hexagonal. Grand claims were made for a series of events
taking place all over Europe. But there was no ‘great
confinement’ in England or for that matter anywhere else outside
of France. I am much more troubled by the fact that a French
scholar, Pierre Quétel, argues that it is not good history of
madness even in Paris.
3
It is disquieting that nobody wanted to
listen to him.
4
He is now the historian running a war museum at
Cherbourg. But for better or worse, we do not read any version
of the book, long or short, for history, but for a complete
rethinking of the very idea of madness.
To pass to the end of the story, in June, 1972 we have the
‘doubling’. The text of 1961 was printed again, untouched. The
old preface was now wholly deleted. A brief new preface was
inserted, with its mention of doubling. Two appendices were
added. One bears on the spat with Derrida about madness in the
Cartesian Meditations. The other is very curious piece to which I
shall return. It, I shall ague, is the transition from madness to the
Mots et les choses, 40 years on
10
3
Pierre Quétel, “Fautil critiquer Foucault ?” in Penser la folie.
Essais sur Michel Foucault, Paris: Galilée, 1992, 81102.
4
“Résumé du débat”, ibid., 103105
disappearance of man., but for that we shall wait a few minutes.
The most obvious thing to notice is that the main title, Folie et
Déraison,was dropped. The latest doubling, to appear early next
year, is the complete English translation of everything : text, old
preface, new preface, appendices, plus further appendices, plus
an introduction by the editor, plus a very short foreword by me.
6 Unreason
‘Unreason’ is only barely an English word, but it is the only
possible translation of déraison, an old French word that Le petit
Robert classifies as ‘Vx ou lit’ – old or literary.
Everybody professing to be in the know has long dismissed
the English version of 1965 as a callow abridgement. In fact it is
highly instructive, for Foucault decided to reinsert the most vivid
assertions about Unreason to be found in the entire work. They
include sentences like this, in the new translation: ‘How can we
avoid summing up this experience by the single word Unreason?
By that we mean all that for which reason is at once nearest and
most distant, fullest and most empty.’ [The word Unreason is
italicized and capitalized.] Or in the old translation,
All that madness can say of itself, is merely reason,
although it is the negation of unreason. In short, a
rational hold over madness is always possible and
necessary, to the very degree that madness is nonreason.
There is only one word which summarizes this
experience: Unreason: all that, for reason, is closest and
most remote, emptiest and most complete; all that
presents itself to reason in familiar structures –
authorizing a knowledge, and then a science, which seeks
to be positive – and all that is constantly in retreat from
Mots et les choses, 40 years on
11
reason, in the inaccessible domain of nothingness.
[original italics, my underlining]
Foucault continued immediately by playing on dazzlement.
And if, now, we try to assign a value, in and of itself,
outside its relations with the dream and with error, to
classical unreason, we must understand it not as reason
diseased, or as reason lost or alienated, but quite simply
as reason dazzled. (M&C 108)
What follows in Foucault is itself deliberately so dazzling that
we, or at any rate I, am left in a state of bedazzlement.
Unreason is in the same relation to reason as dazzlement
to the brightness of daylight itself. And this is not a
metaphor. (M&C 109)
I have not found the noun ‘dazzlement’ in any onevolume
English dictionary, but it perfectly translates a genuine French
noun (éblouissement).
5
Continuing this single paragraph we get
an amazing play of ‘day and night’ (italicized), which is
no longer the fatal time of the planets; it is not yet the
lyrical time of the seasons; it is the universal but
absolutely divided time of brightness and darkness. A
Mots et les choses, 40 years on
12
5
Foucault’s footnote here (deleted from the abridged edition)
directs us to Pierre Nicole, the famous Jansenist, whose name the
text misspelled as Nicolle (in both editions). Nicole wrote a
semipopular series of essays on morality in 14 volumes, starting
in 1671. Foucault explains (?) that ‘dazzlement’ is meant in the
sense of a sentence from the eighth volume, where Nicole
wondered whether the heart took part in the ‘dazzlement of the
mind’.
form which thought entirely masters in a mathematical
science …
By the end of the paragraph we have been taken from ‘Cartesian
physics, a kind of mathesis of light’, through Racine and
Georges de La Tour.
6
It is stupendous stuff, to be kept on the
back shelf away from sophomores. A display of Foucault’s
brilliant eye for paint: ‘the page holding the torch reveals under
the shadow of the vault the man who was his master – a grave
and luminous boy encounters all of human misery; a child brings
death to light’. Then, in the twinkling of an eye, to the plays of
Racine. All interspersed with observations about madness itself:
‘the madman, conversely, finds in daylight only the
inconsistencies of the night’s figures; he lets the light be
darkened by all the illusions of the dream’. Then a pause, a
break in the page, followed by a halfway summary, in two brief
paragraphs:
Madness designates the equinox between the vanity of
night’s hallucinations and nonbeing of light’s
judgements.
And this much, which the archaeology of knowledge
has been able to teach us bit by bit, was already offered to
us […] in the last words of Andromaque. (M&C 111112)
We are far from finished. ‘The curtain that falls on the last scene
of Andromaque falls on the last great tragic incarnation of
madness’ …
This had to be the last scene of the first great classical
tragedy; or if one prefers the first time in which the
classical truth of madness which is the last of the
Mots et les choses, 40 years on
13
6
Foucault writes Georges de la Tour, but of course he was
Georges de La Tour.
preclassical theatre. A truth, in any case, that is
instantaneous, since its appearance can only be its
disappearance; the lightening flash is seen only in the
advancing night.
After two more pages brilliantly reenacting tragedies, we return
to unreason:
The movement proper to unreason, which classical
learning followed and pursued, had already accomplished
the whole of its trajectory in the conclusion of tragic
language. After which silence could reign, and madness
disappear in the – always withdrawn – presence of
unreason.
What we now know of unreason affords us a better
understanding of what confinement was.
Foucault does not mention dates here, but of course he expected
his readers to know them. The ‘great confinement’ of Chapter 2
was said to begin in 1656; Andromaque was first produced in
1667. De La Tour is earlier; his most famous painting in the
Louvre is 1645 (St Joseph) and he died in 1652.
One can imagine several explanations for the fact that all this
material was deleted in 1964, and reinserted in 1965. We read
simply these words in the front matter: ‘The author has added
some material from the original edition.’
7 Archaeology
I have already quoted that strange passage where we first find
the expression, ‘the archaeology of knowledge’:
Mots et les choses, 40 years on
14
Madness designates the equinox between the vanity of
night’s hallucinations and nonbeing of light’s
judgements.
And this much, which the archaeology of knowledge
has been able to teach us bit by bit, was already offered to
us […] in the last words of Andromaque. (M&C 111112)
Foucault in fact spoke of archaeology in the part of the preface
to the 1961 book which was retained for the 1964 abridgement.
There is a moving paragraph stating that there is no longer a
common language by means of which madness and reason can
communicate.
the constitution of madness as a mental illness, at the end
of the eighteenth century, affords the evidence of a
broken dialogue, […]. The language of psychiatry, which
is a monologue of reason about madness, has been
established only on the basis of such a silence.
I have not tried to write the history of that language,
but rather the archaeology of that silence. (M&C, xiixiii.)
An archaeology of silence: neither a history of psychiatry, nor an
archaeology of psychiatric discourse. Foucault wanted to
understand how a certain absence of discourse became possible.
The theme of silence was reiterated in the first sentence of
chapter 2, the famous chapter called ‘The Great Confinement’:
By a strange act of force, the classical age was to reduce
to silence the madness whose voices the Renaissance had
just liberated, but whose violence it had already tamed.
(M&C 38)
What is absent? The voice of unreason, of which there is
a simulacrum only in those artworks that were the
Mots et les choses, 40 years on
15
veritable paradigms of genius in the 1960s, Nerval,
Nietzsche, Hölderlin, and above all Artaud.
I suggest this is a very different kind of archaeology
from his later use of the term. It is not an attempt to
understand how sentences were possible, but rather their
absence.
8 Unreason and art
Unreason was not wholly pruned from the 1964
abridgement. The whole book is about the dialectical play
between madness and unreason.
In the classical period, the awareness of madness and the
awareness of unreason had not separated from one
another. The experience of unreason that had guided all
the practices of confinement so enveloped the awareness
of madness that it very nearly permitted it to disappear,
sweeping it along a road of regression where it was close
to losing its most specific elements.
But in the anxiety of the second half of the eighteenth
century, the fear of madness grew at the same time as
terror in the face of unreason: and thereby these two
Mots et les choses, 40 years on
16
forms of obsessive dread, each supporting the other,
constantly grew stronger. (M&C 211
7
)
That is, alongside madness there is also unreason; the one
provokes fear, the other terror. Unreason had much fuller
play in the Renaissance, an unreason that Foucault evokes
by the ship of fools and the paintings of Bosch, especially
the temptation of Saint Anthony.
Still later we read that since the end of the eighteenth
century ‘the life of unreason’ no longer manifests itself
Mots et les choses, 40 years on
17
7
The original text reads :
Mais dans l’inquiétude de la seconde moitié du xviiie siècle, la
peur de la folie croît en même temps que la frayeur devant la
déraison : et par là même les deux formes de hantise, s’appuyer
l’une sur l’autre, ne cessent de se renforcer.
It does not matter much to the overall meaning, and I make no
pretense to have got mine just right – I am no translator. But for
an example of the difficulties, the 1965 translation has :
But in the anxiety of the second half of the eighteenth century,
the fear of madness grew at the same time as the dread of
unreason: and thereby the two forms of obsession, leaning upon
each other, continued to reinforce each other. (M&C 211)
In the new translation:
But in the anxiety of the second half of the eighteenth century,
the fear of madness grew at the same rate as the dread of
unreason: and for that reason these twin obsessions constantly
leant each other mutual support.
except in the flashes of lightning found in works like
those of ‘Hölderlin, Nerval, Nietzsche and Artaud’ which
resisted ‘through their own strength that gigantic moral
imprisonment’. Standard French history and iconography
of psychiatry represents Pinel casting off the chains of
madmen soon after the Bastille had fallen, so that is a
story of liberating the mad. Here we are told that the new
‘moral’ treatment of insanity was also the imprisonment
of unreason that had flashed so openly on the canvasses
of Hieronymous Bosch.
Casual reference books diagnose Hölderlin as a
schizophrenic poet, Nietzsche as a philosopher suffering
from dementia caused by terminal syphilis, and Artaud as
a bipolar (manicdepressive) playwright. Nerval killed
himself: We say suicide is caused by severe depressive
illness. It is of course a central thesis of this book, that far
from these being inevitable ways of conceiving of four
very strange men, it requires a specific system (yes,
system) of thought to categorize these men – and myriad
other people – in terms of mental disorder. But Foucault
does not pander to the thought that genius and mental
disturbance are of a piece. The art of these men, as he
shouts in italics on the last page of the book, is not their
insanity, but its opposite. In the old translation which you
may know: ‘Where there is a work of art, there is no
madness.’ (M&C 2889.) In the new translation, ‘Where
there is art, there is no madness.’ In the original, Là où il
y a œuvre, il n’y a pas de folie (HF 557).
9 Madness, the absence of a work
Mots et les choses, 40 years on
18
In May 1964, Michel Foucault contributed to a
magazine issue dicussing the curent situaton in
psychiatry. [D&É no. 25, 440448; from La Table ronde.]
Title: ‘La folie, l’absence d’œuvre’. Deleuze urged
Foucault to include this as one of the two appendices for
1972 edition of the big book on madness. The translator
of the forthcoming big book translates this as ‘Madness,
the absence of a work’.
8
In the new preface of 1972,
Foucault explains that he has added two appendices, one
about Derrida, and the other, ‘already published, in which
I discuss a sentence that I had used a little in the dark [un
Mots et les choses, 40 years on
19
8
I will not make any suggestion as to what should be the
correct translation for I am not sure there is one. The
phrase is a problem for one of Foucault’s most careful
readers, Frédéric Gros. He, together with Arnold
Davidson, edited Michel Foucault: Philosophie.
Anthologie, the standard French book of some 900 pages
of selections from Foucault for examination purposes.
[Paris: Gallimard 2004.] Gros writes [p. 35]: ‘Madness
for Foucault is […] absence d’œuvre. And the
“productions” (one hardly dares say “œuvres”) (by
Nietzsche, Artaud, etc) are witnesses of that absence’. [p.
35.] I take it that whatever Frédéric Gros meant, he
wanted to distance himself from the pun on, or play with,
the word ‘œuvre’. Or was there just an intended
equivocation?
peu à l’aveugle], namely “La folie, l’absence d’œuvre”.’
[HF 8.]
9
I do not want to ‘interpret’ this short essay of May
1964, and reprinted as an appendix in 1972. I do not
understand it and I do not like it. It is full of stuff like this
– and I quote from the forthcoming translation:
Since Freud, Western madness has become a nonlanguage
because it has become a double language (a language which
only exists in this speech, a speech that say nothing but its
language) – i.e. a matrix of the language which, strictly
speaking says nothing. A fold of the spoken which is an
absence of work.
Note that the translator here writes ‘absence of work’, not
‘absence of a work’. There are lot of ‘folds’ – plis – in
this text. ‘Speech’ renders parole and ‘language’ renders
langue. The next paragraph runs:
One day it will have to be acknowledged that Freud did not
make speak a madness that had genuinely been a language
for centuries […] he dried it out; he forced its words back to
their source, all the way back to that blank region of auto
implication where nothing is said.
But here ‘language’ renders langage. My first observation
is that the entire essay is totally caught up in the play of
very, very, fanciful talk about what I will just call
‘language’. A strange hyperSaussurianism of the day that
cannot be recognized in anything Saussure ever wrote.
Mots et les choses, 40 years on
20
9
This sentence does not occur in the galley proofs of the
translation. I hope that is just a slip! Or were there 2 prefaces of
1972, one of which is in front of me, and the other of which was
in front of the translator?
That play with words about words was fashionable in
France 40 years ago. It is a type of linguistic philosophy
so at odds with my English philosophy of language that,
let us say, I fail to understand. But the bizarre discussion
of the folds in language is, I claim, important for
understanding the famous conclusion of The Order of
Things.
10 Madness and Man
The essay on the absence d’œuvre opens by saying,
‘Perhaps one day we will no longer know what madness
was.’ That is the last ‘perhaps’ in the piece. ‘Artaud’, we
read two sentences down, ‘will then belong to the
foundation of our language, and not to its rupture […]’.
That’s langage. Artaud has the biggest part in the story
about to unfold. Next comes Foucault’s creation,
Raymond Rousssel. Nietzsche too, of course.
After the initial announcement that madness may
disappear, there is a long speculation about how we will
be seen, in our relation to madness, by this future world.
Maybe near future. ‘In the eyes of I know not which
future culture – and perhaps it is very near […].’ Even in
the second paragraph the imagined future asks:
Why, since the nineteenth century, but also since the
classical age, had [Western culture] clearly stated that
madness was the naked truth of man, only to place it in a
pale, neutralized space, where it almost entirely cancelled
out? Why had it accepted the words Nerval and Artaud, and
recognized itself in their words but not in them?
Mots et les choses, 40 years on
21
But let us pass to the end. The penultimate paragraph says
that madness is separating itself from mental illness
because it ‘gathers itself in the transgressive Fold known
as literature’. This time the Pli is capitalized.
Madness and mental illness are undoing their belonging to
the same anthropological unity. That unity is disappearing,
with man, a passing postulate. Madness, the lyrical halo of
sickness, is ceaslessly dimming its light. And far from
pathology, in language, where it folds in on itself without
yet saying anything, an experience is coming into being
where our thinking is at stake; its imminence, visible
already but absolutely empty, cannot yet be named.
That’s May 1964. On the 5
th
of January 1965, Foucault
watched out the window of his airplane took off from the
Island of Djerba, seeing the curve of the earth at the edge
of the sea. On a postcard he wrote that final sentence that
ends, ‘[…] one can certainly wager that man would be
erased, like a face drawn in sand at the edge of the sea’.
(D&é 34.) But man had been a passing postulate earlier,
the anthropological unity was coming to an end. And
earlier in absence d’œuvre the ‘human sciences’ had been
called a ‘derisory’ classification, lumping psychoanalysis
with linguistics and ethology.
11 Man and language
We all remember that image of the final sentence, the
face drawn in the sand at the edge of the sea. But when I
looked again at the left hand side of the same open page
of The Order of Things, I realized how much I had
suppressed the extent to which claims about language
Mots et les choses, 40 years on
22
dominate the argument for the erasure of man. I shall just
quote the half of a sentence: ‘man is in the process of
perishing as the being of language continues to shine ever
brighter on our horizon’. [OT 386.]
Go back a couple more pages:
From within language experienced and traversed as
language, in the play of its possibilities extended to their
furthest point, what emerges is that man has ‘come to an
end’, [and so on, a sentence getting longer and longer, but
more and more powerful]. [p. 383]
Immediately afterwards we get the usual suspects,
namely the ones who star in absence d’œuvre. Artaud and
Raymond Roussel. I do not find this any more compelling
than I did 35 years ago. But alongside these
hyperlinguistic musings, there is an important story to tell
about man. It does not have Foucault’s ending. Only his
beginning.
Although we remember the last line of the book about
the disappearance of man, the last chapter is titled, ‘The
Human Sciences’. The human sciences (that ‘derisory’
name) are going down the same drain as man. Indeed they
are the real heart of the tale, because, as seen from Paris
in the early 1960s, they are becoming purely linguistic.
12 Anthropology
Foucault submitted two theses for his doctorate, the
big book on madness, and a thesis about Kant’s
Anthropology from a practical point of view. The latter
was never published, and officially the copies deposited
in the usual archives have been lost or stolen. In fact the
Mots et les choses, 40 years on
23
best of authorities told me that he did have a typescript,
but he did not want to show it to anyone for fear it would
be lost or pirated, which would get him in trouble with the
executors, Foucault’s family. When I said in a socalled
seminar in Paris that the text was effectively lost, about
80 of the 160 persons present shouted in one voice, well, I
have a copy. There is even an English translation in free
circulation. Live and learn.
Foucault liked to choose ‘minor’ texts by Kant for
building rather grand analyses – think of ‘What is
Enlightenment? These were not idle or showoff choices:
he was always on to something. There is a curios fact that
in the course of his annual logic lectures Kant, who had
long a list the three fundamental questions, added a fourth
before 1800. The three are, in various versions, What can
we know, What ought we to do, and For what may we
hope? Then he added, ‘What is Man?’ The Anthropology
is not exactly an answer to that question, but it is in the
same ball park. The important point is that questions
about man are not to be answered a priori, by reflection,
but by observation.
The aspect of this, that I personally know best, is the
transition from 18
th
century moral science (that was the
name) to 19
th
century moral science. The former was a
priori theory of man, the latter was empirical, and often
involved counting, and tabulating, and what came to be
caled statistics. That was the beginning of social science.
Man, in the sense to which Foucault pointed, is
inextricable from the sciences of man.
Mots et les choses, 40 years on
24
As is well known, Foucault would have taken such a
truism for what it is: boring. He made us see it in a new
and exciting way: ‘Man and his doubles’ is the title and
the lynchpin of chapter 9. One simple meaning of that
complex play of allusions is that Man is the object of
study and it is Man who does the studying. Of course
there is vastly more, before which one can only be
nonplussed. The way in which Las Meninas, the tour de
force that opens the book, reappears as so central in the
ninth chapter. But allow me to continue reading the book
backwards.
13 Life, labour and language
From time to time I rethink for my own ends
something about the evolution of biology, or of
linguistics. I always astonished by what a rich source
Foucault’s story is, of a fairly sharp creation of new
sciences. I always find illumination. I cannot vouch for
economics because I do not think about it much. One of
the central ideas was that the predecessor fields of
thought, natural history, general grammar, and the theory
of wealth, all ordered their objects in terms of relations of
representation or exchange. The new sciences resulted
from transforming the objects into historical ones.
General grammar became philological, the species came
to have histories. I am not sure this works so well for the
labour theory of value, but that is just ignorance on my
part.
At any rate I am wholly content, except for trivial
details, with the way in which Foucault structured whole
fields of inquiry in ways that were quite unexpected. But
Mots et les choses, 40 years on
25
then comes the part where I lose my grip. Many people
say that the human sciences waste their energies imitating
‘hard’ science. But Foucault has a very complex,
interesting, but I think mistaken, vision of the structure
that the human sciences must acquire, because of their
need to copy the form of the sciences of biology,
philology and economics. I suspect that the argument is
more dense that casual readers notice. Since I now need to
be brief, I shall state one lemma that is essential to the
argument. He first provides his analysis of the model
sciences. And then he states that the human sciences can
have no different organization upon which to work: there
is no space left. Here is the proposition with which he
concludes. It begins with a ‘thus’. I do not think the ‘thus’
holds.
Thus, these three pairs of function and norm, conflict and
rule, signification and system completely cover the entire
domain of what can be known about man. [OT 357]
He makes clear that he is not saying that one human
science adopts one pair, while another chooses another.
All these concepts occur throughout the entire volume
common to the human sciences and are valid in each of the
regions included within it: hence the frequent difficulty in
fixing limits not merely between the objects, but also
between the methods proper to psychology, sociology, and
the analysis of literature and myth.
Mots et les choses, 40 years on
26