Deleuze, Gilles. Bergsonism




Transcript of Deleuze, Gilles. Bergsonism

Page 1: Deleuze, Gilles. Bergsonism
Page 2: Deleuze, Gilles. Bergsonism
Page 3: Deleuze, Gilles. Bergsonism

T r a n s l a t e d b y H u g h T o m l i n s o n

a n d B a r b a r a H a b b e r j a m

Page 4: Deleuze, Gilles. Bergsonism


G i l l e s D e l e u z e

Z O N E B O O K S • N E W Y O R K

1 9 9 1

Page 5: Deleuze, Gilles. Bergsonism

© 1988 Ur/one, Inc.


611 Broadway

New York, NY 10012

All rights reserved.

No part of this book may be reproduced, stored in a

retrieval system, or transmitted in any lorni, or by

any means including electronic, mechanical, photo­

copying, microfilming) recording, or otherwise

(except lor that copying permitted by Sections 107

'and 108 of the U.S. Copyright law and except bv

reviewers for the public press) without written

permission from the Publisher.

Printed in the United States of America

Originally published as Lc Bergsonisme

© 1966 Presses Univcrsitaircs de France

Distributed by The MIT Press

Cambridge, Massachusetts, and London, England

Library of Congress Cataloging in Publication Data

Deleuze, Gilles.

[Bergsonisme. English]

Bergsonism / Gilles Deleu/.e; translated by

Hugh Tomlinson.

p. cm.

Translation of: Bergsonisme.

Bibliography: p.

ISBN 0-942299-06-x ISBN 0-942299-07-8 (pbk.)

1 . Bergson, Henri, 1859-1941. [.Title.

B24 )0 .B4JDJ{13 1988 87-34051

194—<lc i9 C IP

9 412 7 5 4 4

Page 6: Deleuze, Gilles. Bergsonism

C o n t e n t s

Translator's Introduction 7

References to Rerason's Works 11

I Intuition as Method 13

II Duration as Immediate Datum 3 7

III Memory as Virtual Coexistence 51

IV One or Many Durations? 7 3

V HIan Vital as Movement of Differentiation 9 l

.1 Return to Rerc/son 11 5

Sous l l 9

Index 13 7

Page 7: Deleuze, Gilles. Bergsonism

T r a n s l a t o r s ' I n t r o d u c t i o n

Thi s book was originally published in 1966 as part ol a series

ol short studies known as " In i t ia t ion Phi Iosophk |ue ." On lirst

impression, the subject matter appears unpromising. Although

I lenri Bergson was one of the most important and widely read

ph i lo sophe r s o l the lirst decades o l t h e t w e n t i e t h century ,

nowadays his work seems to be almost forgotten. As Kolakowski

says, 'Today 's philosophers, both in thei r research and in their

teaching are almost ent i re ly indifferent to his legacy." 1 Berg­

son i sm is r educed to t h e s ta tus of a f o o t n o t e in h i s to r i e s of

philosophy, making a brief appearance in studies of "v i t a l i sm"

or " i r ra t ional ism."

But this first impression is misleading, I or Deleuze, Bergson

forms part of a "coun te r history" of philosophy. I le was a writer

like Lucre t ius , Spinoza, Hume or Nie tzsche " w h o seemed to

be part of the history of philosophy, but who escaped from it

in one respect or a l toge ther . " 2 In the 1950s and 1960s , it was

writing about philosophers of this kind that enabled Deleuze

to make his escape from the scholas t ic ism of post-war French

academic philosophy. He has descr ibed this task of escaping

the history of philosophy as follows:

Page 8: Deleuze, Gilles. Bergsonism


My way of get t ing out ol it at that t ime , was, I really think,

to conce ive of the history ol philosophy as a kind ol bug­

gery or, what c o m e s to the same thing, immacula te con­

c e p t i o n . I imagined m y s e l f ge t t i ng o n t o the back of an

author, and giving him a ch i ld , which would be his and

which would at the same t ime be a monster. It is very impor­

tant that it should be his ch i ld , because the author actu-

allv had to say everything that I made him say. But it also

had to be a monster because it was necessary to go through

all kinds of decenter ings, slips, break ins, secret emissions,

which 1 really enjoyed. My book on Bergson seems to me

a classic case of t h i s . '

But Bergson is not just an exemplary target lor the philo­

sophical perversion of the early Deleuze . Bergson's work has

provided Deleuze with materials lor his own tool box , lor the

manufacture of his own c o n c e p t s and his own war mach ines .

As he said to Claire Parnet,

Bergson, of course , was also caught up in French-style his­

tory of philosophy, and yet in him there is something which

cannot be ass imi la ted , wh ich enab led him to provide a

shock, to be a rallying point for all the opposition, the objec t

of so many hatreds: and this is not so much because of the

t heme of duration, as ol the theory and pract ice of b e c o m ­

ings ol all kinds, of coex i s t en t mu l t i p l i c i t i e s . 4

Deleuze has himself taken up and transformed these Bergsonian

no t ions in his own errant campaigns lor cons t ruc t ive plural­

ism, recent ly describing himself as an empir ic i s t engaged in

tracing the becomings of which mul t ip l ic i t ies are made up . '


Page 9: Deleuze, Gilles. Bergsonism

T R A N S L A T O R ' S I N T R O D U C T I O N

T h e allinities between Deleuze and Bergson led Ciillian Rose

to speak of his work as " t h e new Bergsonism." 6 But this may

lead to a misunderstanding as Deleuze 's work is character ized

not by a f idel i ty to any master , but by a ser ies of transfor­

mat ions of c o n c e p t s borrowed from a range of wri ters from

many discipl ines . Nevertheless , Deleuze and Bergson do have

a number of important " p r o b l e m s " in c o m m o n . In particular,

Deleuze 's work has been increasingly p reoccup ied with the

p rob lems o f " m o v e m e n t " and " t i m e " wh ich s o c o n c e r n e d

Bergson. His recent isolation of the c inematographic concep t s

of the " m o v e m e n t - i m a g e " and the " t i m e - i m a g e " grows out of

four "commentar ies" on Bergson's notions of movement, image,

recogni t ion and t i m e . 7

T h e t rans la t ion o f the Bergsonian t e rms in t h e b o o k pre­

sents a special difficulty. Bergson's m o t h e r was from the north

of Ing land and he spoke the language from ch i ldhood . Many

of his major works were translated during his l ifetime and per­

sonally revised by h i m . 8 We have not followed the t e rmino l ­

ogy adopted in these translations in three respects .

First, in the authorized translations, the key term "elan Vitaf

is rendered as "vital impetus ." Th i s version is not an entirely

happy one and has of ten been c r i t i c i z e d . T h e F rench word

"clan" has a m u c h b roade r range of sense than t h e F.nglish

" impetus ," from " m o m e n t u m " through "surge" to "vigor." We

have thus followed the pract ice of recent writers on Bergson

and have left "elan vital" in the French. Second , the authorized

t ransla t ions do not make a sys t ema t i c d i s t inc t ion b e t w e e n

" r e c o l l e c t i o n " and " m e m o r y " in the English. We have invari­

ably rendered "souvenir" as " r e c o l l e c t i o n " and "me'moire" as

" m e m o r y " and have al tered ext rac ts from the Bergson transla­

t ions accordingly. Third , the authorized translations have used


Page 10: Deleuze, Gilles. Bergsonism


an English neo log i sm " d e t e n s i o n , " as the i r rendering of the

word "detente." However, this only suggests one ol the range

of senses in which Bergson uses the word, that is " re laxa t ion ,"

in contras t to " c o n t r a c t i o n " (in o ther words, "de- tens ion") -

the N ixon-Brezhnev sense . I t does not , however, convey the

more act ive senses of the word: meaning "spring" or "expan­

s ion . " Bergson often draws on this last sense w h i c h is used

technical ly in thermodynamics to mean the expansion ol a gas

that has been previously sub jec t to pressure. We have there­

fore rendered "detente" by e i the r " r e l axa t ion" or "expans ion"

depending on the c o n t e x t , with the original in parentheses.

We have followed the authorized translations in translating

"duree" as "duration" and adopting "ex tens i ty" and "extension"

to translate Bergson's terms "e'tendue" and "extension." We have

translated bo th "e'eart" and "intervalle" as " in t e rva l " wi th the

F rench word in pa ren theses . D e l e u z e of ten uses Kant ' s dis­

t i n c t i o n b e t w e e n t h e "quaestio quid juris" and the "quaestio

quid facti" be tween the "question de droit" and the "question de

fait:** We have translated "en fait" and "en droit" by "in fact"

and "in pr inc ip le ."

We are grateful to l l rzone , Inc. and particularly to Ramona

Naddaff for suggesting that we translate this book . A number

of friends and colleagues have made suggestions and commen t s

and tried to remind us how English is supposed to read. In par­

t i cu la r we would l ike to thank: Caro l ine Davidson, R o b e r t

Cialeta, Martin Joughin and Richard Wi l l i ams .

I lugh Tomlinson

Barbara I labberjam


D e c e m b e r 1987

1 0

Page 11: Deleuze, Gilles. Bergsonism

R e f e r e n c e s t o B e r g s o n ' s W o r k s

T F Time and Free Will, t rans la ted by F .L . Pogson, L o n d o n :

George Allen & Unvvin Ltd. New York: Macmillan & Co. , 1919.

Essai sur les donnees imme'diatcs de la conscience, 1889 .

MM Matter and Memon, translated by Nancy Margaret Paul and

W. Sco t t Palmer, London: George Allen & Unwin Ltd . , 1911.

Mattered Memoire, 1 8 9 6 .

CE Creative Evolution, translated by Arthur Mitchel l , New York:

Henry Holt & C o . , 1911 (New York: Macmi l lan & C o . , 1 9 4 4 ) .

l.'Evolution creatrice, 1 9 0 7 .

ME Mind-Eneryv, t ranslated by I I . W i l d o n Carr, New York:

Henry Holt & C o . , 1 9 2 0 . L'Eneryicspiritucllc, 1919.

DS Duration and Simultaneity, translated by Leon J acobson ,

Indianapolis: Bobbs-Merr i l l , 1 9 6 5 . Duree ct Simultaneity, 1922 .

MR The Two Sources of Morality and Religion, translated by R.

Ashley Audra and Cloudeslev Brc rc ton with the assistance ol

VY. 1 lorslall Carter. New York: Henry Holt & C o . , 1 9 3 5 . Les deux

sources de la morale et de la religion, 1 9 3 2 .

CM The Creative Mind, t r ans la ted by M a b e l l e L. And i son ,

Westport, C o n n e c t i c u t : Greenwood Press, 1 9 4 6 . La Pensc'e ct

lcMouvant, 1941.


Page 12: Deleuze, Gilles. Bergsonism


References to the original French are in parentheses. The 1)S

references are to the 4th Edi t ion . For all the o the r works, the

French references are, first, to the Centenary Edit ion (Presses

LIniversitaires de France) , and then to the 19 39 -1941 reprints.


Page 13: Deleuze, Gilles. Bergsonism


I n t u i t i o n a s M e t h o d

Duration, Memory, lllan Vital mark the major stages of Bergson's

philosophy. T h i s b o o k sets ou t to d e t e r m i n e , first, the rela­

tionship between these three notions and, second, the progress

they involve.

Intuition is the m e t h o d of Bergsonism. Intui t ion is ne i ther

a feeling, an inspiration, nor a disorderly sympathy, but a fully

developed me thod , one ol the most lullv developed methods

in philosophy. It has its s t r ic t rules, cons t i tu t ing that which

Bergson calls "precision" in philosophy. Bergson emphasizes this

point: Intuition, as he understands it methodologically, already

presupposes duration. "These conclusions on the subject ol dura­

t ion were, as it seemed to m e , decisive. S tep by step they led

me to raise intui t ion to the level of a phi losophical me thod .

I he use of the word intuition, however, caused me some degree

of hesitation." 1 And to llotfding, he writes: " T h e theory of intu­

ition which you stress more than that of duration only became

clear to me long afterwards." 2

But first and second have many meanings. Intuition certainly

is second in relation to duration or to memory. But while these

not ions by themselves deno t e lived reali t ies and expe r i ences .

• J

Page 14: Deleuze, Gilles. Bergsonism


they do not give us any means ol knowing (connaitrc) them with

a precision analogous to that ol science. We might say, strangely

enough, that duration would remain purely intuit ive, in the

ordinary sense of the word , i f i n tu i t ion — in t h e proper ly

Bergsonian sense — were not there as me thod . T h e lact is that

Bergson relied on the intuit ive me thod to establish philoso­

phy as an absolutely "p rec i se" discipl ine, as precise in its field,

as capable ol being prolonged and transmitted as sc ience itself

is. And wi thou t the methodica l thread ol intui t ion, the rela­

t ionsh ips b e t w e e n Dura t ion , M e m o r y and Ulan Vital would

themselves remain inde te rmina te from the point ol view ol

knowledge. In all of these respects , we must bring intui t ion

as rigorous or precise method to the forefront of Our discussion. 5

T h e most general methodolog ica l quest ion is this: How is

intuition — which primarily denotes an immediate know ledge

(connaissance) — capable of forming a method, once it is accepted

that the method essentially involves one or several mediations?

Bergson often presents intuition as a simple act. But, in his view,

s impl ic i ty does not e x c l u d e a qual i ta t ive and virtual mul t i ­

plicity, various d i rec t ions in which it c o m e s to be actual ized.

It is in this sense, then, that intui t ion involves a plurality of

meanings and i r r educ ib le mu l t i p l e a s p e c t s . 4 Bergson dis t in­

guishes essentially three distinct sorts of acts that in turn deter­

mine the rules of the me thod : T h e lirst conce rns the stating

and creat ing ol problems; the second , the discovery ol genu­

ine differences in kind; the third, the apprehension ol real t ime.

It is by showing how we move from one meaning to another

and what the "fundamental meaning" is, that we are able to

red iscover the s imp l i c i ty ol in tu i t ion as l ived ac t , and thus

answer the general methodolog ica l ques t ion .

* • *

' 4

Page 15: Deleuze, Gilles. Bergsonism


FIRST R U L E : Apply the test of true and false to problems themselves.

Condemn false problems and reconcile truth and creation at the level

of problems.

We are wrong to bel ieve that the true and the false can only

be brought to bear on solutions, that they only begin with solu­

tions. This prejudice is social (lor society, and the language that

t ransmi t s i ts o rder -words [mots d'ordre], " s e t u p " [donnent]

ready-made problems, as if they were drawn out of " t h e city's

administrative filing c a b i n e t s , " and force us to " so lve" t hem,

leaving us only a thin margin of freedom). Moreover, this preju- '

dice goes back to ch i ldhood, to the classroom: It is the school

teacher who "poses" the problems; the pupil's task is to dis­

cover the solut ions . In this way we are kept in a kind of slav­

ery. True f reedom l ies in a power to d e c i d e , to c o n s t i t u t e

problems themselves. And this "semi-divine" power entails the

disappearance of false problems as much as the creative upsurge

of true ones . " T h e truth is that in philosophy and even else­

where it is a question of finding the problem and consequent ly

of positing it, even more than of solving it . For a speculat ive

problem is solved as soon as it is properly stated. By that 1 mean

that its solut ion exists then, although it may remain hidden

and, so to speak, covered up: The only thing left to do is to

uncover it. But stating the problem is no t s imply uncovering,

it is inventing. Discovery, or uncovering, has to do with what

already exis ts , actually or virtually; it was therefore cer tain to

happen sooner or later. Invention gives being to what did not

exist ; it might never have happened. Already in ma themat ics ,

and still more in metaphysics , the effort of invention consis ts

most often in raising the p rob lem, in creat ing the t e rms in

which i t will be stated. T h e stating and solving of the prob-

' 5

Page 16: Deleuze, Gilles. Bergsonism


I cm are here very c l o se to be ing equivalent : T h e truly great

problems are set forth only when they are so lved ." 5

It is not just the whole history of ma themat i c s that sup­

ports Bergson. We might c o m p a r e the last s en t ence of this

ex t rac t from Bergson with Marx 's formulation, which is valid

for pract ice itself: "Humani tv only sets itself problems that it

is capable of solving." In ne i the r example is it a case of saying

that problems are like the shadow of pre-existing solutions ( the

whole con tex t suggests the contrary) . Nor is it a case of*saying

that only the problems coun t . On the contrary, it is the solu­

t ion that coun t s , but the problem always has the solut ion it

deserves, in t e rms of" the way in which it is stated ( i . e . , the

condi t ions under which it is de te rmined as p r o b l e m ) , and of

the means and terms at our disposal for stating it. In this sense,

the history ol man, from the theore t ica l as much as from the

practical point of view is that of the const ruct ion of problems.

It is here that humanity makes its own history, and the b e c o m ­

ing consc ious of that act ivi ty is like the conques t of freedom.

( I t is true that, in Bergson, the very not ion of the problem has

its roots beyond history, in life itself or in the elan vital: Life is

essentially de te rmined in the act of avoiding obs tac les , stat­

ing and solving a problem. T h e cons t ruc t ion ol the organism

is bo th the stating of a p roblem and a s o l u t i o n . ) 6

But how can this cons t i tu t ive power which resides in the

problem be r econc i l ed with a norm ol the true? W h i l e it is

relatively easy to de l ine the true and the false in relation to

solut ions whose problems have already been stated, i t seems

much more difficult to say in what the true and the false con­

sist when applied to the process of stating problems. Th i s is

how many philosophers fall into circular arguments: Conscious

ol the need to take the test of true and false beyond solut ions

Page 17: Deleuze, Gilles. Bergsonism


into problems themselves, they are conten t to define the truth

or falsity of a problem by the possibility or impossibility of its

being solved. Bergson's great vir tue, on the o ther hand, is to

have a t tempted an intrinsic de te rmina t ion of the false in the

expression "false p rob lem." This is the source of a rule that is

complemen ta ry to the preceding general rule.

COMPLEMENTARY RULE: Take problems arc of two sorts, "nonexistent

problems," defined as problems whose very terms contain a confusion

of the "more" and the "less"; and "badh stated" questions, so defined

because their terms represent badlv analv/ed composites.

Ib il lustrate the first kind of problem Bergson c i t e s the prob­

lems of 'non being, of disorder or of the possible ( the problems

of knowledge and being); as examples of the second type, there

are the problems of freedom or of intensi ty . 7 His analyses of

these are famous. In the first case, they consist in showing that

there is not less, but more in the idea of nonbeing than that of

being, in disorder than in order, in the possible than in the real.

In the idea of nonbeing there is in fact the idea of being, plus

a logical operat ion ol generalized negat ion, plus the particu­

lar psychological mot ive for that opera t ion (such as when a

being does not correspond to our expec ta t ion and we grasp it

purely as the lack , the absence ol what in teres ts us) . In the

idea ol disorder there is already the idea ol order, plus its nega­

t ion, plus the mot ive lor that negat ion (when we e n c o u n t e r

an order that is not the one we e x p e c t e d ) . And there is more

in the idea ol the possible than there is in the idea of the real:

' T o r the possible is only the real with the addition ol an act

ol mind that throws its image back into the past o n c e it has

been e n a c t e d , " and the mot ive of that act (when we confuse

' 7

Page 18: Deleuze, Gilles. Bergsonism


the upsurge of a reality in the universe wi th a success ion ol

states in a c losed s y s t e m ) . 8

W h e n we ask " W h y is there something rather than noth­

ing?" or " W h y is there order rather than disorder?" or " W h y

is there this rather than that (when that was equally possible)?"

we fall in to the same error: We mistake the more lor the less,

we behave as though nonbeing existed before being, disorder

before order and the |*>ssible before exis tence. As though being

c a m e to lill in a void, order to organize a preceding disorder,

the real to realize a primary possibility. Being, order or the exis­

ten t are truth itself; but in the false problem there is a funda­

mental illusion, a "retrograde movement ol the true," according

to which being, order and the existent are supposed to pre­

cede themselves, or to precede the creative act that const i tutes

t hem, by pro jec t ing an image of themselves back into a possi­

bility, a disorder, a nonbeing which are supposed to be primor­

dial. This t h e m e is a central o n e in Bergson's philosophy: It

sums up his c r i t ique of the negative and of negat ion, in all its

forms as sources of false p rob lems .

Badly stated problems, the second type of false p rob lem,

introduce a different mechanism: This t ime it is a case of badly

analyzed compos i t e s that arbitrarily group things that differ in

kind. Take for example , the quest ion ol whe ther happiness is

reducible to pleasure or not : Perhaps the te rm pleasure sub­

sumes very varied i rreducible states, just like the idea of hap­

piness. II the terms do not correspond to "natural articulations"

then the problem is lalse, lor it does not affect " the very nature

ol things.'"' I lere again, Bergson's analyses are famous: for exam­

ple, the one in which he c o n d e m n s intensi ty as such a c o m ­

posi te . W h e t h e r the quality ol the sensation is confused with

the muscular space that corresponds to it, or with the quan-


Page 19: Deleuze, Gilles. Bergsonism


tity ol the physical cause that produces it, the not ion of inten­

sity involves an impure mix tu re be tween de terminat ions that

differ in kind, so that the question "by how much does the sen­

sation g row?" always goes back to a badly stated p r o b l e m . 1 0

l ikewise the problem of freedom, in which two types ol "multi­

p l i c i ty" are contused: that ol terms juxtaposed in space and

that of states which merge together in duration.

Let us return to the lirst type of false problem. I lere, Berg­

son says, the more is mistaken for the less. But there are also

t imes when Bergson says that the less here is mistaken lor the

more : just as doubt about an act ion only apparently adds to

the act ion, when in reality it indicates a half-willing; negation

is not added to what it denies, but only indicates a weakness in

the person who denies. " F o r we feel that a divinely crea ted will

or thought is t o o full of itself, in the immensi ty of its reality,

to have the slightest idea of a lack of order or a lack of being.

To imagine the possibili ty of absolute disorder, all the more

the possibili ty of nothingness , would be tor i t to say to itself

that it might not have existed at all, and that would be a weak­

ness incompat ib le with its nature, which is force It is not

something more but something less; it is a deticit ol the wil l ." 1 1

Is there a contradiction between these two formulations, where

nonbeing is somet imes presented as a more in relation to being

and some t imes as a less? T h e r e is no cont rad ic t ion if we bear

in mind that what Bergson is condemning in nonexistent prob­

lems is the obsession in all its aspects wi th thinking in terms of

more and less. T h e idea of disorder appears when, instead of

seeing that there are two or more irreducible orders (for exam­

ple, that of life and that of mechanism, each present when the

other is absent ) , we retain only a general idea of order that we

conf ine ourselves to opposing to disorder and to thinking in


Page 20: Deleuze, Gilles. Bergsonism


cor re la t ion wi th the idea of disorder. T h e idea of nonbe ing

appears when, instead of grasping the different reali t ies that

are indefinitely substi tuted lor one another, we muddle them

toge ther in the homogene i ty of a Being in general , which can

only IK- opposed to nothingness, be related to nothingness. T h e

idea of the possible appears when, instead of grasping each exis­

ten t in its novelty, the whole of e x i s t e n c e is related to a pre­

formed e lement , from which everything is supposed to emerge

by s imple "real izat ion."

In short, each t i m e that we think in terms of more or less,

we have already disregarded the differences in kind between the

t w o orders, or be tween beings, be tween exis tents . In this way

we can see how the first type of false problem rests, in the final analy­

sis, on the second: T h e idea ol disorder emerges from a general

idea of order as badly analyzed composi te , e t c . And conceiving

everything in terms of more and less, seeing nothing but dif­

ferences in degree or differences in intensity where , more pro­

foundly, there are differences in kind is perhaps the most general

error of thought, the error c o m m o n to sc ience and metaphysics.

We are therefore vict ims of a fundamental illusion that cor­

responds to the t w o aspects of the false p rob lem. T h e very

no t ion of the false problem indeed impl ies that we have to

struggle not against simple mistakes (false solutions), but against

something more profound: an illusion that carries us along, or

in which we are immersed , inseparable from our cond i t ion .

A mirage, as Bergson describes the project ion backward of the

possible. Bergson borrows an idea from Kant although he com­

pletely transforms it: It was Kant who showed that reason deep

within i tself engenders not mistakes but inevitable illusions, only

the effect of wh ich cou ld be warded off. Although Bergson

de te rmines the nature of false problems in a comple t e ly dil-

2 0

Page 21: Deleuze, Gilles. Bergsonism


lerent way and although the Kantian cr i t ique i t se l f seems to

him to be a co l l e c t i on of badly stated problems, he treats the

illusion in a way similar to Kant. T h e illusion is based in the

deepest part of the intel l igence: It is not , strictly speaking, dis­

pelled or dispel lable , rather it can only be repressed.^2 We tend

to think in terms of more and less, that is, to see d i l l e rcnccs

in degree where there are differences in kind. We can only react

against this intel lectual tendency by bringing to life, again in

the intelligence, another tendency, which is critical. But where,

precisely, does this s econd tendency c o m e from? Onlv intu­

ition can produce and act ivate it, because it rediscovers dil­

terences in kind beneath the differences in degree, and conveys

to the in te l l igence the cr i ter ia that enable i t to dist inguish

be tween true and false p rob lems . Bergson shows clearly that

the in te l l igence is the faculty that states problems in general

( the instinct is rather a faculty for finding s o l u t i o n s ) . " But only

intuition decides be tween the true and the false in the prob­

lems that are stated, even if this means driving the intel l igence

to turn back against itself.

* * *

S E C O N D R U L E : Struggle against illusion, rediscover the true differ­

ences in kind or articulations of the real.14

I he Bergsonian dual i sms are famous: d u r a t i o n - s p a c e , qual­

ity-quantity, heterogeneous-homogeneous, continuous-discon­

tinuous, the two mult ipl ici t ies, memory-mat ter , r eco l l ec t ion-

pe rcep t ion , c o n t r a c t i o n - r e l a x a t i o n * (detente), i n s t i n c t - i n t e l ­

l igence , the t w o sou rces , e t c . Even the running heads that

KTor a discussion of the problem of translating detente, see Preface (Trans.).


Page 22: Deleuze, Gilles. Bergsonism


Bergson puts at the top ol each page ol his hooks indicate his

taste lor dualisms - which do not , however, have the last word

in his philosophy. W h a t , there-lore, do they mean? According

to Bergson, a c o m p o s i t e must always he divided accord ing to

its natural ar t iculat ions, that is, in to e l e m e n t s which differ in

kind. Intui t ion as method is a method ol division, P la tonic in

inspiration. Bergson is aware that things are mixed together in

reality; in (act, exper i ence itsell oilers us nothing but compos­

i tes . But that is not where the difficulty l ies. For example , we

make t ime into a representation imbued with space. I he awk­

ward thing is that we no longer know how to distinguish in

that representation the two c o m p o n e n t e l emen t s which differ

in kind, the t w o pure presences ol duration and extensi ty. W e

mix extensity and duration so thoroughly that we can now only

oppose their mix tu re to a principle that is assumed to be both

nonspatial and nontemporal , and in relation to which space and

t ime , duration and extensity, are now only de te r io ra t ions . 1 5 To

take yet another example , we mix recol lect ion and perception;

but we do not know how to recognize what goes back to per­

cept ion and what goes back to reco l lec t ion . We no longer dis­

t inguish the t w o pure p r e sences ol m a t t e r and m e m o r y in

representat ion, and we no longer see anything but differences

in degree between percept ion-recol lec t ions and reco l l ec t ion-

perceptions. In short, we measure the mixtures with a unit that

is itsell impure and already mixed . We have lost the ground

of composi tes . T h e obsession with the pure in Bergson goes back

to this restoration ol differences in kind. Only that which dif­

fers in kind can be said to be pure, but only tendencies differ in

kind." ' T h e compos i t e must therefore be divided according to

quali tat ive and qualified tendenc ies , that is, according to the

way in which it c o m b i n e s duration and ex tens i ty as they are

2 2

Page 23: Deleuze, Gilles. Bergsonism


2 3

defined as movements, d i rec t ions of movements ( h e n c e dura­

t ion-con t rac t ion and mat ter-expansion [detente]). Again, there

is some resemblance be tween intui t ion as method of division

and transcendental analysis: If the composi te represents the fact,

it must be divided into tendencies or into pure presences that

only exis t in principle (en droit).17 W e go beyond e x p e r i e n c e ,

toward the condi t ions of expe r i ence (but these are not , in the

Kantian manner, the condit ions of all possible experience: They

are the condi t ions of real e x p e r i e n c e ) .

This is the Bergsonian le i tmotif : People have seen only dif-

ferences in degree where there are differences in k ind . And

Bergson groups his major c r i t iques , which take many differ­

ent forms, under this heading. His fundamental c r i t i c i sm of

metaphys ics is that it sees d i f ferences in degree be tween a

spatialized t ime and an eternity which it assumes to be primary

( t ime as de ter iora t ion , re laxat ion [detente] or d iminut ion of

b e i n g . . . ) : All beings are defined on a scale of intensity, between

the two ex t remes of perfection and nothingness. But he directs

a similar cri t icism at sc ience; there is no definition of mechanism

other than that which invokes a spatialized t ime, according to

which beings no longer present anything but differences of

degree, of position, of dimension, of proportion. There is even

" m e c h a n i s m " in evolut ionism, to the ex ten t that it postulates

a unilinear evolution and takes us from one living organization

to another by simple intermediaries, transitions and variations

ol degree . T h e w h o l e sou rce of the false p rob lems and the

illusions that overwhelm us lies in this disregard for true dif­

ferences in k ind: As early as the first chap te r oi Matter and

Memory, Bergson shows how the forgett ing of differences in

kind — on the one hand be tween percept ion and affection, on

the o the r hand be tween percept ion and r eco l l ec t ion — gives

Page 24: Deleuze, Gilles. Bergsonism


rise to all kinds ol falsi- problems by making us think that our

percept ion is inextensive in charac ter : ' T h e r e are, in the idea

that we project outside ourselves states which are purely inter­

nal, so many misconcep t ions , so many lame answers to badly

stated quest ions " I K

No text shows more clearly than this lirst chap te r oi Matter

and Memory how c o m p l e x the manipulat ion of intui t ion is as a

me thod of division. T h e representat ion has to be divided into

the e l emen t s that condi t ion it, in to pure presences or tend­

encies that differ in kind. I low does Bergson proceed? He asks,

first, be tween what two things there may be (o r may not b e )

a difference in kind. His lirst response is that, s ince the brain

is an " i m a g e " among o ther images, or ensures certain move­

ments among o the r movements , there cannot he a difference in

kind between the faculty of the brain which is said to be per­

cep t ive and the reflex functions of the c o r e . Thus, the brain

does not manufacture representat ions, but only compl i ca t e s

the relationship between a received movement (exc i ta t ion) and

an executed movement ( response) . Between the two, it estab­

l ishes an interval (ecart)., w h e t h e r it divides up the r ece ived

movemen t infinitely or prolongs it in a plurality of possible

react ions . Hven if r eco l l ec t ions take advantage- ol this interval

or, strictly speaking, "interpolate themselves," nothing changes.

We can, lor the m o m e n t , d i scount them as being involved in

another " l i ne . " On the line that we are tracing, we only have,

we can only have mat ter and movemen t , movemen t which is

more or less c o m p l i c a t e d , more or less delayed. T h e whole-

quest ion is knowing whether, in this way, we also alre-ady have-

perception. By virtue ol the cerebral interval, in effect, a being

can retain Irom a material o b j e c t and the ac t ions issuing from

it only those e l e m e n t s that interest h im. 1 ' ' So that percep t ion

Page 25: Deleuze, Gilles. Bergsonism


is not the o b j e c t plus someth ing , hut the o b j e c t minus some­

thing, minus everything that does not interest us. It could be

said that the o b j e c t itsell merges with a pure virtual percep­

tion, at the same t i m e as our real percept ion merges with the

o b j e c t from which it has abstracted only that which did not

in teres t us. H e n c e Bergson 's famous thes is ( the lull c o n s e ­

quences ol which we will have to analyze): We perceive things

where they are, perception puts us at once into matter, is imper­

sonal, and co inc ides with the perceived ob j ec t . Cont inuing on

this same l ine , the whole ol Bergson's me thod consis ts , lirst

of all, in seeking the terms between which there could not be a

difference in kind: T h e r e cannot be a difference in kind, but

only a difference in degree be tween the faculty of the brain

and the function of the c o r e , be tween the percept ion of mat •

ter and mat te r itself.

We are now in a position to trace out the second line, which

differs in kind from the first. In order to establish the first we

needed fictions: We assumed that the body was l ike a pure

mathemat ica l point in space, a pure instant, or a succession

ol instants in t ime . But these fictions were not simply hypoth­

eses: They consisted in pushing beyond exper ience a direct ion

drawn from exper ience itself, ft is only in this way that we can

extract a whole aspect of the condit ions of exper ience . All that

is left now is to ask ourselves what fills up the cerebra l inter­

val, what takes advantage of it to b e c o m e embod ied . Bergson's

response is three-told, f i rs t , there is allectivitv, which assumes

that the body is some th ing o the r than a mathemat ica l point

and which gives it vo lume in space. Nex t , it is the r eco l l ec ­

tions ol memory that link the instants to each o ther and inter­

polate the past in the present, finally, it is memory again in

a n o t h e r form, in the form Of a c o n t r a c t i o n ol m a t t e r that

Page 26: Deleuze, Gilles. Bergsonism


makes the quality appear. (It is therefore memory that makes

the body s o m e t h i n g o t h e r than ins tan taneous and gives i t a

duration in t i m e ) . We are consequen t ly in the presence ol a

new l ine , that ol subjec t iv i ty , on which affcctivity, r e c o l l e c ­

t i o n - m e m o r y , and c o n t r a c t i o n - m e m o r y are ranged: T h e s e

terms mav be said to differ in kind from those of the preced­

ing line ( p e r c e p t i o n - o b j e c t - m a t t e r ) . 2 0 In short, representation

in general is divided into two d i rec t ions that differ in kind,

into two pure presences that do not al low themselves to be

represented: that o f perception which puts us at once in to mat­

te r and that o f memory which puts us at once in to the mind .

O n c e again, the quest ion is no t whe the r the t w o lines mee t

and mix together . Th i s m ix tu re is our e x p e r i e n c e i tsel l , our

representat ion. But all our false problems derive from the fact

that we do not know how to go beyond exper ience toward the

condi t ions o f expe r i ence , toward the ar t iculat ions o f the real,

and rediscover what differs in kind in the compos i t e s that arc-

given to us and on which we live. T h e s e two acts , percept ion

and reco l l ec t ion , "always interpenetrate each other , are always

exchanging something of their substance as by a process of end-

osmosis . T h e proper office of psychologists would be to dis­

socia te t hem, to give back to each its natural purity; in this

way many d i f f icu l t ies raised by psychology and perhaps also

by metaphysics might be lessened. But they will have it that

these mixed states, c o m p o u n d e d , in unequal proport ions, of

pure pe rcep t ion and pure r e c o l l e c t i o n , are s imp le . And so

we are c o n d e m n e d to an ignorance alike of pure r eco l l ec t i on

and of pure percep t ion , to knowing only a single kind of phe­

nomenon that will be cal led now reco l l ec t ion and now per­

cep t ion , accord ing to the predominance in i t ol one or o the r

o l the t w o a s p e c t s ; and, c o n s e q u e n t l y , to finding b e t w e e n

2 6

Page 27: Deleuze, Gilles. Bergsonism


p e r c e p t i o n and r e c o l l e c t i o n on ly a d i f ference in degree and

not in k ind . " 2 1

In tu i t ion leads us to go beyond the s tate of e x p e r i e n c e

toward the condi t ions of exper ience . But these condi t ions are

ne i ther general nor abstract . They are no broader than the con-

di t ioned: they are the condi t ions of real expe r i ence . Bergson

speaks ol going "to seek experience at its source, or rather above

that decis ive turn, where , taking a bias in the d i rec t ion of our

utility, i t b e c o m e s properly human e x p e r i e n c e . " 2 2 Above the

turn is precisely the point at which we finally discover differ­

ences in kind. But there are so many difficulties in trying to

reach this focal point that the acts of intuition, which are appar­

ently contradictory, have to be multiplied. Bergson, thus, some­

t imes speaks of a m o v e m e n t that is exact ly appropriate to the

expe r i ence , some t imes a broadening out , s o m e t i m e s a tight­

ening and narrowing. For, in the first place, the determinat ion

of each " l i n e " involves a sort of con t rad ic t ion in which appar­

ently diverse facts are grouped according to their natural affini­

t ies , drawn toge ther accord ing to the i r ar t icula t ion. But , on

the other hand, we push each line beyond the turn, to the point

where i t g o e s beyond our own e x p e r i e n c e : an extraordinary

broadening out that forces us to think a pure percept ion iden­

tical to the whole of matter , a pure memor y identical to the

total i ty of the past. It is in this sense that Bergson on several

occas ions compares the approach of philosophy to the proce­

dure of infini tesimal ca lcu lus : W h e n we have benef i t t ed in

expe r i ence from a l i t t le light which shows us a l ine of articu­

lation, all that remains is to extend it beyond exper ience — just

as mathemat ic ians recons t i tu te , with the infinitely small ele­

ments that they perce ive of the real curve, " t h e curve itsell

s t re tching out into the darkness behind t h e m . " 2 3 In anv case,

2 7

Page 28: Deleuze, Gilles. Bergsonism


Bergson is not o n e of those philosophers who ascribes a prop­

erly human wisdom and equi l ib r ium to philosophy. To open

us up to the inhuman and the superhuman (durations which are

inferior or superior to our o w n ) , to go beyond the human con­

dit ion: This is the meaning of philosophy, in so far as our condi­

t ion condemns us to live among badly analyzed c o m p o s i t e s ,

and to be badly analyzed compos i t e s ou r se lves . 2 4

But this broadening out , or even this going-beyond does not

consis t in going beyond exper ience toward concep t s . For con­

cep t s only define, in the Kantian manner, the condi t ions of all

possible e x p e r i e n c e in general . Here , on the o the r hand, i t is

a case of real exper ience in all its peculiari t ies. And if we must

broaden it, or even go beyond it, this is only in order to find

the ar t iculat ions on which these peculiar i t ies depend. So that

the condi t ions of expe r i ence are less de te rmined in c o n c e p t s

than in pure p e r c e p t s . 2 5 And, while these percepts themselves

are uni ted in a c o n c e p t , it is a c o n c e p t mode led on the thing

itself, which only suits that thing, and which , in this sense, is

no broader than what it must accoun t for. For when we have

followed each of the " l i n e s " beyond the turn in e x p e r i e n c e ,

we must also rediscover the point at which they intersect again,

where the d i rec t ions cross and where the t endenc ies that dif­

fer in kind link toge ther again to give rise to the thing as we

know it. ft might be thought that noth ing is easier, and that

expe r i ence i t s e l f has already given us this point . But it is not

as simple as that. After we have followed the lines of d ivergence

beyond the turn, these lines must intersect again, not at the point

from which we started, but rather at a virtual point, at a virtual

image of the point of departure, which is itself located beyond

the turn in exper ience; and which finally gives us the sufficient

reason of the thing, the sufficient reason of the compos i t e , the

2 8

Page 29: Deleuze, Gilles. Bergsonism


sufficient reason of the point of departure. So that the expres­

sion "beyond the dec is ive turn" has two meanings: First , i t

denotes the moment when the lines, setting out from an uncer­

tain c o m m o n point given in expe r i ence , diverge increasingly

according to the differences in kind. T h e n , it denotes another

m o m e n t when these l ines converge again to give us this t i m e

the virtual image or the dist inct reason of the c o m m o n point .

Turn and return. Dualism is therefore only a m o m e n t , which

must lead to the re-formation of a m o n i s m . Th i s is why, after

the broadening out , a final narrowing follows, just as integra­

t ion follows dif ferent ia t ion. " W e have alluded e l sewhere to

those ' l ines of fact," each one indicating but the d i rec t ion of

truth, because it does not go far enough: Truth itself, however,

will be reached if two of them can be prolonged to the point ( ,

where they in tersec t In our opinion this me thod of inter­

sect ion is the only one that can bring about a decisive advance

in me taphys ic s . " 2 6 T h e r e are, therefore , t w o successive turns

in exper i ence as it were, both in a reverse di rect ion: They con­

st i tute what Bergson calls precision in philosophy.

Hence, a C O M P L E M E N T A R Y R U L E to the second rule: The real is

not only that which is cut out fse decoupe ) according to natural articu­

lations or differences in kind; it is also that which intersects again Ise

recoupe) along paths converging toward the same ideal or virtual point.

T h e particular function of this rule is to show how a prob lem,

when it is properly stated, tends to be solved of its own accord.

For example , still in the lirst chapter of Matter and Memory, the

problem of memory is correct ly stated, s ince, starting from the

pe rcep t ion- reco l lec t ion compos i t e , we divide this compos i t e

into two divergent and expanded di rec t ions which correspond

2 9

Page 30: Deleuze, Gilles. Bergsonism


to a true difference in kind be tween soul and body, spirit and

mat ter . But we can only reach the solut ion to the problem by

narrowing: W h e n we attain the original point at which the two

divergent directions converge again, the precise point at which

recol lec t ion inserts i tself into perception, the virtual point that

is like the reflection and the reason of the departure point. Thus

the problem of soul and body, of matter and spirit, is only solved

by an ex t reme narrowing in which Bergson shows how the lines

of ob jec t iv i ty and of subject ivi ty, the lines of external obser­

vation and of internal expe r i ence , must converge at the end

ol the i r different processes, all the way to the case of aphasia . 2 7

Bergson shows, similarly, that the problem of the immor­

tality of the soul tends to be solved by the convergence of two

very different l ines: that of an expe r i ence of m e m or y and that

of a qui te different, myst ical , e x p e r i e n c e . 2 8 T h e problems that

are unraveled at the point at which three lines of facts converge

are even m o r e c o m p l e x : Such is the nature of consc iousness

in the first chapter of Mind-Energy. It should be noted that this

method of in tersect ion forms a genuine probabil ism: bach line

defines a probabi l i ty . 2 9 But it is a qualitative probabil ism, lines

of fact being quali tat ively d i s t inc t . In thei r divergence, in the

disart iculat ion of the real that they brought about according

to the differences in kind, they already cons t i tu ted a superior

empi r i c i sm, capable of stating p rob lems and of going beyond

exper ience toward conc re t e condi t ions . In their convergence ,

in the intersection of the real to which they proceed, they now

define a superior probabilism, one capable of solving problems

and of bringing the condi t ion back to the cond i t ioned so that

no dis tance remains be tween t h e m .

3 0

Page 31: Deleuze, Gilles. Bergsonism


I HIRD R U L E : State problems and solve them in terms of time rather

than of space.50

This rule gives the "fundamental meaning" of intuit ion: Intui­

t ion presupposes duration, it consists in thinking in t e rms of

duration. 1 1 We can only understand it by returning to the move­

ment of the division de termining the differences in kind. At

lirst sight it would seem that a difference in kind is established

be tween t w o things, o r rather be tween t w o t endenc ies . This

is t rue, but only superficially. Let us cons ide r the principal

Bergsonian division: that be tween duration and space . All the

o the r divisions, all the o t h e r dualisms involve it, derive from

it, or result in it. Now, we cannot s imply conf ine ourselves to

affirming a difference in kind between duration and space. The

division occurs between ( I ) duration, which "tends" for its part

to take on or bear all the d i f ferences in kind ( b e c a u s e i t is

endowed with the power of qualitatively varying with i tself) ,

and ( 2 ) space, which never presents anything but differences

of degree ( s ince i t is quantitative homogene i ty ) . T h e r e is thus

not a difference in kind be tween the two halves of the divi­

sion; the quali tat ive difference is ent i re ly on one side. W h e n

we divide someth ing up accord ing to its natural ar t iculat ions

(as with proport ions and figures that vary greatly from case to

case) , we have: on the one hand, the aspect of space, by which

the thing can only ever dilfer in degree from o ther things and

from itself (augmenta t ion , d iminu t ion) ; and on the o the r hand,

the aspect of duration, by which the thing differs in kind from

all o thers and from itself ( a l t e ra t ion) .

lake a lump of sugar: It has a spatial configurat ion. But il

we approach it from that angle, all we will ever grasp are dif­

ferences in degree between that sugar and any o ther thing. But

Page 32: Deleuze, Gilles. Bergsonism


it also has a duration, a rhythm of duration, a way of being in

t ime that is at least partially revealed in the process of its dis­

solving, and that shows how this sugar differs in kind not only

from other things, but lirst and foremost from itself. Th i s alter­

ation, which is one with the essence or the substance of a thing,

is what we grasp when we conce ive of it in terms of Duration.

In this respect, Bergson's famous formulation, "I must wait until

the sugar dissolves" has a still broader meaning than is given

to it by its c o n t e x t . 3 2 It signilies that my own duration, such

as I live it in the impa t ience of waiting, for example , serves to

reveal o the r durations that beat to o the r rhythms, that differ

in kind from m i n e . Dura t ion is always the l o c a t i o n and the

environment of differences in kind; it is even their totality and

mult ipl ic i ty . T h e r e are no differences in kind e x c e p t in dura­

t ion — whi l e space is no th ing o t h e r than the l o c a t i o n , the

env i ronment , the total i ty of differences in degree .

Perhaps we now have the means to resolve the mos t gen­

eral of methodolog ica l ques t ions . W h e n Pla to formulated his

method of division, he t o o intended to divide a compos i te into

two halves, or along several lines. But the whole problem lay

in knowing how to c h o o s e the right half: W h y was what we

were looking for on one side rather than on the o ther? Divi­

s ion c o u l d t he re fo re be c r i t i c i z e d for no t be ing a g e n u i n e

method s ince it lacked a "middle t e r m " and still depended on

an inspiration. In Bergsonism, the difficulty seems to disappear.

For bv dividing the c o m p o s i t e accord ing to two t endenc ies ,

with only o n e showing the way in which a thing varies quali­

tatively in t i m e , Bergson effectively gives h imse l f the means

of choos ing the "right s ide" in each case; that of the essence .

In short , in tui t ion has b e c o m e m e t h o d , or rather me thod has

been reconc i led with the immedia te . Intuit ion is not duration

3 2

Page 33: Deleuze, Gilles. Bergsonism


itself. Intui t ion is rather the movemen t by which we emerge •••

from our own duration, by which we make use of our own dura­

t ion to affirm and immedia te ly to recognize the ex i s t ence of

other durations, above or below us. "Only the method of which

we are speaking allows one to pass beyond idealism as well as

realism, to affirm the e x i s t e n c e of o b j e c t s both inferior and

superior to us, though nevertheless , in a cer tain sense, inte­

rior to us O n e perceives any number of durations, all very

dilfercnt from one another" (in fact the words inferior and superior

should not mislead us, they denote differences in k i n d ) . 5 3 With­

out intui t ion as me thod , duration would remain a s imple psy­

chologica l expe r i ence . Conversely, if it did not co inc ide with

duration, intui t ion would not be capable of carrying out the

program that corresponds to the preceding rules: the determi­

nation of true problems or of genuine differences in kind

Let us return, therefore, to the illusion of false p rob lems .

W h e r e does it c o m e from and in what sense is it inevitable?

Bergson calls into quest ion the order of needs, of ac t ion , and

of socie ty that predisposes us to retain only what interests us

in things; the order of in te l l igence , in its natural affinity with

space; and the order of general ideas that tends to obscure dif­

ferences in kind. Or rather there are very varied general ideas

that t h e m s e l v e s differ in k ind , s o m e referr ing to o b j e c t i v e

resemblances in living bodies, o thers to ob jec t ive ident i t ies

in inanimate bodies , and others again to subject ive demands

in manufactured o b j e c t s . But we are qu i ck to form a general

idea of all general ideas and to dissolve differences in kind in

this e lement of general i ty. 3 4 " W e make differences in kind melt

into the homogene i ty of the space which subtends t h e m . " 3 5

It is true that this c o l l e c t i o n of reasons is still psychological

and inseparable from our own condi t ion . We must take into


Page 34: Deleuze, Gilles. Bergsonism


considerat ion more profound reasons. For while the idea of a

homogeneous space impl ies a sort of art if ice or symbol sepa­

rating us from reality, it is nevertheless the case that mat te r

and extensi ty are realities, themselves prefiguring the order of

space. Although it is i l lusion, space is not merely grounded in

our nature, but in the nature of things. Mat te r is effectively

the " a s p e c t " by which things tend to present to each other ,

and to us, only differences in degree. Exper ience gives us com­

posites. Now the state of the c o m p o s i t e does not cons is t onlv

in unit ing e l e m e n t s that differ in kind, but in uni t ing them in

condi t ions such that these const i tuent differences in kind can­

not be grasped in it . In short , there is a point ol view, or rather

a state ol things, in which differences in kind can no longer

appear. T h e retrograde movement of the true is not merely an illu­

sion about the true, but belongs to the true itsell. Bergson adds

(dividing the compos i te "rel igion" into two directions — static

and dynamic r e l ig ion ) that in placing ourselves at a ce r ta in

standpoint "we should perceive a series of transit ions and, as

it were, differences of degree , whereas really there is a radical

difference in k i n d . " 1 6

The illusion, therefore, does not result onlv from our nature,

but from the world in which we live, from the side of being

that manifests i t se l f to us in the lirst p lace . Bergson evolved,

in a cer tain sense, from the beginning to the end ol his work.

T h e two major aspects ol his evolution are the following: Dura­

t ion seemed to him to be less and less reducib le to a psycho­

logical expe r i ence and b e c a m e instead the variable essence of

things, providing the t heme of a complex ontology. But, simul­

taneously, space seemed to him to be less and less reducible

to a fiction separating us from this psychological reality, rather,

it was itself grounded in being and expressed one of its two


Page 35: Deleuze, Gilles. Bergsonism


slopes, one of its two d i rec t ions . T h e absolute , said Bergson,

has two sides (aspects): spirit imbued with metaphysics and mat­

ter known by s c i e n c e . 5 7 But the point is that s c i ence is not a

relative knowledge, a symbol ic discipline that c o m m e n d s itsell

only by its successes or its e f fec t iveness ; s c i e n c e is part of

ontology, it is one ol ontology's two halves. T h e Absolute is

di lference, but difference has two facets, differences in degree

and differences in kind. It can, therefore , be seen that when

we grasp s imple differences in degree be tween things, when

sc i ence itself invites us to see the world in this way, we are

again in an absolute ( " W i t h modern physics more and more

clearly revealing to us differences in number behind our dis­

t inc t ions of quality " ) . 5 8 I t is, however, an il lusion. But i t

is only an illusion to the ex ten t that we project the real land­

scape of the first slope o n t o the other . If the illusion can be

repressed it is because of that o the r s lope , that ol durat ion,

which gives us differences in kind corresponding in the final instance

to differences of proportion as they appear in space, and already

in mat ter and ex tens ion .

Thus intuition does form a method with its three (or five) rules.

This is an essentially problemati/.ing me thod (a c r i t ique of false

p rob lems and the invent ion of genu ine o n e s ) , differentiating

(carvings out and intersections), temporali/ing (thinking in terms

of durat ion) . But how does intuition presuppose duration, and

how, on the o the r hand, does it give duration a new ex tens ion

Irom the point of view of being and knowledge? Th i s is what

remains to be de te rmined .


Page 36: Deleuze, Gilles. Bergsonism
Page 37: Deleuze, Gilles. Bergsonism


D u r a t i o n a s I m m e d i a t e D a t u m

Wo shall assume that the reader is familiar with the descrip­

t ion of duration as psychological expe r i ence as it appears in

lime and Iree Will and in the first pages of Creative Evolution: It

is a case of" a " t rans i t ion ," of a " c h a n g e , " a becoming, but it is a

becoming that endures, a change that is substance itself. T h e

reader will no t e that Bergson has no difficulty in reconc i l ing

the two fundamental characteristics of duration; continuity and

heterogenei ty . 1 However, defined in this way, duration is not

merely lived exper i ence ; it is also exper ience enlarged or even

gone beyond; it is already a condi t ion of exper ience . For expe­

r ience always gives us a c o m p o s i t e of space and duration. Pure

duration offers us a succession that is purely internal, wi thout

exter ior i ty; space, an exter ior i ty wi thout succession (in effect,

this is the memory of the past; the r eco l l ec t ion of what has

happened in space would already imply a mind that endures) .

I he two c o m b i n e , and into this combina t ion space introduces

the lorms ol its ext r ins ic dis t inct ions or ol its homogeneous

"iid d iscont inuous " s e c t i o n s , " while duration con t r ibu tes an

internal succession that is both heterogeneous and continuous.

We are thus able to "preserve" the instantaneous states of space


Page 38: Deleuze, Gilles. Bergsonism


and to juxtapose them in a sort of "auxil iary space" : But we

also i n t roduce ex t r i n s i c d i s t i nc t ions in to our dura t ion; we

decompose it into external parts and align it in a sort of homo­

geneous t i m e . A c o m p o s i t e of this kind (whe re homogeneous

t ime merges with auxiliary space) must be divided up. Even

before Bergson had b e c o m e consc ious of intuition as method ,

he had to face the task of dividing up the c o m p o s i t e . Should

it be divided along two pure d i rec t ions? So long as Bergson

does not expl ic i t ly pose the problem of an ontologica l origin

ol space, it is rather a case of dividing the c o m p o s i t e in t w o

d i rec t ions , only one of which (dura t ion) is pure, the o the r

( space ) is the impuri ty that denatures i t . 2 Durat ion will be

attained as " i m m e d i a t e da tum" because it is associated with

the right s ide, the good side of the c o m p o s i t e .

T h e important thing here is that the decompos i t ion of the

c o m p o s i t e reveals to us t w o types of mult ipl ici ty . O n e is rep­

resented by space (o r rather, if all the nuances are taken into

accoun t , by the impure combina t ion o f homogeneous t i m e ) :

I t i s a mul t ip l ic i ty of exter ior i ty , of simultaneity, of juxtapo­

si t ion, of order, of quant i ta t ive differentiation, oi difference in

degree; it is a numer ica l mul t ip l ic i ty , discontinuous and actual.

T h e o the r type of mul t ip l ic i ty appears in pure duration: I t is

an internal mul t ip l ic i ty of success ion, of fusion, of organiza­

t ion, of heterogenei ty , of quali tat ive d iscr iminat ion , or of dif­

ference in kind; it is a virtual and continuous m u l t i p l i c i t y that

cannot be reduced to n u m b e r s . 5

* * w

Too little importance has been attached to the use of this word

"multiplicity." It is not part of the traditional vocabulary at all -

this is particularly not the case when denoting a continuum. We

Page 39: Deleuze, Gilles. Bergsonism


shall see not only that it is fundamental in terms of the con­

struction ol the method , but also that, even at this early stage,

it tel ls us about the p rob lems that appear in Time and Tree Will.

(These will be developed later) . T h e word "mu l t i p l i c i t y " is

not there as a vague noun corresponding to the well-known

phi losophica l no t ion ol the Mul t ip l e in genera l . In fact for

Bergson it is not a question of opposing the Multiple to the One hut,

on the contrary, oj distinguishing two types of multiplicity. Now, this ^

problem goes back to a scholar of genius , G . B . R . R iemann , a

physicist and mathemat ic ian . Riemann defined as "mul t ip l i ­

c i t i e s " those things that could be de te rmined in terms ol their

d imensions or their independent variables. He dist inguished

discrete multiplicities and continuous multiplicities. T h e former con­

tain the pr inciple of the i r own met r ics ( the measure of one of

their parts being given by the number of e l emen t s they con­

tain) . T h e lat ter found a metr ica l pr inciple in someth ing e lse ,

even if only in phenomena unfolding in them or in the forces

acting in t h e m . 4 It is c lear that Bergson, as a philosopher, was

well aware of Riemann 's general p rob lems . Not only his inter­

est in ma themat i c s points toward this, but, more specifically,

Duration and Simultaneity is a book in which Bergson opposes

his own doc t r ine to the theory of Relativity, which is direct ly

dependent on Riemann. If our hypothesis is co r rec t , this book

loses its doubly strange character. In the lirst place, it does not

appear abruptly and without explanation. Rather, it brings into

the open a conf ron ta t ion that until then , had been impl i c i t

be tween Riemannian and Bergsonian interpretat ions of con­

tinuous mult ipl ici t ies . Second, Bergson's renunciation and con­

demnat ion of this book is perhaps due to the fact that he did

not feel able to pursue the mathematical implicat ions of a the­

ory of mul t ip l i c i t i es . He had, in fact, profoundly changed the

3 9

Page 40: Deleuze, Gilles. Bergsonism


direction ol the Riemannian distinction. Continuous multiplici­

ties seemed to him to be long essentially to the sphere of dura­

t ion . In this way, lor Bergson , durat ion was not s imply the

indivisible, nor was it the nonmeasurable . Rather , it was that

which divided only by changing in kind, that which was sus­

cep t ib le to measurement onlv by varying its metr ical princi­

ple at each stage ol the division. Bergson did not confine himsell

to opposing a philosophical vision of duration to a sc ient i f ic

c o n c e p t i o n ol space but t o o k the problem into the sphere of

the two kinds of multiplicity. I le thought that the mult ipl ic i ty

proper to duration had, lor its part, a "p rec i s ion" as great as

that of s c i ence ; moreover, that i t should react upon sc i ence

and open up a path lor it that was not necessarily the same as

that of R iemann and Einste in . Th i s is why we must at tach so

much importance to the way in which Bergson, borrowing the

notion of multiplici ty, gives it renewed range and distribution.

I low is the qualitative and cont inuous mult ipl ic i ty ol dura­

t ion del ined, in opposi t ion to quant i ta t ive or numerical mul­

tiplicity? A difficult passage from Time and Tree Will is particularly

significant in this respect as it foreshadows the deve lopments

in Matter and Memory. It distinguishes the subjec t ive and the

ob jec t ive : " W e apply the te rm subject ive to what seems to In­

comple te ly and adequately known; and the term objec t ive , to

what is known in such a way that a constant ly increasing num­

ber ol new impressions could be substituted lor the idea which

we actually have ol i t . " ' II we confine ourselves to these for­

mulations, we run the risk ol misunderstandings, which are for­

tunately dispelled by the con tex t . Bergson in fact specifies that

an object can be divided up in an infinity of ways. Now, even

before these divisions are made, they are grasped by thought

as possible, without anything changing in the total aspect ol the

4 0

Page 41: Deleuze, Gilles. Bergsonism


o b j e c t . I hey are t he re fo re already v i s ib l e in t h e image of

the ob jec t : I von when not realized (but simply possible) , they

are actually perceived, or at least perceptible in principle. "This

actual, not merely virtual, apperception of subdivisions in the

undivided is precisely what we call object ivi ty." Bergson means

that the ob jec t ive is that which has no virtualitv — whe the r real­

ized or not , whe the r possible or real, everything is actual in

the o b j e c t i v e . The first chapter ol Matter and Memory develops

this t heme more clearly: Mat te r has ne i ther virtualitv nor hid­

den power, and that is why we can assimilate it to " the image."

No doubt there can be more in mat te r than in the image we

have of it, but there cannot be anything else in it, of a differ­

ent k i n d . 6 And in a n o t h e r passage Bergson praises Berke ley

lor having assimilated body and idea, precisely because mat ter

"has no in ter ior , no u n d e r n e a t h , . . . h ides no th ing , con ta ins

no th ing . . . possesses nei ther power nor virtualitv ol any k i n d . . .

is spread out as mere surface a n d . . . i s no more than what it

presents to us at any given m o m e n t . " 7

In short , " o b j e c t " and " o b j e c t i v e " deno te not only what is

divided, but what, in dividing, does not change in kind. It is

thus what divides by differences in degree . 8 T h e ob jec t is char­

acterized by the perfect equivalence of the divided and the divi­

sions, of n u m b e r and uni t . In this sense, the o b j e c t wil l be

called a "numer ica l mul t ip l ic i ty ." f o r number, and primarily

(In- ari thmetical unit i tsell , is the model ol that which divides

without changing in kind. Th i s is the same as saying that num­

ber has only d i f fe rences in d e g r e e , or tha t its d i f f e rences ,

whe ther realized or not, are always actual in it. " T h e units by

means of which a r i t hme t i c forms numbers are provisional units

which can be subdivided without l imi t , a n d . . . e a c h ol them

is the sum ol fractional quant i t ies , as small and as numerous

1 1

Page 42: Deleuze, Gilles. Bergsonism


as we like to imagine While- all mul t ipl icat ion implies the

possibility ol treating any number whatever as a provisional unit

that can be added to i tsel l , conversely the units in thei r turn

are true numbers which are as big as we l ike, but are regarded

as provisionally indivisible lor the purpose of c o m p o u n d i n g

them with one another. Now, the very admission that it is pos­

sible to divide the unit into as many parts as we l ike , shows

that we regard it as ex tended. ' " '

On the o ther hand, what is a qualitative multiplici ty? Wha t

is the sub jec t or the subjec t ive? Bergson gives the following

example : "A c o m p l e x feeling will con ta in a fairly large num­

ber of s imple e l emen t s ; but as long as these e l emen t s do not

stand out with perfect clearness, we cannot say that they were

c o m p l e t e l y rea l ized , and as soon as consc iousnes s has a dis­

t inc t percept ion of them, the psychic state which results from

their synthesis will have changed lor this very r ea son . " 1 0 (For

example , a c o m p l e x of love and hatred is actualized in con­

sciousness, but hatred and love b e c o m e consc ious under such

condi t ions that they differ in kind from one another and also

differ in kind from the unconsc ious c o m p l e x ) . It would there­

fore be a serious mistake to think that duration was simply the

indivisible, although for conven ience , Bergson often expresses

himself in this way. In reality, duration divides up and does so

constantly: That is why it is a multiplicity. But it does not divide-

up without changing in kind, it changes in kind in the pro­

cess of d i v i d i n g up: Th i s is why it is a nonnumcr i ca l mult i ­

plicity, where we can speak ol " ind iv i s ib les" at each stage of

the division. T h e r e is other wi thout there being several; num­

ber exists only po ten t ia l ly . " In o ther words, the sub jec t ive ,

or duration, is the virtual. To be more precise , it is the virtual

insofar as it is actualized, in the course of being actual ized, it

4 2

Page 43: Deleuze, Gilles. Bergsonism


is inseparable from the movement of its actualization. For actu­

alization c o m e s about through differentiation, through diver­

gent l ines, and crea tes so many differences in kind by virtue

of its own movement . Everything is actual in a numerical mul­

t ip l ic i ty; everything is not " rea l ized ," but everything there is

actual . T h e r e are no relat ionships o the r than those be tween

actuals, and no differences o the r than those in degree. On the

o ther hand, a nonnumerical mult ipl ic i ty by which duration or

subjectivity is defined, plunges into another dimension, which

is no longer spatial and is purely temporal : It moves from the

virtual to its actual izat ion, it actualizes itself by creat ing lines

of differentiation that correspond to its differences in kind. A

mul t ip l ic i ty of this kind has, essentially, the three propert ies

ol cont inu i ty , he terogenei ty , and s impl ic i ty . In this ins tance

Bergson does not have any real difficulty in r econc i l ing het­

erogenei ty and cont inui ty .

The aforementioned passage from Time and Free Will, wherein

Bergson distinguishes the subjective and the objec t ive , appears

to be all the more important insofar as it is the first to intro­

duce indirect ly the no t ion o f the virtual. Th i s not ion o f the

virtual wil l c o m e to play an increasingly impor tan t ro le in

Bergsonian phi losophy. 1 2 For, as we shall see, the same author

who re jec ts the c o n c e p t oi possibility — reserving a use for it

only in relation to mat te r and to c losed systems, but always

seeing it as the source of all kinds of false problems — is also

he who develops the no t ion of the virtual to its highest degree

and bases a whole philosophy of memory and life on it .

A very important aspect of the not ion of mult ipl ic i ty is the

way in which it is dist inguished from a theory of the O n e and

the Mul t ip le . T h e not ion of mul t ip l ic i ty saves us from think­

ing in terms of " O n e and Mul t ip l e . " T h e r e are many theor ies


Page 44: Deleuze, Gilles. Bergsonism


in philosophy that c o m b i n e the one and the mul t ip le . They

share the characteristic of claiming to reconstruct the real with

general ideas. We are told that the S e l f is one ( thes i s ) and it is

mul t ip le (ant i thes is ) , then i t is the unity of the mul t ip le (syn­

thesis) . Or e lse we are to ld that the O n e is already mul t ip le ,

that Being passes into nonbeing and produces becoming . T h e

passages where Bergson condemns this movemen t ol abstract

thought are among the finest in his oeuvre. To Bergson, it seems

that in this type of dialectical method , one begins with concepts

that, like baggy c lo thes , are much too b i g . " T h e O n e in gen­

eral, the mul t ip le in general , nonbeing in general In such

cases the real is r e c o m p o s e d with abstracts ; but ol what use

is a d ia lec t ic that believes i t se l f to be reunited with the real

when it compensa tes lor the inadequacy of a c o n c e p t that is

t o o broad or t o o general by invoking the oppos i te c o n c e p t ,

which is no less broad and general? T h e c o n c r e t e will never

be attained by combin ing the inadequacy of one c o n c e p t with

the inadequacy o l its o p p o s i t e . T h e s ingular wil l never be

attained by co r rec t ing a general i ty with another generality. In

all this, Bergson clearly has in mind l l ame l in whose Essai sur

les elements principaux de la representation dates from 1 9 0 7 . Berg-

sonism's incompat ib i l i ty with I legelianism, indeed with any

dialect ical m e t h o d , is also evident in these passages. Bergson

criticizes the dialectic lor being a false movement, that is, a move­

ment ol the abstract c o n c e p t , which goes from one opposi te

to the o ther only by means of i m p r e c i s i o n . 1 4

O n c e again there is a P la ton ic tone in Bergson. P la to was

the lirst to deride those who said " the O n e is mul t ip le and the

mul t ip le one — Being is nonbe ing , " e t c . In each case he asked

how, how main, when and where. " W h a t " unity of the mul t ip le

and "wha t " mul t ip le o l the o n e ? " T h e combina t ion o l oppo-


Page 45: Deleuze, Gilles. Bergsonism


si lcs tells us nothing; it forms a net so slack that everything

slips through. Those me taphor s ol P la to about carving and

the good c o o k (which Bergson likes so m u c h ) correspond to

Bergson's invocation ol the good tailor and the well-fitted out-

lit. This is what the precise concept must be like. " W h a t really

matters to philosophy is to know what unity, what mult iplici ty,

what reality superior to the abstract one and the abstract mul­

tiple is the mul t ip le unity ol the person C o n c e p t s . . . o r d i ­

narily go by pairs and represent the t w o opposi tes . T h e r e is

scarcely any conc re t e reality upon which one cannot take two

opposing views at the same t i m e and that is consequen t ly not

subsumed under the two antagonistic concep t s . I l ence a thesis

and an ant i thesis which it would be vain for us to try logically

to r econc i l e , lor the s imple reason that never, with c o n c e p t s

or points ol view, will you make a thing If I try to analyze

duration, that is, to resolve it in to ready-made c o n c e p t s , I am

obliged by the very nature of the c o n c e p t and the analysis to

take two opposing views of Juration in general, with which I shall

then claim to recompose it. This combinat ion can present nei­

ther a diversity ol degrees nor a variety of forms: It is, or it is

no t . I shall say, for e x a m p l e , that the re is, on the one hand,

a multiplicity of successive states ol consciousness and, on the

o ther hand, a unity which binds t hem together . Duration will

be the synthesis of this unity and multiplicity, but how this mys­

terious operat ion can admit of shades or degrees, 1 repeat , is

not qu i t e c l e a r . " 1 6

What Bergson calls lor — against the dialect ic , against a gen­

eral c o n c e p t i o n ol opposi tes ( the O n e and the Mul t ip l e ) — is

an acute perception of the "what" and the "how many," ol what

he cal ls the " n u a n c e " or the potent ia l number . Dura t ion is

opposed to b e c o m i n g precisely because it is a mult ipl ic i ty , a

4 5

Page 46: Deleuze, Gilles. Bergsonism


type ol mul t ip l i c i ty that is not r educ ib le to an overly broad

combina t ion in which the opposi tes , the O n e and the Mult i ­

ple in general, only co inc ide on condi t ion that they are grasped

a t the e x t r e m e point o l t h e i r gene ra l i za t i on , e m p t y ol all

"measu re" and of all real substance. Th i s mul t ip l ic i ty that is

duration is not at all the same thing as the mul t ip le , any more

than its s impl ic i ty is the same as the O n e .

Two forms of the negative are often distinguished: The nega­

tive of s imple l imitat ion and the negative of oppos i t ion . We

are assured that the substitution of the second lorm tor the first

by Kant and the post-Kantians was a revolut ion in philosophy.

It is all the more remarkable that Bergson, in his c r i t ique of

t h e nega t ive , c o n d e m n s b o t h forms. Bo th s e e m to h im to

involve and to demonstrate the same inadequacy. For if we con­

sider negative not ions l ike disorder or nonbeing, their very con­

cept ion (from the starting-point of being and order as the limit

of a "deterioration" in whose interval all things are [analytically]

included) amounts to the same thing as our conceiving of them

in opposit ion to being and order, as forces that exercise power

and c o m b i n e with thei r opposi tes to produce ( synthe t ica l ly)

all things. Bergson's c r i t ique is thus a doub le one insofar as it

c o n d e m n s , in bo th forms of the negative, the same ignorance

of differences in kind, which are some t imes treated as "deter ior­

a t i o n s , " s o m e t i m e s as o p p o s i t i o n s . T h e heart o f Bergson 's

p r o j e c t is to th ink d i f ferences in kind independen t ly of all

lorms ol negation: The re are differences in being and vet noth­

ing negative. Negation always involves abstract c o n c e p t s that

are much t o o general . W h a t is, in fact, the c o m m o n root of

all negat ion? We have already seen it. Instead of starting out

from a difference in kind be tween two orders , from a differ­

e n c e in kind be tween two beings, a general idea of order or

4 6

Page 47: Deleuze, Gilles. Bergsonism


being is c rea ted , which can no longer be thought e x c e p t in

opposi t ion to a nonbeing in general , a disorder in general , or

else which can only be posited as the starting point ol a deter i­

oration that leads us to disorder in general or to nonbeing in

general. In any case, the question of difference in kind - "what"

order? "wha t " being? — has been neg lec ted , l i k e w i s e the dif­

ference in kind between the two types ol multiplicity has been

n e g l e c t e d : T h u s a general idea ol the O n e is c rea ted and is

combined with its opposi te , the Mul t ip le in general, to recon­

struct all things from the standpoint ol the lorce opposed to

the m u l t i p l e or to the de te r io ra t ion of the O n e . In fact, i t

i s the ca tegory of mul t ip l ic i ty , wi th the d i f ference in kind

be tween t w o types that i t involves, which enables us to con­

demn the myst if icat ion of a thought that operates in terms of

the O n e and the Mul t ip le . We see, therefore, how all the cri t i­

cal aspects of Bergsonian philosophy are part of a single theme:

a cri t ique of the negative of l imitation, of the negative of oppo­

sit ion, of general ideas.

" I f we analyze in the same way the c o n c e p t o f m o t i o n " l 7

In fact, movement as physical exper i ence is itself a compos i t e :

on the one hand, the space traversed by the moving o b j e c t ,

which forms an indefinitely divisible numerical mult ipl ic i ty ,

all of whose parts - real or possible - are actual and differ only

in degree; on the other hand, pure movement , which is alter­

ation, a virtual quali tat ive mult ipl ici ty, like the run of Achil­

les that is divisible into steps, but which changes qualitatively

each t ime that it d iv ides . 1 8 Bergson discovers that beneath the

local transfer there is always a conveyance of another nature.

And what seemed from outside to be a numerical part, a com-

4 7

Page 48: Deleuze, Gilles. Bergsonism


portent of the run, turns out to he, expe r i enced from inside,

an obs tac le avoided.

But in doubling the psychological expe r i ence of duration

with the physical expe r i ence of movemen t , one problem be­

c o m e s pressing. The ques t ion " D o externa l things endure?"

remained indeterminate from the standpoint of psychological

expe r i ence . Moreover , in Time and Free Will, Bergson invoked

on two occas ions an " inexpres s ib l e , " an " i n c o m p r e h e n s i b l e "

reason — " W h a t duration is there exist ing outside us? The pre­

sent only, or, if we prefer the expression, simultaneity. No doubt

external things change, but the i r m o m e n t s do not succeed (in

the ordinary sense of the word) one another, e x c e p t for a con­

sciousness that keeps t hem in mind H e n c e we must not

say that external things endure, but rather that there is s o m e

inexpressible reason in t hem which accoun ts for our inability

to examine them at successive m o m e n t s of our own duration

without observing that they have changed." - "Although things

do not endure as we do ourselves, nevertheless , there must be

s o m e incomprehens ib l e reason why phenomena are seen to

succeed one another instead of being set out all at o n c e . " ' 1 '

However, Time and Free Will already had an analysis of move­

m e n t . But movemen t had been primarily posited as a "fact of

consciousness" implying a conscious and enduring subject con­

fused with duration as psychological e x p e r i e n c e . It is only to

the ex tent that movement is grasped as belonging to things as

much as to consciousness that i t ceases to be confused with

psychological duration, whose point of application it will dis­

place , thereby necessi ta t ing that things part icipate direct ly in

duration itself. II quali t ies exis t in things no less than they do

in consc iousness , if there is a movemen t ol qualit ies outside

myself, things must, of necessity, endure in their own way. I'sy-

4 8

Page 49: Deleuze, Gilles. Bergsonism


etiological duration should be only a clearly de te rmined case,

an opening on to an on to log ica l duration. On to logy should, of

necessity, be possible, f o r duration was delined from the start

as a mult ipl ici ty . Wi l l this mul t ip l ic i ty not — thanks to move­

ment — b e c o m e contused with be ing i tsell? And, s ince it is

endowed with very special propert ies , in what sense can it be

said that there are several durations; in what sense can there be

said to be a single one; in what sense can one get beyond the

ontological alternative of one several? A related problem now

b e c o m e s more urgent. If things endure, or i f there is duration

in things, the ques t ion ol space will need to be reassessed on

new loundations. For space will no longer simply be a form

ol exteriority, a sort of screen that denatures duration, an impu­

rity that c o m e s to disturb the pure, a relative that is opposed

to the absolu te : Space itsell will need to be based in things,

in relat ions be tween things and be tween durations, to be long

i t se l f to the absolute , to have its own "purity." This was to be

the double progression of the Bergsonian philosophy.

4 9

Page 50: Deleuze, Gilles. Bergsonism
Page 51: Deleuze, Gilles. Bergsonism

C H A P T E R 111

M e m o r y a s V i r t u a l C o e x i s t e n c e

Duration is essentially memory , consciousness and freedom.

It is consciousness and freedom because it is primarily memory.

Now, Bergson always presents this iden t i ty ol m e m o r y and

duration in t w o ways: " t h e conservat ion and preservation of

the past in the present." Or else "whether the present dist inctly

con ta ins t h e ever-growing image of t h e past, or w h e t h e r by-

its cont inual changing ol quali ty attests rather to the increas­

ingly heavy burden dragged along behind one the o lder one

grows." Or again: " m e m o r y in these two forms, covering as it

does wi th a c loak of r e c o l l e c t i o n s a c o r e of immed ia t e per­

cept ion, and also contract ing a number ol external moments . " 1

In fact we should express in two ways the manner in which

duration is distinguished Irom a discontinuous series of instants

repeated identically: On the one hand, " the following m o m e n t

always conta ins , over and above the preceding one , the mem­

ory the latter has left i t" ; 2 on the other hand, the two moments

c o n t r a c t o r c o n d e n s e in to each o t h e r s ince one has no t yet

disappeared when another appears. T h e r e are, therefore , t w o

memor ies — or two indissolubly linked aspects ol m e m o r y -

s i

Page 52: Deleuze, Gilles. Bergsonism


r e c o l l e c t i o n - m e - m o r v and c o n t r a c t i o n - m e m o r y (II we ask

what, in the linal analysis, is the basis ol this duality in dura­

t ion , doubt less we find ourselves in a movement — which we

shall e x a m i n e la ter — by wh ich the " p r e s e n t " that endures

divides at each " ins tant" into two direct ions , one or iented and

d i la ted toward the past, the o t h e r c o n t r a c t e d , c o n t r a c t i n g

toward the future) .

But pure duration is i tself the result ol a division that is only

opera t ive " in p r i n c i p l e " (en droit). It is c l e a r that m e m o r y

is identical to duration, that it is coex tens ive with duration,

but this pro|X)sition is valid in principle more than in lact. I he

special problem ol memory is: I low, by what mechanism, does

duration b e c o m e memory in lact? I low does that which exists

in principle actual ize itsell? In the same way, Bergson shows

that consciousness is, in pr inciple , coex tens ive with lile; but

how, and unde r what c o n d i t i o n s , does l ife in fact b e c o m e

se l l -consc iousness? 3

* * *

Le t us resume the analysis of the lirst chap te r of Matter and

Memory. We are led to distinguish live senses or aspects ol sub-

j ec t iv i ty : ( I ) need-subjectivity, the moment ol negation (need

makes a hole in the cont inu i ty of things and holds back every­

thing that interests i t about the o b j e c t , let t ing the rest go by);

( 2 ) brain-subjectivity, the m o m e n t ol interval or ol indetermi-

nat ion ( the brain gives us the means ol " c h o o s i n g " that which

corresponds to our needs in the o b j e c t ; int roducing an inter­

val be tween received and execu ted movement , it is i tse l f the

cho ice between two ways because, in itsell, by virtue of its net­

work of nerves , i t divides up e x c i t a t i o n infini tely and also

because , in relat ion to the mo to r ce l l s ol the core it leaves us

Page 53: Deleuze, Gilles. Bergsonism


to c h o o s e be tween several possible r eac t ions ) ; (I) affection-

subjectivity, the m o m e n t ol pain (because affection is the price

paid by the brain or by conscious percept ion; perception does

not reflect possible ac t ion , nor does the brain bring about the

interval wi thout the assurance that ce r ta in organic parts are

c o m m i t t e d to the immobi l i ty of a purely recept ive role that

surrenders t h e m to pa in) ; ( 4 ) recollection-subjectivity, the pri­

mary aspect ol memory ( reco l l ec t ion being what c o m e s to till

the interval , be ing e m b o d i e d or ac tual ized in the properly

cerebral interval [intcrvallc]); ( 5 ) contraction-subjectivity, the sec­

ond aspect of memory ( the body being no more a punct i lbrm

instant in t i m e than a mathemat ica l point in space, and bring­

ing about a con t rac t ion ol the exper ienced exc i ta t ions from

which quality is bo rn ) .

Now, these live aspects are not mere ly organized in order

ol increasing depth, but are distributed on two very different lines

of facts. The lirst chapter ol Matter and Memory sets out to decom­

pose a compos i te (Representa t ion) in two divergent direct ions:

matter and memory, perception and recol lect ion, object ive and

subject ive ( c f . the t w o mul t ip l i c i t i e s of Time and Tree Will). Of

the live aspects ol subject ivi ty, the lirst t w o obviously be long

to the ob jec t ive l ine, s ince the lirst conf ines itsell to abstract­

ing Irom the o b j e c t , and the second conf ines i t se l f to estab­

lishing a zone of indeterminat ion . T h e case of affect ion, the

third sense, is more c o m p l e x ; it undoubtedly depends on the

in tersect ion of the two l ines. But the positivity of affection,

in its turn, is not yet the presence of a pure subject iv i ty that

would be opposed to pure objectivity, it is rather the "impurity"

that disturbs the la t te r . 4 T h e province of the pure line of sub­

jec t iv i ty is thus the fourth, and then the fifth sense. Only the

two aspects of memory str ict ly signify subjectivi ty, the o ther

5 i

Page 54: Deleuze, Gilles. Bergsonism

8 E R G S O M S M

meanings conf ine themselves to making way for or bringing

about the insertion ol one line into the other , the intersect ion

ol one line with the other .

T h e quest ion " W h e r e are r eco l l ec t ions preserved?" involves

a false p rob lem, that is to sav, a badly analyzed c o m p o s i t e . It

is as though r eco l l ec t ions had to be preserved somewhere , as

though, lor example, the brain were capable of preserving them.

But the brain is wholly on the line of ob j ec t i v i t y : T h e r e can­

not be any difference in kind be tween the o the r states of mat­

te r and the brain. For in the lat ter everything is movemen t , as

in the pure percept ion that it de te rmines . (And yet the te rm

movement obviously must not be understood in the sense of

enduring movemen t , but on the contrary as an " instantaneous

s e c t i o n . " ) 5 R e c o l l e c t i o n , on the contrary, is part of the line of

subject ivi ty. It is absurd to mix the two lines by conce iv ing

ol the brain as the reservoir or the substratum of reco l lec t ions .

Moreover , an examinat ion of the second l ine would be suffi­

c i en t to show that r eco l l ec t ions do not have to be preserved

anywhere o the r than " i n " durat ion. Recollection therefore is pre­

served in itself. Only then "did I b e c o m e aware of the fact that

inward expe r i ence in the pure state, in giving us a ' subs tance '

whose very essence is to endure and consequent ly to prolong

continually into the present an indestructible past, would have

relieved me from seeking and would even have forbidden me

to seek, where recollection is preserved. It preserves itself " < >

Moreover, we have no interest in presupposing a preservation

ol the past e lsewhere than in itself, for e x a m p l e , in the brain.

T h e brain, in its turn, would need to have the power to pre­

serve itself; we would need to i onfer this power ol preserva-

Page 55: Deleuze, Gilles. Bergsonism


tion that we have denied to duration on a state of matter, or

even on the whole of ma t t e r . 7

We are touching on one of the most profound, hut perhaps

also one of the least unders tood, aspects of Bergsonism: the

theory of memory. The re must be a difference in kind between

mat ter and memory, be tween pure percept ion and pure rec­

ollect ion, between the present and the past, as there is between

the two lines previously distinguished. We have great difficulty

in unders tanding a survival of the past in i tself because we

bel ieve that the past is no longer, that it has ceased to be . We

have thus confused Being with being-present. Nevertheless, the

present Is not; rather, it is pure becoming, always outside itself.

It is no t , but it ac ts . Its proper e l e m e n t is not being but the

.11 five or the useful. The past, on the o ther hand, has ceased

to act or to be useful. But i t has not ceased to be . Useless and

inact ive, impassive, it IS, in the full sense of the word: It is

ident ica l wi th be ing in i tself . It should not be said that it

"was , " s ince it is the in-itself of being, and the form under

which being is preserved in itself ( in opposi t ion to the pre­

sent, the form under which being is consummated and places

itsell O u t s i d e of i tself) . At the l imi t , the ordinary determina­

tions are reversed: of the present , we must say at every instant

that it "was , " and of the past, that it " i s , " that it is eternally,

lor all t ime . Th i s is the difference in kind between the past

and the present ." But this first aspect of the Bergsonian the­

ory woidd lose all sense if its extra-psychological range were

not emphasized. W h a t Bergson calls "pure r e c o l l e c t i o n " has

no psycho log ica l e x i s t e n c e . T h i s is why it is ca l l ed virtual,

inactive, and unconsc ious . All these words are dangerous, in

particular, the word " u n c o n s c i o u s " which , s ince Freud, has

b e c o m e inseparable from an especially effective and active psy-

Page 56: Deleuze, Gilles. Bergsonism


chologic.i l ex i s t ence . We will have occas ion to compare the

Freudian unconscious with the Bergson ian, s ince Bergson him-

sell made the compar i son . 4 We must nevertheless be c lear at

this point that Bergson does not use the word " u n c o n s c i o u s "

to deno t e a psychological reality outside consc iousness , but

to deno te a nonpsychological reality — being as it is in itself.

S t r ic t ly speaking, the psychological is the present. Only the

present is "psychological" ; but the past is pure ontology; pure

r eco l l ec t ion has only on to log ica l s i g n i f i c a n c e . 1 0

Let us now quote the admirable passage where Bergson sum­

marizes the whole of his theory. W h e n we look for a r eco l l ec ­

t ion that escapes us, " W e b e c o m e conscious of an act sui aeneris

by wh ich we detach ourselves from the present in order to

replace ourselves, first in the past in general , then in a cer tain

reg ion of the past — a work of a d j u s t m e n t , s o m e t h i n g l ike

the focusing of a camera. But our reco l lec t ion still remains vir­

tual; we simply prepare ourselves to receive it by adopting the

appropriate a t t i tude. Li t t le by l i t t le it c o m e s into view like a

condens ing c l o u d ; from the virtual s ta te i t passes in to the

actual "" Here again, one must avoid an overly psychologi­

cal interpretat ion of the t e x t . Bergson does indeed speak of a

psychological ac t ; but if this ac t is "sui generis," this is because

it has made a genuine leap. We place ourselves at once in the

past; we leap into the past as into a proper e l e m e n t . 1 2 In the

same way that we do not perceive things in ourselves, but at

the place where they are, we only grasp the past at the place

where it is in itself, and not in ourselves, in our present. I here

is therefore a "past in genera l" that is not the particular past

of a particular present but that is like an onto logica l e l e m e n t ,

a past that is e ternal and for all t i m e , the c o n d i t i o n ol the

"passage" ol every particular present. It is the past in general


Page 57: Deleuze, Gilles. Bergsonism


that makes possible all pasts. According to Bergson, we first

put ourselves back into the past in general: He describes in this

way the leap into ontology. We really leap into being, into being-

in-itself, into the being in itsell ol the past. It is a case ol leaving

psychology al together . It is a case ol an immemor ia l or on to l ­

ogical Memory. It is only then , o n c e the leap has been made,

that r e c o l l e c t i o n will gradually take on a psychological ex i s ­

t ence : "from the virtual it passes into the actual state " We

have had to search at the place where it is, in impassive Being,

and gradually we give it an embodiment , a "psychologi/at ion."

T h e parallels be tween this text and s o m e others must be

emphasized. For Bergson analyzes language in the same way as

memory. T h e way in which we understand what is said to us

is identical to the way in which we find a recol lect ion. Far from

recompos ing sense on the basis ol sounds that are heard and

associated images, we place ourselves at once in the e l e m e n t of

sense, then in a region ol this e l ement . A true leap into Being.

It is only then that sense is actualized in the psychologically

perceived sounds, and in the images that are psychologically

associated wi th the sounds. Here there is a kind ol transcen-

dance of sense and an ontological foundation ol language that,

as we shall see , are particularly important in the work of an

author whose c r i t ique ol language is considered to have been

overly has ty . ' 3

We must place ourselves at o n c e in the past — in a leap, in

a jump. I lere again, this a lmost Kirkegaardian idea ol a " l e ap"

is strange in the work ol a phi losopher who is considered to

be so at tached to continuity. What does it mean? Bergson con­

stantly says: You will never r ecompose the past with presents,

no matter what they may be : " T h e image pure and s imple will

not take me back to the past unless, indeed, it was in the past

Page 58: Deleuze, Gilles. Bergsonism


that I sought i t . " 1 4 T h e past, i t is t rue, s e e m s to be caught

b e t w e e n t w o presents : t h e old present that i t o n c e was and

the actual present in relation to which it is now past. Two lalse

beliefs are derived from this: On the one hand, we believe that

the past as such is only cons t i tu ted after having been present;

on the o ther hand, that it is in s o m e way recons t i tu ted In the

new present whose past it now is. Th i s double illusion is at

the heart ol all physiological and psychologica l t heo r i e s of

m e m o r y . W h e n o n e i s i n f luenced by such an i l lus ion , o n e

assumes that there is only a difference in degree be tween rec­

o l lec t ion and perception. We are thus entangled in a badly ana­

lyzed compos i t e . Th i s c o m p o s i t e is the image as psychological

reality. T h e image in effect retains someth ing of the regions

where we have had to look for the r eco l l ec t i on that it actual­

izes or e m b o d i e s . But i t does not actualize this r eco l l ec t i on

without adapting it to the requirements of the present; it makes

i t in to someth ing of the present. Thus , we subst i tute the sim­

ple d i f ferences in degree b e t w e e n r e c o l l e c t i o n - i m a g e s and

perception-images for the difference in kind between the pres­

ent and the past, be tween pure percep t ion and pure memory.

We are t o o accus tomed to th inking in terms of the "pres­

ent ." We bel ieve that a present is only past when it is replaced

by another present . Nevertheless , let us s top and reflect for a

momen t : I low would a new present c o m e about if the old pres­

ent did not pass at the same t ime that it is present? How would

any present whatsoever pass, if it were not past at the same time

as present? T h e past would never be cons t i tu ted if it had not

been cons t i tu ted first of all, at the same t ime that it was pres­

en t . T h e r e is he re , as it were , a fundamental posit ion ol t i m e

and also the most profound paradox ol m e m o r y : T h e past is

"con temporaneous" with the present that it has been. I I the past

5 8

Page 59: Deleuze, Gilles. Bergsonism


had to wait in order to be no longer, if it was not immediately

and now that it had passed, "past in genera l , " it could never

b e c o m e what it is, it would never be that past. II it were not

cons t i tu ted immediately, nei ther could it be reconst i tuted on

the basis of an ul ter ior present . T h e past would never be con­

s t i tu ted if it did not c o e x i s t with the present whose past it

i s . 1 5 T h e past and the present do not deno t e t w o successive

moments , but two e lements which coexis t : One is the present,

which does not cease to pass, and the o the r is the past, which

does not cease to be but through which all presents pass. It is

in this sense that there is a pure past, a kind of "past in gen­

eral": T h e past does not follow the present , but on the con­

trary, is presupposed by it as the pure condi t ion without which

it would not pass. In o the r words, each present goes back to

itsell as past. T h e only equiva lent thesis is Plato 's no t ion of

R e m i n i s c e n c e . T h e r emin i s cence also aflirms a pure being of

the past, a being in itsell ol the past, an on to log ica l Memory

that is capable of serving as the foundation for the unfolding

of t ime . Yet again, a Platonic inspiration makes itsell profoundly

felt in B e r g s o n . 1 6

I he idea of a con temporane i ty of the present and the past

has one final consequence : Not only does the past coexis t with

the present that has been , but , as it preserves i t s e l f in i tself

(whi le the present passes), it is the who le , integral past; it is

all ou r past, which c o e x i s t s wi th each present . T h e famous

metaphor of the c o n e represents this c o m p l e t e state of c o e x ­

i s tence . But such a state implies , finally, that in the past itsell

there appear all kinds ol levels ol profundity, marking all the

possible intervals in this c o e x i s t e n c e . 1 7 T h e past AB coex i s t s

with the present S, but by including in itself all the sec t ions

A ' B ' , A " B " , e t c . , that measure the degrees o l a purely ideal

J 9

Page 60: Deleuze, Gilles. Bergsonism


A '


proximity or d is tance in relat ion to S. Each ol these sec t ions

is itself virtual, belonging to the being in itsell ol the pas t . 1 8

Each of these sect ions or each ol these levels includes not par­

t icular e lements of the past, but always the total i ty of the past.

It includes this totality at a more or less expanded or contracted

level. Th i s is the precise point at which con t r ac t i on -Memory

tits in with r e c o l l e c t i o n - M e m o r y and, in a way, takes over from

it. I lence this consequence : Bergsonian duration is, in the final

analysis, defined less by succession than by c o e x i s t e n c e .

In Time and tree Will duration is really defined by succession,

coex i s t ences referring back to space, and by the power of nov­

elty, repetition referring back to Matter. But, more profoundly,

duration is only success ion relatively speaking (we have seen

in the same way that it is only indivisible relatively). Duration

is indeed real success ion, but it is so only because , more pro­

foundly, it is virtual coexistence: the c o e x i s t e n c e with itsell ol

all the levels, all the tensions, all the degrees of con t rac t ion

and relaxation (detente). Thus, with coexis tence , repetition must

be reintroduced into duration — a "psychic" repetition of a com­

pletely dilferent type than the "physical" repet i t ion of mat-


Page 61: Deleuze, Gilles. Bergsonism


tor; a repet i t ion ol "planes" rather than ol e lements on a single

plane; virtual instead ol actual repe t i t ion . T h e whole of our

past is played, restarts, repeats itself, at the same time, on all the

levels that i t ske tches o u t . 1 9 Let us return to the " l eap" that

we make when, looking for a r e co l l e c t i on , we place ourselves

at o n c e in the past. Bergson gives the following clar i f icat ion:

We place ourselves "firstly into the past in general , then into

a certain region of the past." It is not a case ol one region con­

taining particular e lements ol the past, particular recol lec t ions ,

in opposi t ion to another region which conta ins o the r reco l ­

l ec t ions . It is a case ol there being dis t inct levels, each one of

which conta ins the whole of our past, but in a more or less

con t rac ted s tate . It is in this sense that one can speak of the

regions of Being itself, the on to log ica l regions of the past "in

genera l , " all coexis t ing , all " repea t ing" one another.

Later we shall see how this doc t r ine revives all the prob­

lems of Bergsonism. I fowever, at this point it is enough to sum­

marize the four main propositions that form as many paradoxes:

( I ) we place ourselves at o n c e , in a leap, in the ontological ele­

ment of the past (paradox of the leap); ( 2 ) there is a difference

in kind be tween the present and the past (paradox of B e i n g ) ;

( 3 ) the past does not follow the present that it has been , but

coexists with it (paradox of coexis tence) ; ( 4 ) what coexists with

each present is the whole of the past, integrally, on various lev­

els of con t rac t ion and relaxation (detente) (paradox of psychic

repe t i t ion) . T h e s e paradoxes are i n t e r connec t ed ; each one is

dependent on the others. Conversely, the propositions that they

attack also form a group, insolar as these proposit ions are char­

acter ized by thei r being ordinary theories ol memory, l o r it is

a single illusion about the essence of T i m e , a single badly ana­

lyzed c o m p o s i t e that makes us bel ieve that: ( I ) we can recon-

Page 62: Deleuze, Gilles. Bergsonism


st i tute the past with the present; ( 2 ) we pass gradually from

o n e to the o ther ; ( 3 ) that they are dist inguished by a before

and an alter; and ( 4 ) that the work ol the mind is carried out

by the addition ol e l emen t s (rather than by changes of level,

genuine jumps , the reworking o l s y s t e m s ) . 2 0

* • *

O u r problem is: I low can pure r eco l l ec t i on take on a psycho­

logical existence? I low will this pure virtual be actualized? Thus

the present makes an appeal, accord ing to the requi rements

or needs ol the present situation. We make the "leap": We place

ourselves not simply in the e l e m e n t of the past in general , but

in a particular region, that is, on a particular level which , in a

kind ol R e m i n i s c e n c e , we assume corresponds to our actual

needs. Each level in effect contains the totality of our past, but

in a more or less con t rac ted s tate . And Bergson adds: T h e r e

are also dominant r eco l l ec t i ons , like remarkable points, which

vary from one level to the o t h e r . 2 1 A foreign word is spoken

in my presence : Given the situation this is not the same thing

as wondering what the language in general, of which this word

is a part, could be or what person o n c e said this word, or a

similar one , to me . Depending on the case , I do not leap into

the same region of the past; I do not place myself on the same

level; 1 do not appeal to the same essential character is t ics . Per­

haps I fail: Look ing for a r e co l l e c t i on , I may place myself on a

level that is too contracted, t o o narrow, or on the contrary, t o o

broad and expanded for it . I would then have to start from the

beginning again in order to find the c o r r e c t leap. We must

emphasize that this analysis, which seems to have so much psy­

chologica l finesse, really has a qu i te di l lerent meaning. It is

related to our affinity with being, our relationship with Being,

6 2

Page 63: Deleuze, Gilles. Bergsonism


. I I K I to the variety ol this relationship. Psychological conscious­

ness has no t yet been born . I t wil l be born , but p rec i se ly

because it has found its proper on to log ica l condi t ions here.

Faced with these ex t r eme ly difficult t ex t s , the task of the

c o m m e n t a t o r is to mult iply the d is t inc t ions , even and above

all when these tex ts confine themselves to suggesting the dis­

t inc t ions , rather than to establ ishing them strictly. First , we

must not confuse the appeal to r eco l l ec t ion and the "recal l of

the image" (or its evocat ion) . T h e appeal to recol lect ion is this

j ump by which I place myself in the virtual, in the past, in a

particular region of the past, at a particular level of con t rac­

t ion. It appears that this appeal expresses the properly on to ­

logica l d i m e n s i o n of man or, rather , o f m e m o r y : " B u t our

recol lec t ion still remains vi r tual ." 2 2 W h e n , on the o ther hand,

we speak of evoca t ion , or o l this recall of the image, some­

thing c o m p l e t e l y different is involved: O n c e we have put our­

selves on a particular level where r eco l l e c t i ons l ie, then , and

only then, do they tend to be actual ized. T h e appeal of the

present is such that they no longer have the ineffectiveness,

the impassivity that character ized them as pure r eco l l ec t ions ;

they b e c o m e recol lect ion- images , capable of being " reca l led ."

They are actualized or embodied . This actualization has all kinds

of d is t inc t aspects , stages, and d e g r e e s . 2 3 But through these

stages and these degrees it is the actual izat ion (and it a l o n e )

that cons t i tu tes psychological consc iousness . In any case , the

Bergsonian revolution is c lear : We do not move from the pre­

sent to the past, from percept ion to reco l l ec t ion , but from the

past to the present , from reco l l ec t ion to percept ion .

"Memory, laden with the whole ol the past, responds to the

appeal ol the present state by two s imul taneous movements ,

one ol translation, by which it moves in its entirety to meet expe-

Page 64: Deleuze, Gilles. Bergsonism


r i cnce , thus contracting more or less, though without dividing,

with a view to action; the other ol rotation upon itself, by which

it turns toward the situation of the moment , presenting to it that

side ol itsell which may prove to be the most usefu l . " 2 ' Thus

we already have two aspects of actualization here: t ranslat ion-

contract ion and rotat ion-orientat ion. Our question is: Can this

t ransla t ion-contract ion be identical with the variable cont rac­

t ion of regions and levels of the past that we were discussing

earlier? Bergson's c o n t e x t seems to suggest that i t is, s ince he

constantly invokes t ransla t ion-contract ion with regard to sec ­

t ions of the c o n e , that is, levels ol the pas t . 2 5 Many considera­

t ions , however, lead us to the conc lus ion that whi le there is

obviously a relat ionship be tween the two con t rac t ions , they

are by no means ident ical . W h e n Bergson speaks of levels or

regions of the past, these levels are no less virtual than the past

in general; moreover , each o n e of them con ta ins the whole of

the past, but in a more or less con t rac ted state, around ce r ­

tain variable dominant r eco l l ec t i ons . T h e ex ten t o f the con­

traction, therefore, expresses the difference between one level

and another. On the other hand, when Bergson speaks of trans­

lat ion, it involves a movement that is necessary in the actual­

ization of a r e c o l l e c t i o n taken from a part icular level. Here

c o n t r a c t i o n no longer expresses trie o n t o l o g i c a l d i f ference

between two virtual levels, but the movement by which a rec­

o l l ec t ion is (psychological ly) actualized, at the same time as the

level that belongs to i t . 2 6

It would , in fact , be a mi s t ake to th ink tha t , in o r d e r to

be ac tua l i zed , a r e c o l l e c t i o n must pass th rough m o r e and

more cont rac ted levels in order to approach the present as the

supreme point of cont rac t ion or the summit of the c o n e . Th i s

would be an untenable interpretat ion lor several reasons. In


Page 65: Deleuze, Gilles. Bergsonism


the metaphor of the t o n e , even a level that is very cont rac ted ,

very c lose to the summit — so long as it is not actualized —

displays a genuine difference in kind from this summit , that

is, from the present, fu r the rmore , in order to actual ize a rec­

ol lec t ion , we do not have to change levels; if we had to do this,

the operat ion of memory would he imposs ib le . I or each rec­

o l l ec t ion has its own proper level; i t is t o o d i smembered or

dispersed in broader regions, too confined and muddled in nar­

rower regions. If we had to pass from one level to another in

order to actual ize each r eco l l e c t i on , each r eco l l ec t ion would

thus lose its individuality. Th i s is why the movement of trans­

lation is a movemen t by which the r eco l l ec t i on is actualized

at t h e same t i m e as its level : T h e r e is c o n t r a c t i o n because

recol lect ion-becoming-image enters into a "coa l e scence" with

the present. It therefore passes through "planes of consc ious­

ness" that put it in to e f lec t . But it does not pass through the

in termediate levels (which would prevent it from being put

into e f l e c t ) . Hence the need t o avoid confusing the planes of

consciousness, through which r eco l l ec t i on is actual ized, and the

regions, the sections or the levels of the past, according to which the

always virtual s ta te of r eco l l ec t i on varies. H e n c e the need to

distinguish intensive, on to log ica l con t rac t ion — where all the

levels coex i s t virtually, con t rac ted or relaxed (detendus) — and

translative, psychological con t rac t ion through which each rec­

ol lect ion on its own level (however relaxed [de'tendu] it is) must

pass in order to be actualized and thereby b e c o m e image.

But , on the o the r hand, Bergson says, there is rota t ion. In

its process of actualization, recol lec t ion does not confine itself

to carrying out this translation that unites it to the present; i t

also carries out this rotat ion on itsell in order to present its

"useful facet" in this union. Bergson does not clarify the nature


Page 66: Deleuze, Gilles. Bergsonism


of this rotation. We must make hypotheses on the basis of other

tex ts . In the movement of translat ion, it is therefore a whole

level of the past that is actual ized at the same t i m e as a par­

t icular r e c o l l e c t i o n . Each level thus linds itsell con t rac ted in

an undivided representat ion that is no longer a pure r eco l l ec ­

t ion, but is not yet , s tr ict ly speaking, an image. This is why

Bergson specifies that, from this point of view, there is no divi­

sion at this p o i n t . 2 7 R e c o l l e c t i o n undoubtedly has its individ­

uality. But how do we b e c o m e c o n s c i o u s ol it, how do we

distinguish it in the region that is actualized with it? We begin

From this undivided r ep re sen t a t i on ( tha t Bergson wil l cal l

"dynamic s c h e m e " ) , where all the reco l lec t ions in the process

of actualization are in a relationship ol reciprocal penetration;

and we develop it in dis t inct images that are external to one

another, that correspond to a particular r e c o l l e c t i o n . 2 8 Here

again, Bergson speaks of a success ion of "planes of consc ious­

ness." But the movement is no longer that of an undivided con­

t r a c t i o n . I t is, on the cont rary , that of a d iv i s ion , a deve lop­

men t , an expansion. R e c o l l e c t i o n can only be said to be actu­

alized when it has b e c o m e image. It is then , in lact , that it

enters not only into " c o a l e s c e n c e , " but in to a kind of circuit

with the p resen t , the r e c o l l e c t i o n - 7 m a g e referring back to

the percept ion- image and v ice versa. 2 1* H e n c e the preceding

m e t a p h o r o l " r o t a t i o n " wh ich prepares the ground lor this

launch into the c i rcu i t .

Thus , we have here t w o movements o f actual izat ion: one

of con t rac t ion , one ol expansion. We can see clearly that they

correspond c losely to the mul t ip le levels of the c o n e , some-

expanded (dc'tcnilus), some con t r ac t ed . For what happens in a

creature that conf ines i tsel f to dreaming? S ince sleep is l ike a

present s i tuat ion requiring noth ing but rest , with no interest

Page 67: Deleuze, Gilles. Bergsonism


Other than "dis in teres t , " it is as if the con t rac t ion were miss­

ing, as if the ex t remely expanded (de'tendu) relat ionship of the

r eco l l ec t ion with the present reproduced the most expanded

(de'tendu) level of the past itself. Conversely, what would hap­

pen in an au tomaton? It would he as though dispersion were

imposs ible , as though the d is t inct ion be tween images was no

longer carried into ef lect and onlv the most con t rac ted level

ol the past r e m a i n e d . 5 0 T h e r e is thus a c l o s e analogy be tween

the different levels of the c o n e and the aspects of actualization

for each level. It is inevitable that the latter will come to include the

former (hence the ambiguity that has already been pointed ou t ) .

Neve r the l e s s , we must no t confuse t h e m b e c a u s e t h e first

t heme conce rns the virtual variations of r eco l l ec t i on in itself;

the o ther , r e co l l ec t i on for us, the actualization of the recol ­

lec t ion in the r eco l l ec t ion - image .

Wha t is the framework c o m m o n to recol lec t ion in the pro­

cess of actualization (the recol lect ion-becoming-image) and the

p e r c e p t i o n - i m a g e ? T h i s c o m m o n f ramework i s m o v e m e n t .

Thus , it is in the relat ionship be tween the image and move­

ment , in the image's way of extending itself in movement , that

we must find the final m o m e n t s of actual izat ion: " the recol ­

lections need, for their actualization, a motor ally." 5 1 I lere again,

the ally is double . Some t imes percept ion is ex tended naturally

in movement ; a m o t o r tendency, a motor scheme, carries ou t a

d e c o m p o s i t i o n o f the pe rce ived i n t e r m s o f u t i l i t y . 5 2 I b i s

movement -percep t ion relationship would, on its own, be sul-

l ic ient to de l ine a recogni t ion that is purely au tomat ic , with­

out the i n t e rven t ion of r e c o l l e c t i o n s (o r , i f you prefer, an

instantaneous m e m o r y cons i s t ing ent i re ly in m o t o r m e c h a ­

n i s m s ) . However, r e c o l l e c t i o n s do i n te rvene . For, insofar as

recol lec t ion- images resemble actual percept ion, they are nec -


Page 68: Deleuze, Gilles. Bergsonism


cssarily ex tended into the movements that correspond to per­

cep t ion and they b e c o m e "adop ted" by i t . 5 '

Let us assume for a moment that a disturbance arises in this

movement -percep t ion-a r t i cu la t ion , a mechanical disturbance o f

the m o t o r s c h e m e : R e c o g n i t i o n has b e c o m e imposs ib l e (a l ­

though another type of recogni t ion subsists, as we see in those

patients who clearly describe an o b j e c t that is named to them,

but who do not know how to "make use" of it; or who cor ­

rect ly repeat what is said to t h e m , but no longer know how

to speak spontaneously) . T h e patient no longer knows how to

o r i en t himself , how to draw, that is, how to d e c o m p o s e an

o b j e c t according to the mo to r tendencies: His perception only

provokes diHuse movement s . Nevertheless, the r eco l l ec t ions

are there. Moreover, they cont inue to be evoked, to be embod­

ied in dis t inct images, that is, to undergo the translation and

rotat ion that character ize the first m o m e n t s ol actualizat ion.

W h a t is lacking therefore is the final m o m e n t , the final phase:

that of ac t ion . Just as the concomi t an t movements of percep­

t ion are disorganized, the reco l l ec t ion- image also remains as

useless, as ineffective as a pure reco l l ec t ion , and can no longer

ex tend itself into ac t ion . Th i s is the lirst important (act: T h e r e

are cases where r eco l l ec t ions survive despi te psychic or ver­

bal blindness or deafness . 5 4 ,

Let us move on to the second type ol movement-percept ion

re la t ionsh ip that de l ines the c o n d i t i o n s o f an a t t en t ive r ec ­

ogni t ion . It is no longer a mat te r of movements that " ex t end

our perception in order to draw useful el lects from it" and that

d e c o m p o s e the o b j e c t according to our needs , but of move­

men t s that abandon the effect , that bring us back to the o b j e c t

in order to restore its detail and completeness . Then the recol­

lect ion-images — which are analogous to present perception —

6 8

Page 69: Deleuze, Gilles. Bergsonism


take on a role that is "p reponderan t and no longer mere ly

accessory," regular and no longer acc iden t a l . 5 5 Let us assume

that this second kind of movement is disturbed (dis turbance

of the sensory m o t o r (unctions that is dynamic, and no longer

mechanical).* It is possible for automatic recognition to remain,

but what does appear to have disappeared is recollection itself.

Because such cases are the most frequent they have inspired

the traditional concept ion of aphasia as the disappearance of

recol lec t ions stored in the brain. Bergson's whole problem is:

W h a t has really disappeared?

First hypothesis: Is it pure recollection? Obviously not, since

pure recol lec t ion is not psychological in nature and is imper­

ishable. Second hypothesis: Is it the capacity to evoke reco l ­

lection, that is, to actualize it in a recollection-image? At times,

Bergson does express h imse l f in this way. 3 7 Nevertheless, it is

more compl ica ted than this. For the first two aspects of actu­

alization ( translat ion and ro ta t ion) depend on a psychic atti­

tude; the last t w o ( the t w o types of m o v e m e n t ) depend on

sensory-motrici ty and the attitudes of bodies . Whatever the

solidarity and complementar i ty of these two dimensions, the

one cannot comple te ly cance l out the other. W h e n only the

automatic movements of recognit ion are affected (mechanical

disturbances of sensory-motr ic i ty) , recol lec t ion nevertheless

comple t e ly retains its psychic actual iza t ion; i t preserves its

"normal a s p e c t , " but can no longer e x t e n d i tself in move­

ment , the corporeal stage of its actualization having b e c o m e

impossible. W h e n the movements of attentive recognition are

alfectcd (dynamic disturbances of sensory-motr ic i ty) , psychi­

cal actualization is undoubtedly more endangered than in the

preceding case for here the corporeal attitude really is a con­

dition of the mental attitude. Bergson nevertheless maintains


Page 70: Deleuze, Gilles. Bergsonism


that , o n c e again, no r eco l l ec t i on is " i n a t t e n t i v e . " T h e r e is

merely a "disturbance of the equ i l ib r ium." 5 8 We must perhaps

understand that the t w o psychic aspects of actualization sub­

sist but are, as it were, dissociated lor want ol a corporeal atti­

tude in which they could be inserted and combined. Sometimes

then, translation-contraction would occur, but would lack the

complementary movement of rotation, so that there woidd be

no distinct recol lec t ion- image (or, at least, a whole category

of recol lect ion- images would seem to have been abol ished) .

Some t imes , on the contrary, rotation would occu r , d is t inct

images would form, but they would be detached from memory

and abandon their solidarity with the others . In any case , it is

not sufficient to say that, according to Bergson, pure reco l l ec ­

t ion always preserves itself; we must add that i l lness never

abolishes the recol lec t ion- image as such, but merely impairs

a particular aspect of its actualization.

These , therefore, are the four aspects of actualization: transla­

tion and rotation, which form the properly psychic moments ;

dynamic movement , the attitude of the body that is necessary

to the stable equilibrium of the two preceding determinations;

and finally, mechanical movement , the motor scheme that rep­

resents the final stage of actualization. All this involves the adap­

tation of the past to the present, the utilization of the past in

terms of the present - what Bergson calls "at tent ion to l i fe ."

T h e first moment ensures a point of con tac t be tween the past

and the present: T h e past literally moves toward the present

in order to find a point of con tac t (o r of con t rac t ion) with it.

T h e second m o m e n t ensures a transposition, a translation, an

expans ion of the past in t h e present : R e c o l l e c t i o n - i m a g e s

restore the distinctions of the past in the present - at least those

that are useful. T h e third moment , the dynamic attitude of the


Page 71: Deleuze, Gilles. Bergsonism


body, ensures the harmony of the two preceding moments , cor­

rect ing the one by the o the r and pushing them to thei r l imit .

The fourth m o m e n t , the mechanica l movemen t of the body,

ensures the proper uti l i ty of the whole and its performance in

the present. But this utility, this performance, would be noth­

ing if the lour m o m e n t s were not c o n n e c t e d with a cond i t ion

that is valid lor them all . We have seen that pure r eco l l ec t ion

was con temporaneous with the present that it had been. R e c ­

o l l ec t i on , in the course of actualizing itself, thus tends to be

actualized in an image that is itself con temporaneous to this

present. Now it is obvious that such a recol lect ion-image, such

a " r e c o l l e c t i o n of the present ," would be comple t e ly useless

since it would simply result in doubling the perception-image.

R e c o l l e c t i o n must be e m b o d i e d , not in terms of its own pre­

sent (with which it is contemporaneous) , but in terms of a new

present, in relation to which it is now past. Th i s condi t ion is

normally realized by the very nature of the present, which con­

stantly passes by, moving forward and hol lowing out an inter­

val. Th i s is therefore the fifth aspect of actual izat ion: a kind

ol d isplacement by which the past is e m b o d i e d only in terms

of a present that is different from that which it has been. ( T h e

dis turbance corresponding to this last aspect would be param­

nesia, in wh ich the " r e c o l l e c t i o n of the p r e s e n t " would be

actualized as s u c h . ) 5 9

* * *

In this way a psychological unconscious, distinct from the onto­

logical unconscious, is defined. T h e latter corresponds to a rec­

o l lec t ion that is pure, virtual, impassive, inactive, in itself. T h e

former represents the movement of recol lec t ion in the course

ol actualizing itself: Like Leibnizian possibles, recol lect ions try

Page 72: Deleuze, Gilles. Bergsonism


to b e c o m e e m b o d i e d , they exert pressure to be admit ted so

that a lull-scale repression originat ing in the present and an

"a t ten t ion to l i fe" are necessary to ward oi l useless or danger­

ous r e c o l l e c t i o n s . 4 0 T h e r e i s no con t rad ic t ion be tween these

two descriptions of two distinct unconsciousnesses. Moreover,

the whole of Matter and Memory plays be tween the two , with

c o n s e q u e n c e s that we shall analyze later.

Page 73: Deleuze, Gilles. Bergsonism


O n e o r M a n y D u r a t i o n s ?

Thus far, the Bergsonian method has shown two main aspects,

the one dualist, the o the r monis t . First, the diverging lines or

the differences in kind had to be followed beyond the "turn

in e x p e r i e n c e " ; then, still further beyond, the point of con­

vergence of these l ines had to be rediscovered, and the rights

of a new monism res tored . 1 T h i s program is in fact realized in

Matter and Memory. First , we bring out the difference in kind

between the t w o lines of o b j e c t and subject : between percep­

t ion and r eco l l e c t i on , mat te r and memory, present and past.

W h a t happens then? I t certainly seems that when the reco l ­

l ec t ion is ac tual ized, its di f ference in kind from percep t ion

tends to be o b l i t e r a t e d : T h e r e are no longer , the re can no

longer be , anything but differences in degree be tween recol ­

l ec t ion - images and p e r c e p t i o n - i m a g e s . 2 It is for this reason

that, wi thout the method of intui t ion, we inevitably remain

prisoners of a badly analyzed psychological c o m p o s i t e whose

original differences in kind we are unable to discern.

But it is c lear that, at this level, a genuine point of unity is

not yet available. T h e point of unity must account for a c o m ­

posi te from the other side of the turn in expe r i ence ; it must not

7 ?

Page 74: Deleuze, Gilles. Bergsonism


be contused with the one in expe r i ence . And in l a d , Bergson

is not c o n t e n t to say that t h e r e are now only dif ferences in

degree b e t w e e n the r e c o l l e c t i o n - i m a g e and the percep t ion-

image. He also presents a much more important on to log ica l

proposi t ion: While the past coexists with its own present, and while

it coexists with itself on various levels of contraction, we must recog­

nise that the present itself is only the most contracted level of the past.

This t ime it is pure present and pure past, pure perception and

pure r eco l l ec t i on as such, pure mat ter and pure memory that

now have only differences of expansion (detente) and con t rac ­

t ion and thus rediscover an onto logica l unity. But discovering

a deepe r c o n t r a c t i o n - m e m o r y at the heart of r e c o l l e c t i o n -

memory we have thus laid the foundations for the possibili ty

ol a new monism. At each instant, our percept ion cont rac ts "an

inca lcu lab le m u l t i t u d e of r e m e m o r i / e d e l e m e n t s " ; a t each

instant, our present infini tely con t r ac t s our past: " T h e two

terms which had been separated to begin with cohere c losely

t o g e t h e r . . . . " 5 W h a t , in fact, is a sensation? It is the opera t ion

of con t rac t ing t r i l l ions ol vibrations o n t o a recept ive surface.

Qual i ty emerges from this, quality that is nothing o ther than

cont rac ted quantity. Th i s is how the not ion of cont rac t ion (o r

of tension) allows us to go beyond the duality ol homogeneous

quanti ty and he terogeneous quality, and to pass from one to

the o the r in a con t inuous movemen t . But , conversely, if our

present, through which we place ourselves inside matter, is the

most con t rac ted degree of our past, ma t t e r itsell will be like

an infinitely dilated or relaxed (detendu) past ( so relaxed that

the preceding m o m e n t has disappeared when the following

appears). Th i s is how the idea of relaxat ion (detente) — or of

extens ion — will ove rcome the duality of the unextended and

the extended and i»ive us the means of passim; from one to the


Page 75: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S ?

other , l o r percept ion itsell is extensi tv, sensation is extens ive

insofar as what i t c o n t r a c t s is p rec i se ly the e x t e n d e d , tin-

expanded (detendu). (It makes space available to us "in the exact

propor t ion" in which we have t i m e avai lable) . 4

Hence , the impor tance of Matter and Memory: Movement is

attr ibuted to things themselves so that material things partake

directly of duration, and thereby form a l imit case of duration.

The immediate data (les donees imme'diates) are surpassed: Move­

ment is no less outside me than in m e ; and the Sell itsell in

turn is only o n e case among others in durat ion. 5 But then all

kinds ol problems arise. Let us single out two important ones .

( I ) Is there not a cont rad ic t ion be tween the two m o m e n t s

of the me thod , be tween the dualism of di l lerences in kind and

the m o n i s m of c o n t r a c t i o n - r e l a x a t i o n (detente)? Lor, in the

name of the first, philosophies that confine themselves to dif­

ferences oi'degree, oi'intensity were condemned . Moreover, what

were c o n d e m n e d were the false not ions of degree, of inten­

sity, as no t ions of cont rar ie ty or negation, sources of all false

problems. Isn't Bergson now in the process of restoring all that ^

he o n c e dismissed? W h a t differences can the re be b e t w e e n

relaxation (detente) and con t rac t ion e x c e p t for the differences

of d e g r e e , of intensity? T h e present is only the most contracted

degree of the past, mat te r the most relaxed (de'tendu) degree

ol the present (mens momentanca).1' And if we seek to co r r ec t

what is too "gradual" here , we can only do so by re int roduc­

ing into durat ion all the contrar ie ty , all the oppos i t ion that

Bergson had previously c o n d e m n e d as so many abst ract and

inadequate c o n c e p t i o n s . We will only escape from mat te r as

deter iorat ion ol duration by embrac ing a c o n c e p t i o n ol mat­

ter that is a "reversal" of durat ion. 7 What then becomes of the

Bergsonian pro jec t of showing that Difference, as difference


Page 76: Deleuze, Gilles. Bergsonism


in kind, could and should be understood independently ol the

negative ( t he negative ol de ter iora t ion as well as the negative

ol opposi t ion)? T h e worst cont rad ic t ion of all seems to be set

up at the heart of" the s y s t e m . Every th ing is r e i n t r o d u c e d :

degrees , intensity, oppos i t ion .

( 2 ) Even suppos ing that th i s p r o b l e m is so lved , can we

speak of a rediscovered moni sm? In one sense, yes, insofar as

everything is durat ion. But , s ince duration is dissipated in all

these differences in degree, intensity, re laxat ion (detente), and

con t rac t ion that affect it, we tend instead to fall in to a kind

of quanti tat ive pluralism. H e n c e , the impor tance of the fol­

lowing ques t ion: Is duration one or many, and in what sense?

Have we really ove rcome dualism, or have we been engulfed

in pluralism? We must begin wi th this ques t ion .

* * *

Bergson's t ex t s seem to vary considerably on this point . Matter

and Memory goes furthest in the affirmation of a radical plurality

of durations: T h e universe is made up of modif icat ions , dis­

turbances , changes of tension and ol energy, and nothing e lse .

Bergson does indeed speak of a plurality of rhvthms of duration;

but in this c o n t e x t he makes it clear — in relation to durations

that are more or less slow or fast — that each duration is an

absolute , and that each rhythm is i tsel f a dura t ion . 8 In a key

text from 1903 , he insists on the progress made since Time and

Tree Will: Psychological duration, our duration, is now only one

case among o the rs , among an infinity of O t h e r s , "a ce r t a in

wel l -de l ined tension, whose very def in i t iveness s e e m s l ike a

c h o i c e between an infinity of possible durations."*' We can see

that, as in Matter and Memory, psychology is now only an open­

ing on to ontology, a springboard lor an "instal lat ion" in Being.


Page 77: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S ?

But no sooner are we installed, than we perceive that Being

is m u l t i p l e , the very n u m e r o u s dura t ion , our o w n , caught

b e t w e e n m o r e dispersed dura t ions and m o r e taut (tendue),

more intense durations: "Th i s being so one perceives any num­

ber of durations, all very different from one another " T h e

idea of a virtual c o e x i s t e n c e of all the levels of the past, of all

the levels of tension, is thus extended to the whole of the uni­

verse: This idea no longer simply signifies my relationship with

being, but the relationship of all things with being. Everything

happens as if the universe were a t remendous Memory. And

Bergson is pleased with the power of the me thod of intuit ion: xf

It a lone enables us " t o go beyond idealism as well as real ism,

to affirm the ex i s tence of ob jec t s which are inferior and superior

to ourselves, although st i l l , in a cer tain sense, internal to us,

to make them coexist together without difficulty." This extension

of virtual coex i s t ence to an infinity of specific durations stands

out clearly in Creative Evolution, where life i tsel f is compared to

a memory, the genera or spec ies corresponding to coex i s t ing

degrees of this vital m e m o r y . 1 0 Thus we have an on to log ica l

vision that seems to imply a general ized pluralism. But it is

precisely in Creative Evolution that a major l imitat ion is under­

l ined: If things are said to endure, it is less in themselves or

absolute ly than in re la t ion to the W h o l e of the universe in

which they part icipate insofar as their d is t inc t ions are artifi­

c ia l . Thus , the p iece of sugar only makes us wait because , in

spite of its arbitrary c a n i n g out, it opens out on to the universe

as a whole . In this sense, each thing no longer has its own dura­

t ion. T h e only ones that do are the beings similar to us ( p s y - ^

etiological duration), then the living beings that naturally form

relative c losed systems, and finally, the W h o l e of the uni­

v e r s e . " It is thus a l imi ted , not a generalized, pluralism.


Page 78: Deleuze, Gilles. Bergsonism


Finally, Duration and Simultaneity recapitulates all the pos­

s ib le hypotheses : genera l ized plural ism, l imi t ed plural ism,

m o n i s m . 1 2 Acco rd ing to the lirst , there i s a c o e x i s t e n c e of

comple t e ly different rhythms, ol durations that are really dis­

t inc t , hence a radical mul t ip l ic i ty ol T i m e . Bergson adds that

he o n c e advanced this hypothesis , but considered that apart

from ourselves it was valid only lor living species : " W e did not

see then , we still see today, no reason to ex t end this hypothe­

sis ol a m u l t i p l i c i t y ol dura t ions to the mater ia l un iverse . "

Hence , a second hypothesis: Material things outs ide us would

not be distinguished by absolutely different durations but by

a cer tain relative way of participating in our duration and of

giving it emphasis, f fere it seems that Bergson is condensing the

provisional doc t r ine of Time and Tree Will ( t he re is, as it were,

a myster ious part icipat ion of things in our duration, an " i n e x ­

pressible ground") and the more developed doctr ine oi Creative

Evolution (this participation in our duration would be explained

by things belonging to the W h o l e of the universe) . But even

in this second case, the mystery about the nature ol the W h o l e

and our relationship with it remains. Hence , the third hypoth­

esis: T h e r e is only a single t i m e , a single duration, in which

everything would part icipate, including our consc iousnesses ,

including living beings, including the whole material world.

Now, to the reader's surprise, it is this hypothesis that Bergson

puts forward as the most satisfactory: a single Time, one, univer­

sal, impersonal.^ In short, a monism o f T i m e Nothing could

be m o r e surprising; one of the o the r two hypotheses would

s e e m to be a b e t t e r express ion of the s tate of Be rgson i sm,

whether alter Matter and Mcmon or alter Creative Evolution. What

is m o r e : l ias Bergson forgotten that in Time and Tree Will he

defined duration, that is real t i m e , as a "mul t ip l i c i ty"?

7 8

Page 79: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S ?

W h a t has happened? Undoubted ly the confrontat ion with

the theory of Re la t iv i ty . T h i s c o n f r o n t a t i o n was forced on

Bergson because Relativity, for its part, invoked concep t s such

as expansion, con t rac t ion , tension and dilat ion in relation to

space and t ime . But this confrontation did not c o m e about sud­

denly: It was prepared by the fundamental no t ion of Mul t i ­

plicity, which Hinstein drew from Riemann, and which Bergson

lor his part had used in Time and Free Will. Let us recall , briefly,

the principal characteristics of Einstein's theory, as Bergson sum­

marizes t hem: Everything begins from a cer tain idea of move­

ment that entails a contract ion of bodies and a dilation of their

t ime . From this we conc lude that there has been a dislocation

of s imultaneity: W h a t is s imultaneous in a fixed system ceases

to be s imul taneous in a m o b i l e system. Moreover , by virtue

of the relativity of rest and m o v e m e n t , by virtue of the rela­

t ivity even o f a c c e l e r a t e d m o v e m e n t , t he se c o n t r a c t i o n s o f

extensi ty, these di lat ions o f t ime , these ruptures o f s imulta­

neity b e c o m e absolutely rec iproca l . In this sense there would

be a mul t ip l ic i ty of t imes , a plurality of t imes , with different

speeds of flow, all real, each one peculiar to a system of refer­

e n c e . And as it b e c o m e s necessary, in order to situate a point ,

to indicate its posit ion in t i m e as well as in space, the only

unity of t ime is in a fourth d imens ion of space . It is precisely

this S p a c e - T i m e b l o c that actually divides up into space and

into t ime in an infinity of ways, each one peculiar to a system.

To what does the discussion relate? Cont rac t ion , di la t ion,

relat ivi ty of m o v e m e n t , mul t ip l i c i ty — all these no t ions are

familiar to Bergson. He uses them lor his own purposes. Bergson

never gives up the idea that durat ion, that is to say t i m e , is

essent ia l ly m u l t i p l i c i t y . But the p r o b l e m is: W h a t type of

mul t ip l ic i ty? R e m e m b e r that Bergson opposed two types of

7 9

Page 80: Deleuze, Gilles. Bergsonism


mult ipl ici ty — actual mul t ipl ic i t ies that are numerical and dis­

cont inuous and virtual mul t ip l ic i t ies that are cont inuous and

quali tat ive. It is c lea r that in Bergson's terminology, Einstein 's

l i m e belongs to the first category. Bergson cr i t i c izes Einstein

lor having confused the t w o types ol mul t ip l ic i ty and lor hav­

ing, as a result , revived the confusion of t i m e with space . T h e

discussion only apparently deals with the quest ion: Is t ime one

or mul t ip le? T h e true p rob lem is " W h a t is the mul t ip l i c i ty

peculiar to t i m e ? " Th i s clearly surfaces in Bergson's uphold­

ing of the exis tence of a single, universal and impersonal T i m e .

" W h e n we are si t t ing on the bank of a river, the flowing of

the water, the gliding of a boat or the flight of a bird, the unin­

terrupted murmur of our deep life, are for us three different

things or a single one , at will " 1 4 Here Bergson endows atten­

t ion with the power of "appor t ioning wi thout dividing," " o f

being one and.several"; but more profoundly, he endows dura­

tion with the power to encompass itself. T h e flowing of the

water, the flight of the bird, the murmur of my life form three

l luxes; but only because my duration is one of t hem, and also

the e l emen t that contains the two others . W h y not make do

with t w o fluxes, my durat ion and the flight of the bird, for

example? Because the two fluxes could never be said to be coex­

istent or simultaneous if they were not contained in a third one .

T h e flight of the bird and my own duration are only simulta­

neous insofar as my own duration divides in two and is reflected

in another that conta ins it at the same t i m e as it conta ins the

flight of the bird: T h e r e is therefore a fundamental t r ipl ici ty

of l l u x e s . 1 5 It is in this sense that my duration essentially has

the power to disclose other durations, to encompass the others,

and to encompass itsell ad infinitum. But we see that this infin­

ity of ref lect ion or a t tent ion gives duration back its true char-

8 0

Page 81: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S '

actcr is t ics , which must be constantly recalled: It is not simply

the indivisible, but that which has a very special style ol divi­

sion; it is not simply succession but a very special coex i s t ence ,

a s imultanei ty ol l luxes. "Such is our first idea of s imul tane­

ity. We call simultaneous, then, two external lluxes that occupy

the same duration because they hold each o the r in the dura­

t ion ol a third, our own [ I t is this] s imultanei ty of fluxes

that brings us back to internal duration, to real dura t ion . " 1 6

l e t us return to the characteristics by which Bergson defines

durat ion as virtual or c o n t i n u o u s mul t ip l i c i ty . On the o n e

hand, it divides into e lements that diller in kind; on the other ,

these e l emen t s or these parts only actually exist insofar as the

division itself is effectively carried out (II our consciousness

" terminates the division at a given point, there also terminates

d ivis ib i l i ty .") . 1 7 If we take up a position where the division has

not yet been carried out , that is, in the virtual, it is obvious

that there is only a single t ime . T h e n , let us take up another

position at a m o m e n t where the division has been carried out :

two l luxes, lor e x a m p l e , that of Achi l les ' race and that of the

tor toise 's race. We say that they difler in kind (as do each step

of Achil les and each step of the tor to ise , i f we take the divi­

sion still further). T h e fact that the division is sub jec t to the

condi t ion ol actually being carried out means that the parts

( l luxes) must be lived or at least posited and thought of as capa­

ble ol being lived. Now Bergson's w hole thesis consists in dem­

onstrating that they can only be livable or lived in the perspective of a

single time. T h e principle o f the demonst ra t ion is as follows:

W h e n we admit the exis tence of several t imes, we are not con­

tent to cons ider flux A and flux B or even the image that the

sub jec t of A has of B (Achi l les as he conce ives or imagines the

to r to i se ' s race as capab le of be ing lived by the t o r t o i s e ) . In


Page 82: Deleuze, Gilles. Bergsonism


order to posit the e x i s t e n c e o l t w o t i m e s , we are lo rced to

int roduce a strange factor: the image that A has of B, whi le

nevertheless knowing that B cannot live in this way. Th i s lac-

tor is comple t e ly " s y m b o l i c " ; in o the r words, i t opposes and

. r exc ludes the lived expe r i ence and through it (and only i t ) is

the so-called second time realized, f rom this Bergson concludes

that there exis ts o n e T i m e and one T i m e onlv, as much on the

level of the actual parts as on the level ol the virtual W h o l e .

(But what is the significance ol this obscure demonst ra t ion?

We shall soon s e e . )

It we lollow the division in the o ther d i rec t ion , if we go

back , we see the lluxes each t i m e with their differences in kind,

with their differences of contraction and expansion (detente), c o m m u ­

nicat ing in a single and identical l i m e , which is, as it were ,

the i r condi t ion: "A single duration will pick up along its route

the events ol the total i ty ol the material world; and we will

then be able to el iminate the human consciousness that we had

initially had available, every now and then, as so many relays

lor the movement of our thought: there will now only be imper­

sonal t ime in which all things will How." 1 8 H e n c e the t r ipl ic-

ity ol lluxes, our duration ( the duration ol a spec ta to r ) be ing

necessary both as Mux and as representative ol l i m e in which

all l luxes are engulfed. It is in this sense that Bergson's various

tex ts are perfectly r econc i l ab le and conta in no cont rad ic t ion :

There is only one t ime ( m o n i s m ) , although there is an infin­

ity ol actual lluxes (generalized pluralism) that necessarily par­

ticipate in the same virtual whole ( l imited pluralism). Bergson

in no way gives up the idea of a difference in kind be tween

actual fluxes; any more than he gives up the idea ol differences

ot r e l axa t ion (detente) or c o n t r a c t i o n in the vir tual i tv that

encompasses them and is actualized in them. But he considers

8 2

Page 83: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S '

that these two certaint ies do not exc lude , but on the contrary

imply, a single t i m e . In short: Not only do virtual mul t ip l ic i ­

ties imply a single t i m e , but duration as virtual mul t ip l ic i ty is

this single and same Time.

It is nonetheless true that the Bergsonian demonstrat ion of

the con t r ad i c to ry cha rac t e r o f the plurali lv o f t imes s e e m s

obscure . Let us clarify it at the level of the theory of Relat iv­

ity. For, paradoxically, only this theory makes it appear c lea r

and convincing. Insofar as we are dealing with qualitatively dis­

t inct fluxes, it may in fact be difficult to know whether or not

the t w o subjec ts live and perceive the same t i m e : We support

unity, but only as the most "plausible" idea. On the other hand,

the theory ol Relat ivi ty is based on the following hypothesis:

T h e r e are no longer quali tat ive fluxes, but systems, "in a s tate

of reciprocal and uniform r ep l acemen t " where the observers

are in terchangeable , s ince there is no longer a privileged sys­

t e m . 1 9 Let us accept this hypothesis. Einstein says that the t ime

of the two systems, S and S ' , is not the same. But what is this

other t ime? It is no t that of Peter in S, nor that of Paul in S ' ,

s ince, by hypothesis, these two t imes only differ quantitatively,

and this difference is cancel led out when one takes S and S' as

systems of reference in turn. Could it at least be said that this

o the r t i m e is the one that Peter conce ives as lived or capable

ol being lived by Paul? Not at all — and this is the essential point

of the Bergsonian argument. "Undoubtedly Peter s t icks a label on

this T i m e in the name of Paul; but if he imagined Paul con­

scious, living his own duration and measuring it, lor this very

reason he would see Paul take his own system as a system of

reference , and then place himself in this single T i m e , internal

to each system, which we have just been speaking of: more ­

over, also for this very reason, Peter would provisionally sur-

8 ?

Page 84: Deleuze, Gilles. Bergsonism


render his system ol reference and in consequence his exis tence

as physicist, and in c o n s e q u e n c e also his consc iousness ; Pe te r

would now only see himself as a vision of Paul."- 0 In short, the

other t ime is someth ing that can nei ther he lived by Peter nor

by Paul, nor by Paul as Pe te r imagines h im. It is a pure sym­

bol excluding the lived and indicating simply that such a sys­

t em, and not the other , is taken as a reference point . " P e t e r

no longer envisages Paul as a physicist , nor even a consc ious

being, nor even a being: he empt i e s from his consc ious and

living interior the visual image of Paul, onlv retaining the exter­

nal envelope ol the character ."

Thus , in the Relativity hypothesis, it b e c o m e s obvious that

the re can only be a single livable and lived t ime . (Th is dem­

onstrat ion goes beyond the relativist hypothesis, s ince quali­

tative differences, in their turn, cannot cons t i tu t e numerical

dist inct ions.) This is why^Bergson claims that Relativity in fact

demonstrates the opposite of what it asserts about the plurality

of t i m e . 2 1 All Bergson's o ther c r i t ic i sms derive from this, f o r

what s imultanei ty does Einstein have in mind when he states

that i t varies from one system to the o ther? A s imul tanei ty

defined l>\ the readings ol two distant I locks . And it is true

that this s imu l t ane i t y is var iable or re la t ive . But p rec i se ly

because its relativity expresses, not something lived or livable,

but the symbolic factor of which we have just been speaking. 2 2

In this sense, this s imultanei ty presupposes t w o others l inked

in the instant, simultaneit ies that are not variable but absolute:

the s imul tanei ty be tween t w o instants, taken from external

movements (a nearby phenomenon and a moment of the c l o c k ) ,

and the s imul tanei ty of these instants with the instants taken

by them from our duration. And these two simultaneit ies pre­

suppose yet ano the r , that ol the l luxes , w h i c h is even less

8 4

Page 85: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S '

var iab le . 2 ' T h e Bergsonian theory o f simultaneity thus tends to

confirm the c o n c e p t i o n o f duration as the virtual coexistence o f

all the degrees of a single and identical t ime .

In short , from the first page o f Duration and Simultaneity to

the last, Bergson crit icizes Einstein tor having confused the vir­

tual and the actual ( the in t roduct ion of the symbol ic factor,

that is, ol a f ic t ion, expresses this confus ion) . I Ie is c r i t i c ized ,

therefore, for having confused the two types of mult ipl ici ty,

virtual and actual. At the heart of the question "Is duration one

or mul t ip le?" we find a comple t e ly different p rob lem: Dura­

t ion is a mult ipl ic i ty , but of what type? Only the hypothesis o f

a single T ime can, according to Bergson, account for the nature

of virtual mul t ip l ic i t ies . By confusing the two types — actual

spatial mult ipl ici ty and virtual temporal multiplicity - Einstein

has mere ly invented a new way of spatializing t ime . And we

cannot deny the originality of his s p a c e - t i m e and the stupen­

dous achievement it represents for s c i ence . (Spatialization has

never been pushed so far or in such a way . ) 2 4 But this achieve­

men t is that of a symbol for expressing compos i t e s , not that

of something exper ienced that is capable , as Proust would say,

of expressing "a l i t t le t i m e in the pure s ta te ." Being, or T i m e ,

is a multiplicity. But it is precisely not "mul t ip l e" ; it is O n e , in

conformi ty with to type of mult ipl ici ty .

* * *

W h e n Bergson defends the uniqueness of t i m e , he does not

retract anything he has said previously about the virtual c o e x ­

istence of various degrees ol relaxation (detente) and contract ion

and the difference in kind be tween lluxes or actual rhvthms.

W h e n he says that space and t i m e never over lap nor " i n t e r ­

twine , " when he maintains that only their dis t inct ion is r e a l , 2 5


Page 86: Deleuze, Gilles. Bergsonism


he does not retract any of the ambiguity ofMatter and Memory,

which consisted in integrating something of space into dura­

t ion , in o rder to find in durat ion a suff icient reason (raison

wffhante) of ex tens ion . Wha t he condemns from the start is

the whole combination of space and t ime into a badly analyzed

composite, where space is considered as ready made, and t ime,

in c o n s e q u e n c e , as a fourth d imens ion of s p a c e . 2 h And this

spatialization of t i m e is undoubtedly inseparable from sc ience .

But Relativity is characterized by its having pushed this spatiali­

zation forward, welding the composi te together in a completely

new way: For, in prerelat ivist s c i e n c e , t i m e assimilated to a

fourth d imension ol space is nevertheless an independent and

really dist inct variable. In Relativity, on the o the r hand, the

assimilation ol space to t ime is necessary in order to express

the invariance ol dis tance, so that it is expl ic i t ly introduced

into the ca lcula t ions and does not allow any real d is t inct ion

to subsist . In short . Relat ivi ty has formed an especially c lose -

knit mix tu re , but a mix tu re that is part of the Bergsonian cri­

t ique of the " c o m p o s i t e " in general .

On the o the r hand, from Bergson's point of view we can

(in fact we mus t ) conce ive of combina t ions that depend on a

comple t e ly different pr inciple , l e t us consider the degrees of

expansion (detente) and of contract ion, all ol which coexist with

one another : At the l imit of expansion (detente), we have mat­

t e r . 2 7 W h i l e undoubtedly, mat ter is not yet space, it is already

extensi ty. A duration that is infinitely slackened and relaxed

places its m o m e n t s outs ide one another ; one must have dis­

appeared when the o ther appears. W h a t these m o m e n t s lose

in rec iprocal penet ra t ion they gain in respect ive spreading.

W h a t they lose in tension they gain in ex tens ion . So that, at

each moment , everything tends to be spread out into an instan-


Page 87: Deleuze, Gilles. Bergsonism

O N E O R M A N Y D U R A T I O N S ?

t .meous, indefinitely divisible continuum, which will not pro­

long itsell in to the nex t instant, but will pass away, only to be

reborn in the following instant, in a flicker or shiver that con­

stantly begins again . 2 8 It would be sufficient to push this move­

ment of expansion (detente) to its limit in order to obtain space

(but space would then be found at the end of the line of dif­

ferentiation as the e x t r e m e ending that is no longer c o m b i n e d

with durat ion) . Space , in effect , is not mat te r or ex tens ion ,

but the " s c h e m a " of matter, that is, the representat ion of the

l imit where the movement of expansion (detente) would c o m e

to an end as the external envelope of all possible ex tens ions .

In this sense, it is not matter, it is not extensity, that is in space,

but the very o p p o s i t e . 2 9 And if we think that mat ter has a thou­

sand ways of b e c o m i n g expanded (detendu) or ex t ended , we

must also say that there are all kinds of dis t inct ex tens i t ies ,

all related, but still qualified, and which will finish by inter­

mingl ing only in our own schema of space.

T h e essential point is to see how expansion (detente) and con­

t r ac t i on are r e l a t ive , and re la t ive to o n e ano the r . W h a t i s

expanded (detendu) if not the con t rac ted — and what is con­

tracted i f not the ex tended , the expanded (detente)'! This is why

there is always extensity in our duration, and always duration in matter.

W h e n we perce ive , we con t r ac t mi l l ions o f vibrations o r e le ­

mentary shocks into a felt quality; but what we cont rac t , what

we " t e n s e " in this way, is matter, ex tens ion . In this sense there

is no point in wondering if there are spatial sensations, which

ones are or are not : All our sensat ions are ex tens ive , all are

"vo luminous" and ex tended , although to varying degrees and

in different styles, depending on the type of con t rac t ion that

they carry out . And qual i t ies be long to mat te r as much as to

ourselves: They belong to matter, they are in matter, by virtue

8 7

Page 88: Deleuze, Gilles. Bergsonism


of the vibrations and numbers that punctuate them internally.

I xtensi t ies arc- thus --till qualified, since ihe\ arc inseparable

from the contractions that become expanded (detendu) in them;

and matter is never expanded (detendu) enough to IK- pure space,

to stop having this min imum ol con t rac t ion through which i t

participates in duration, through which it is part of durat ion.

Conversely, duration is never contracted enough to be inde­

pendent ol the internal mat ter where it operates , and ol the

extension that it c o m e s to con t rac t . Let us return to the image

ol the inverted cone: Its point (our present) represents the most

con t r ac t ed point ol our durat ion; but i t also represents our

insertion in the least contracted, that is, in an infinitely relaxed

(detendu) matter. This is why, according to bergson, intelligence

has two correlative aspects, forming an ambiguity that is essen­

tial to it: It is acquaintance wi th matter, it marks our*adapta­

t ion to matter , i t molds itsell on mat ter ; but it only does so

by means of mind or duration, by placing itsell in mat te r in a

point of tension that allows it to master matter. In intel l igence,

one must therefore distinguish between form and sense: It has

its form in matter , it finds its form with matter , that is, in the

most expanded (detendu), but it has and finds its sense in the

most contracted, through which it dominates and utilizes mat­

ter. Tt might therefore be said that its form separates intel l i­

g e n c e from its meaning, but that this meaning always remains

present in it, and must be rediscovered by intui t ionTjThis is

why, in the final analysis, Bergson refuses all s imple genesis ,

which would account for intel l igence on the basis ol an already

presupposed order ol matter, or which would account lor the

phenomena ol mat te r on the basis ol the supposed categories

ol in te l l igence . T h e r e can only be a s imul taneous genesis of

mat te r and in te l l igence . O n e s tep for one , o n e s tep lor the


Page 89: Deleuze, Gilles. Bergsonism

O N E O B H A N V D U R A T I O N S ?

other : In te l l igence is con t rac ted in mat te r at the same t ime

as m a t t e r is expanded (detendu) in dura t ion ; bo th l ind the

form that is c o m m o n to them, their equi l ibr ium, in extensity,

even if in te l l igence in its turn pushes this form to a degree of

expansion (detente) that mat ter and extensi tv would never have

attained by themselves — that ol a pure s p a c e . 5 0

Page 90: Deleuze, Gilles. Bergsonism
Page 91: Deleuze, Gilles. Bergsonism


Elan Vital a s M o v e m e n t o f

D i f f e r e n t i a t i o n

Our problem is now this: By moving from dualism to monism,

Irom the idea of differences in kind to that of levels of expan­

sion (detente) and contrac t ion, is Bergson not reintroducing into

his philosophy everything that he had condemned — the dif­

ferences in degree and intensity that he so strongly cr i t ic ized

in Time and Free Will?1 Bergson says in turn that the past and

the present differ in kind and that the present is only the most

c o n t r a c t e d level or deg ree of the past: How can these t w o

proposi t ions be r econc i l ed? T h e problem is no longer that of

monism; we have seen how the coex i s t ing degrees of expan­

sion (detente) and cont rac t ion effectively implied a single t ime

in which even the " f luxes" were s imul taneous . T h e problem

is that of the harmony be tween the dualism of differences in

kind and the monism of degrees of expansion (detente), between

the two moments of the method or the two "beyonds" the turn

in e x p e r i e n c e — recognizing that the m o m e n t of dualism has

not been suppressed at a l l , - but comple t e ly retains its sense .

T h e c r i t i q u e of in tens i ty in Time and Tree Will is highly

ambiguous, Is it d i rec ted against the very not ion of intensive

quantity, or mere ly against the idea of an intensity of psychic

9 1

Page 92: Deleuze, Gilles. Bergsonism


states? It it is t rue that intensity is never given in a pure expe­

r ience , is it not then intensity that gives all the quali t ies with

which we make exper ience? I l ence , Matter and Memory recog­

nizes intensit ies, degrees or vibrations in the quali t ies that we

live as such ou t s ide ourse lves and tha t , as such , be long to

mat ter . T h e r e are numbers e n c l o s e d in qua l i t ies , in tens i t ies

included in durat ion. Here again, must we speak of a contra­

d ic t ion in Bergson? Or are there , rather, different m o m e n t s ol

the me thod , with the emphasis somet imes on one , somet imes

on another , but all coex i s t ing in a dimension of depth?

(1 ) Bergson begins by c r i t i c i z ing any vision of the world

based on differences in degree or intensity. T h e s e in tact lose

sight ol the essential point; that is, the articulations of the real

or the quali tat ive differences, the differences in kind. T h e r e

is a difference in kind between space and duration, mat ter and

memory , present and past, e t c . We only discover this differ­

ence by dint ol decompos ing the compos i t e s given in exper i ­

ence and going beyond the "turn." We discover the differences

in kind be tween t w o actual t endenc ie s , be tween two actual

d i rec t ions toward the pure state into which each c o m p o s i t e

divides. This is the m o m e n t ol pure dualism, or of the divi­

sion ol compos i t e s .

( 2 ) But we can already see that it is not enough to say that

the difference in kind is between t w o tendenc ies , be tween two

d i rec t ions , be tween space and duration For one ol these

two d i rec t ions takes all the differences in kind on itsell and

all the differences in degree tail away into the o the r d i rect ion,

the o t h e r tendency. It is duration that includes all the qualita­

tive differences, to the point where it is defined as alteration in

relation to i tsell . It is space that only presents differences in

degree, to the point where it appears as the schema ol an indel-


Page 93: Deleuze, Gilles. Bergsonism


inite divisibility. Similarly, Memory is essentially difference and

mat te r essentially repet i t ion . T h e r e is therefore no longer any

difference in kind be tween two tendenc ies , but a difference

between the differences in kind that correspond to one tendency

and the differences in degree that refer back to the o the r ten­

dency. Th i s i s the m o m e n t of neutralized, balanced dualism.

( 3) Duration, memory or spirit is difference in kind in i tself

and for itself; and space or ma t t e r is difference in degree out­

side itsell and for us. There fo re , be tween the two there are all

the degrees of difference or, in o the r words, the whole nature of

difference. Duration is only the most cont rac ted degree o f mat­

ter, mat ter the most expanded (detendu) degree o f duration. But

duration is l ike a naturing nature (nature naturante), and mat­

ter a natured nature (nature nature'e). Differences in degree are

the lowest degree ol Difference; differences in kind (nature) are

the highest nature of Dif ference . T h e r e is no longer any dual­

ism be tween nature and degrees . All the degrees c o e x i s t in a

single Nature that is expressed, on the one hand, in differences

in kind, and on the other , in differences in degree. Th i s is the

m o m e n t of mon i sm: All the degrees coex i s t in a single T i m e ,

which is nature in i t se l f . 2 T h e r e is no cont rad ic t ion be tween

this monism and dualism, as m o m e n t s of the me thod . For the

duality was valid be tween actual t endenc ies , be tween actual

d i r ec t ions leading beyond the first turn in e x p e r i e n c e . But

the unity occurs at a second turn: T h e c o e x i s t e n c e of all the

degrees , of all the levels is virtual, only virtual. T h e point of

unification is i t se l f virtual. This point is not wi thout similar­

ity to the O n e - W h o l e of the Platonists. All the levels of expan­

sion (detente) and contract ion coexis t in a single T ime and form

a total i ty; but this W h o l e , this O n e , are pure virtuality. Th i s

W h o l e has parts, this One has a number - but only potentially. 5


Page 94: Deleuze, Gilles. Bergsonism


This is why Bergson is not contradicting himself when he s|x-aks

ol different intensi t ies or degrees in a virtual c o e x i s t e n c e , in

a single T i m e , in a s imple Totality.

• * *

A philosophy like this assumes that the not ion of the virtual

stops being vague and indeterminate . In itself, it needs to have

the highest degree ol precision. Th i s condi t ion is only fulfilled

if, starting from monism, we are able to rediscover dualism and

account for it on a new plane. A fourth m o m e n t must be added

to the three preceding ones — that of dualism recovered, mas­

tered and in a sense, generated.

W h a t does Bergson mean when he talks about elan vital? It

is always a case of a virtualitv in the process of being actual­

ized, a s impl ic i ty in the process of differentiating, a total i ty

in the process of dividing up: Proceeding "by dissociation and

divis ion," by "d i cho tomy ," is the essence of l i f e . 4 In the most

familiar examples, life is divided into plant and animal; the ani­

mal is divided into instinct and intel l igence; an instinct in turn

divides into several d i rec t ions that are actualized in different

species ; in te l l igence itself has its particular modes or actual­

izations. It is as if Life were merged into the very movemen t

of differentiation, in ramified series. Movement is undoubtedly-

explained by the insertion ol duration into mat te r : Durat ion

is differentiated according to the obstacles it mee t s in matter,

according to the materiality through which it passes, according

to the kind ol ex tens ion that i t con t r ac t s . But differentiation

does not mere ly have an external cause. Durat ion is differen­

tiated within itsell through an internal explosive force; it is

only affirmed and prolonged, it only advances, in branching

or ramified ser ies . 5 Durat ion, to be precise, is called life when

9 4

Page 95: Deleuze, Gilles. Bergsonism


it appears in this movement . W h y is differentiation an "actual­

ization"? Because it presupposes a unity, a virtual primordial

total i ty that is dissociated accord ing to the lines ot differen­

t ia t ion , but that st i l l shows its subsis t ing unity and to ta l i ty

in each l ine. Thus , when life is divided into plant and animal,

when the animal is divided into instinct and inte l l igence, each

side of the division, each ramificat ion, carries the whole with

it. From a certain perspective it is like an accompanying nebu­

losity, testifying to its undivided origin. And there is a halo of

inst inct in in te l l igence , a nebula of in te l l igence in ins t inct , a

hint of the animate in plants, and of the vegetable in animals . 6

Differentiat ion is always the actual izat ion of a virtuality that

persists across its actual divergent l ines.

We then encounter a problem that is peculiar to Bergsonism:

T h e r e are t w o types of division that must not be confused.

According to the first type, we begin with a c o m p o s i t e , for

e x a m p l e the s p a c e - t i m e m i x t u r e o r t h e p e r c e p t i o n - i m a g e

and recol lect ion-image mixture . We divide this composi te into

two actual divergent lines that are different in kind and that

we e x t e n d beyond the turn in e x p e r i e n c e (pure ma t t e r and

pure duration, or e lse pure present and pure past) . But now

we are speaking of a c o m p l e t e l y dif ferent type of d iv is ion:

Our starting point is a unity, a simplicity, a virtual totality. This

unity is ac tua l ized a c c o r d i n g to d ivergent l ines differing in

kind; it "exp la ins , " it develops what it had kept enc losed in a

virtual manner . Fo r e x a m p l e , a t each instant pure durat ion

divides in two di rec t ions , one ol which is the past, the o the r

the p resen t ; or e l se the elan vital at every ins tant separa tes

into two movements , one of relaxation (detente) that descends

into matter , the o the r of tension that ascends into durat ion.

I t can be seen that the d ivergent l ines p roduced in the t w o


Page 96: Deleuze, Gilles. Bergsonism


types of division co inc ide and are superimposed, or at least cor­

respond c losely to each other . In the second type of division

we r ed i scove r d i f fe rences in k ind iden t ica l o r ana logous to

those that had been determined in the lirst type. In both cases

a vision of the world is c r i t ic ized for only taking account of

differences in degree where , more profoundly, there are dif­

ferences in kind. 7 In both cases a dualism is established between

t e n d e n c i e s that d i l l e r in k ind . But this is not the same s ta te

of dualism, and not the same division. In the lirst type, it is *

a ref lexive dual ism, wh ich results from the decomposition of an

impure composite: It const i tutes the first moment of the method.

In the second type it is a gene t i c dualism, the result of the dif­

ferentiation of a Simple or a Pure: It forms the linal m o m e n t of

the me thod that u l t imate ly rediscovers the starting point on

this new plane.

O n e quest ion b e c o m e s pressing: Wha t is the nature of this

one and simple Virtual? I low is it that, as early as Time and Tree

Will, then in Matter and Memon, Bergson's philosophy should

have attributed such importance to the idea of virtualitv at the

very m o m e n t when it was chal lenging the category of possi­

bili ty? It is because the "virtual" can be distinguished from the

" p o s s i b l e " from at least two points ol view. I r o m a cer tain

point ol view, in lact , the possible is the oppos i te of the real,

it is opposed to the real; but, in qu i te a different opposi t ion,

the virtual is opposed to the actual . We must take this termi­

nology seriously: T h e possible has no reality (al though it may

have an actual i ty) ; conversely, the virtual is not actual , but as

such possesses a reality. Mere again Proust 's formula best defines

the states ol virtualitv: "real wi thout being actual , ideal with­

out being abstract." On the o ther hand, or Irom another point

of view, the possible is that which is " rea l ized" (o r is not real-


Page 97: Deleuze, Gilles. Bergsonism


ized). Now the process of realization is subjec t to two essen­

tial rules, one o l resemblance and another of l imita t ion. l o r

the real is supposed to be in the image of the possible that it

realizes. ( I t simply has e x i s t e n c e or reality added to it, which

is translated by saying that, from the point of view of the con­

cep t , there is no di l lerence be tween the possible and the real.)

And, every possible is not realized, realization involves a l imi­

tation by which s o m e possibles are supposed to be repulsed

or thwarted, while o thers "pass" into the real. T h e virtual, on

the o the r hand, does not have to be realized, but rather actu­

alized; and the rules of actualization are not those of resem­

blance and l imi ta t ion , but those of difference or divergence

and of c rea t ion . W h e n cer ta in biologis ts invoke a not ion of

organic virtualitv or potentialitv and nonetheless maintain that

this potentiali ty is actualized by s imple l imitation of its global

capacity, they clearly tall in to a confusion of the virtual and

the poss ib le . 8 For, in order to be actualized, the virtual can­

not proceed by e l imina t ion or l imi ta t ion, but must create its

own lines of actual izat ion in positive acts . T h e reason for this

is s imple : W h i l e the real is in the image and likeness of the

possible that it realizes, the actual , on the o the r hand does not

resemble the virtualitv that it embod ie s . It is difference that

is pr imary in the process of ac tua l i za t ion — the d i f fe rence

be tween the virtual from which we begin and the actuals at

which we arrive, and also the d i l le rence be tween the c o m p l e ­

mentary lines according to which actualization takes place. In

short , the charac te r i s t ic of virtualitv is to exis t in such a way

that it is actualized by being differentiated and is forced to dif­

ferentiate i tsell , to crea te its l ines of differentiation in order

to be actual ized.

W h y does Bergson chal lenge the not ion of the possible in


Page 98: Deleuze, Gilles. Bergsonism


favor of that of the virtual? It is precisely because - by virtue

ol these p reced ing c h a r a c t e r i s t i c s — the poss ib le is a false

not ion , the source of false p rob lems . T h e real is supposed to

resemble it. That is to say, we give ourselves a real that is ready-

made, preformed, pre-existent to itsell , and that will pass into

e x i s t e n c e a c c o r d i n g t o a n o r d e r o f succes s ive l i m i t a t i o n s .

Every th ing is already complete!} given: all o f t h e real in t h e

image, in the pseudo-actuality of the possible. Then the sleight

of hand b e c o m e s obvious: If the real is said to resemble the »

possible, is this not in fact because the real was expec t ed to

c o m e about by its own means, to "p ro j ec t backward" a ficti­

tious image of it, and to claim that it was possible at any t ime ,

before it happened? In (act, it is not the real that resembles

the possible, it is the possible that resembles the real, because

i t has been abs t rac ted from the real o n c e made , arbitrari ly

ex t rac ted Irom the real like a s ter i le d o u b l e . 9 H e n c e , we no

longer understand anything e i t he r of the mechanism of differ­

e n c e o r o f the mechanism of c rea t ion .

Evolution takes place from the virtual to actuals. Evolution

is actual izat ion, actual izat ion is c rea t ion . W h e n we speak of

biological or living evolution we must therefore avoid two mis­

c o n c e p t i o n s : that of interpret ing it in terms of the "pos s ib l e "

that is realized, or e lse interpret ing it in t e rms ol pure actu­

als. T h e first misconcept ion obviously appears in prcformism.

And, contrary to preformism, evolut ionism will always have

the meri t ol reminding us that life is product ion, creat ion of

differences. T h e whole problem is that of the nature and the

causes of these differences. T h e vital differences or variations

can certainly be conce ived ol as purely acc identa l . But three

o b j e c t i o n s to an interpretat ion ol this kind arise:

(1 ) s ince they are due to c h a n c e , these variat ions, how-


Page 99: Deleuze, Gilles. Bergsonism



ever small they are, would remain external , " indifferent" to

each other ;

( 2 ) s ince they are external , they could not logically en t e r

in to anything but re la t ions ol assoc ia t ion and addi t ion wi th

one another ;

( 3 ) s ince they are indifferent, they could not even have the

means to really en t e r into such relat ions (tor there would be

no reason why the small successive variations should link up

and add t o g e t h e r in the same d i r e c t i o n ; no r any reason for

sudden and s imul taneous variations to be coordinated into a

livable w h o l e ) . ' "

I f we invoke the ac t ion of the envi ronment and the influ­

e n c e of external cond i t i ons , the three o b j e c t i o n s persist in

another form: For the differences are still interpreted from the

perspect ive of a purely external causality. In thei r nature they

would only be passive effects, e lements that could be abstractly

combined or added together. In their relationships they would,

however, be incapable of functioning "as a b l o c , " so as to con­

trol or utilize their c a u s e s . "

T h e mistake of evolut ionism is, thus, to c o n c e i v e of vital

variations as so many actual de terminat ions that should then

c o m b i n e on a single l ine. T h e three requirements of a philoso­

phy of life are as follows:

(1) the vital difference can only be experienced and thought

of as internal difference; it is only in this sense that the " ten­

dency to c h a n g e " is no t a c c i d e n t a l , and that the variat ions

themselves find an internal cause in that t endency ;

( 2 ) these variations do not en t e r into relationships of asso­

ciat ion and addition, but on the contrary, they en te r into rela­

t ionships o f dissociat ion o r division;

( 3 ) they the re fo re involve a vi r tual i tv that is ac tua l i zed

Page 100: Deleuze, Gilles. Bergsonism


according to the lines oi d ivergence ; so that evolution does not

move irom one actual term to another actual te rm in a h o m o ­

geneous unil inear series, hut Irom a virtual t e rm to the het­

erogeneous terms that actualize it along a ramified s e r i e s . 1 2

But this leads to the ques t ion of how the S imp le or the

O n e , " the original identity," has the power to be differentiated.

T h e answer is already contained in Matter and Memon. And the

linkage between Creative Evolution and Matter and Memory is per­

fectly r igorous. We know that the virtual as virtual has a reality;

this reality, extended to the whole universe, consists in all the

c o e x i s t i n g degrees of expansion (detente) and c o n t r a c t i o n . A

gigant ic memory, a universal c o n e in which everything c o e x ­

ists wi th itself, e x c e p t for the differences of level. On each of

these levels there are s o m e "outs tanding poin ts , " which are

like remarkable points peculiar to it. All these levels or degrees

and all these points are themselves virtual. They be long to a

single T i m e ; they coexis t in a Unity; they are enclosed in a Sim­

plici ty; they form the potent ial parts of a W h o l e that is itself

virtual. T h e y are the reality of this virtual. Th i s was the sense of

the theory of virtual mul t ip l ic i t ies that inspired Bergsonism

from the start. W h e n the virtualitv is actualized, is differenti­

ated, is "deve loped ," when it actualizes and develops its parts,

i t does so accord ing to l ines that are divergent , but each of

which corresponds to a particular degree in the virtual total­

ity. T h e r e is here no longer any c o e x i s t i n g w h o l e ; there are

merely l ines of actual izat ion, some successive, others simultaneous,

but each represent ing an ac tua l iza t ion of the w h o l e in o n e

d i rec t ion and not c o m b i n i n g with o the r lines or o ther d i rec­

t ions. Nevertheless , each of these l ines corresponds to one of

these degrees that all c o e x i s t in the virtual; it ac tua l izes its

l eve l , w h i l e separa t ing i t from t h e o t h e r s ; i t e m b o d i e s its


Page 101: Deleuze, Gilles. Bergsonism


prominent points, while being unaware of everything that hap­

pens on o the r l e v e l s . 1 5 We must think of i t as follows: W h e n

duration is divided into mat te r and life, then life into plant

and animal , different levels of con t rac t ion , which only c o e x ­

ist insofar as they remain virtual, are actualized. And when the

animal instinct is i t se l f divided into various instincts, or when

a particular inst inct is i t s e l f divided according to species , lev­

els are again separated, or arc actually cu t out in the region of

the animal or of the genus. And however s t r ic t ly the lines of

actualization correspond to the levels or the virtual degrees of

expansion (detente) or c o n t r a c t i o n , i t should no t be thought

that the l ines o f ac tual izat ion conf ine themse lves to t racing

these levels or degrees, to reproducing them by s imple resem­

blance . For what coex i s t ed in the virtual ceases to coex i s t in

the actual and is dis t r ibuted in lines or parts that cannot be

summed up, each one retaining the whole , e x c e p t from a cer­

tain p e r s p e c t i v e , from a ce r t a in po in t of view. T h e s e l ines

ol differentiation are therefore truly creat ive: T h e y only actu­

alize by inventing, they c rea te in these condi t ions the phvsi-

cal , vital or psychical representat ive of the on to log ica l level

that they embody.

I f we c o n c e n t r a t e only on the actuals that c o n c l u d e each

l ine, we establish relationships be tween them — whe the r of

gradation or oppos i t ion . Be tween plant and animal , for exam­

ple, be tween animal and man, we now only see differences in

degree. Or we will situate a fundamental opposi t ion in each

one of t hem: We will see in one the negative of the other , the

inversion of the o ther , or the obs tac le that is opposed to the

other. Bergson often expresses h imse l f in this way, in terms of

cont rar ie tv : Mat te r is presented as the obs tac le that the elan

vital must get around, and materiality, as the inversion of the


Page 102: Deleuze, Gilles. Bergsonism

Summary Diagram of Differentiation (CE, Ch. 2)






/ \









Page 103: Deleuze, Gilles. Bergsonism


movemen t of l i f e . ' 4 I t should not , however, he thought that

Bergson is going back to a c o n c e p t i o n of the negative that he

had previously c o n d e m n e d , any more than he returns to a the­

ory of de ter iora t ions . For one only has to replace the actual

terms in the movement that produces them to bring them back

to the virtualitv actualized in t hem, in order to see that dif­

ferentiation is never a negation but a c rea t ion , and that differ­

e n c e is never negative but essentially positive and creat ive.

» * *

We always rediscover the laws c o m m o n to these lines ol actu­

alization or ol differentiat ion. T h e r e is a corre la t ion be tween

life and matter , be tween expansion (detente) and con t rac t ion ,

which shows the c o e x i s t e n c e of the i r respect ive degrees in the

virtual W h o l e , and their essential relativity in the process of

actualizat ion. Fach line of life is related to a type of ma t t e r

that is not mere ly an external env i ronment , but in terms of

which the living being manufactures a body, a form, for itself.

This is why the living being, in relation to matter, appears pri­

marily as the stating of a p rob lem, and the capaci ty to solve

problems: T h e construct ion of an eye, for example , is primarily

the solut ion to a problem posed in terms of l i gh t . 1 5 And each

t ime , we will say that the solution was as good as it could have

been, given the way in which the problem was stated, and the

means that the living being had at its disposal to solve it . ( I t

is in this wav that, if we compare a similar inst inct in various

species , we ought not to say that it is more or less c o m p l e t e ,

more or less perfected, but that it is as perfect as it can be in

varying d e g r e e s . ) 1 6 It is nevertheless c lear that each vital solu­

t ion is not in itself a success : By dividing the animal in two ,

Arthropods and Vertebrates , we have not taken into account


Page 104: Deleuze, Gilles. Bergsonism


the- two o ther d i rec t ions , I c h inodcrms and Mollnsks, which

are a setback lor the clan vital.1"1 Everything takes place as though

living beings themselves also stated false problems lor them­

selves in which they risk losing thei r way. Moreover , if every

solut ion is a relative success in relation to the condi t ions of

the problem or the envi ronment , it is still a relative se tback,

in relat ion to the movement that invents it: l.ile as movement

alienates itsell in the material form that it c rea tes ; by actualiz­

ing i tsel l , by di l lerent ia t ing i tsel l , it loses " c o n t a c t with the

rest ol i tsel l ." Every species is thus an arrest ol movement ; it

could be said that the living be ing turns on i t s e l f and closes

itse//. 1 8 It cannot be otherwise , s ince the W h o l e is only virtual,

dividing itsell by being acted out . It cannot assemble its actual

parts that remain external to each o ther : I he W h o l e is never

"given." And, in the actual, an i r reducible pluralism reigns —

as many worlds as living beings, all " c l o s e d " on themselves .

But we must , in another osc i l la t ion , be del ighted that the

Whole is not given. Th i s is the cons tant t h e m e of Bergsonism

Irom the outset : T h e confusion of space and t i m e , the assimi­

lation of t ime into space, make us think that the whole is given,

even if only in pr inciple , even if only in the eyes of God. And

this is the mistake that is c o m m o n to mechanism and to final-

ism. I he lormer assumes that everything is calculable in terms

of a state; the latter, that everything is de te rminab le in terms

of a program: In any event , t ime is only there now as a screen

that hides the eternal Irom us, or that shows us successively

what a ( i od or a superhuman in te l l igence would see in a sin­

g le g l a n c e . ' 1 ' Now this i l lusion is i nev i t ab l e as soon as we

spatialize t ime. Indeed, in space it is sufficient to have a dimen­

sion supplementary to those where a phenomenon happens lor

the movement in the course ol happening to appear to us as a


Page 105: Deleuze, Gilles. Bergsonism


ready-made form. II we consider t i m e as a fourth d imension

of space , this fourth d imension will thus be assumed to con­

tain all the possible forms of the universe as a whole ; and move­

ment in space, as well as (lowing in t i m e , will now only be

appearances linked to the three d i m e n s i o n s . ' 0 But the lact that

real space has only three dimensions, that T ime is not a dimen­

sion ol space, really means this: T h e r e is an e l l icaci ty , a posi-

tivity ol t ime , that is identical to a "hes i ta t ion" of things and,

in this way, to crea t ion in the wor ld . 2 1

It is c lear that there is a W h o l e of duration. But this whole

is virtual. It is actualized according to divergent lines; but these

lines do not form a whole on thei r own accoun t , and do not

resemble what they actualize. If the c h o i c e is be tween mecha­

nism and finalism, tinalism is preferable; providing that it is cor­

rected in two ways. O n the one hand, it is right to compare the

living be ing to the whole of the universe, but i t is wrong to

interpret this compar ison as if it expressed a kind of analogy

be tween two c losed to ta l i t ies ( m a c r o c o s m and m i c r o c o s m ) .

The finality of the living being exists only insofar as it is essen­

tially open on to a total i ty that is i t se l f open: "finality is ex te r ­

nal , or i t is no th ing at a l l . " 2 2 I t is thus the w h o l e classical

comparison that takes on another meaning; it is not the whole

that c loses like an organism, it is the organism that opens on to

a who le , like this virtual who le .

On the o ther hand, there is a proof of finality to the ex tent

that we discover similar actualizations, identical s tructures or

apparatuses on divergent lines (for example , the eye in the Mol-

lusk and in the Ver tebra te ) . T h e example wil l be all the more

significant the further apart the lines are, and the more the organ

that is s imilar is obta ined by dissimilar m e a n s . 2 ' We see here

how, in the process of actualization, the very category of resem-


Page 106: Deleuze, Gilles. Bergsonism


Nance finds itsell subordinated to that ol divergence, difference

or differentiation. W h i l e actual forms or products can resem­

ble each other, the movements of production do not resemble

each other , nor do the products r e semble the virtualitv that

they embody. Th i s is why actual izat ion, differentiat ion, are a

genuine creat ion. T h e W h o l e must create the divergent lines

accord ing to wh ich it is actual ized and the dissimilar means

that it uti l izes on each l ine. T h e r e is finality because life does

not operate without directions; but there is no "goal ," because

these d i rec t ions do not pre-exist ready-made, and are them­

selves crea ted "along wi th" the act that runs through t h e m . 2 4

Each l ine of actual izat ion corresponds to a virtual level; but

each t ime , it must invent the figure of this correspondence and

crea te the means for the deve lopment of that which was only

enveloped in order to distinguish that which was confused.

* * *

Duration, Life, is in principle (en droit) memory, in principle con­

sciousness, in pr inciple f reedom. " In pr inc ip le" means virtu­

ally. T h e whole quest ion (quid facti?) is knowing under what

c o n d i t i o n s dura t ion b e c o m e s in fact c o n s c i o u s n e s s o f se l l ,

how life actually accedes to a mem or y and freedom of f a c t . 2 5

Bergson's answer is that it is only on the line of Man that the

elan vital successfully "ge ts through"; man in this sense fa " t h e

purpose of the en t i re process of e v o l u t i o n . " 2 6 I t could be said

that in man, and only in man, the actual b e c o m e s adequate to

the virtual. It could be said that man is capable of rediscover­

ing ail t h e levels , all t h e degrees of expans ion (detente) and

con t rac t ion that coexis t in the virtual W h o l e . As if he were

capable of all the frenzies and brought about in h i m s e l f suc­

cess ively everything that , e l s ewhe re , can only be e m b o d i e d


Page 107: Deleuze, Gilles. Bergsonism


in di f ferent s p e c i e s . Even in his d reams he red i scovers or

prepares mat ter . And durations that arc inferior or superior to

him are still internal to him. Man therefore crea tes a differen­

t iat ion that is valid for the W h o l e , and he alone traces out an

open d i r e c t i o n tha t is ab l e to express a w h o l e tha t is i t s e l f

open. W h e r e a s the o ther d i rec t ions are c losed and go round

in c i r c l e s , whereas a dis t inct " p l a n e " of nature corresponds to

each o n e , man is capable of scrambling the planes, of going

beyond his own plane as his own cond i t ion , in order finally to

express naturing N a t u r e . 2 7

f low does this privilege of man c o m e about? At first sight,

its origin is a humble o n e . Every con t rac t ion of duration still

be ing relat ive to an expans ion (detente), and every life to a

ma t t e r , t h e po in t of depar ture is in a ce r t a in s ta te of c e r e ­

bral mat ter . We recall that this la t ter "analyzed" the received

e x c i t a t i o n , s e l e c t e d t h e r eac t ion , made poss ib le an interval

b e t w e e n e x c i t a t i o n and r eac t ion ; noth ing here goes beyond

the physico-chemical propert ies of a particularly c o m p l i c a t e d

type of matter. But, as we have seen, it is the whole of memory

that descends in to this interval, and that b e c o m e s actual. It is

the whole of freedom that is actual ized. On man's l ine of dif­

ferent iat ion, the elan vital was able to use ma t t e r to crea te an

ins t rument of f r eedom, " t o make a m a c h i n e wh ich should

tr iumph over m e c h a n i s m , " " t o use the de te rmin ism of nature

to pass through the meshes of the ne t which this very deter­

m i n i s m had s p r e a d . " 2 8 F r e e d o m has p rec i se ly this physical

sense: " t o de tona te" an explosive, to use it for more and more

powerful m o v e m e n t s . 2 9

But where does this starting point seem to lead? To percep­

t ion; and also to a utilitarian memory , s ince useful r eco l l ec ­

tions are actualized in the cerebral interval; and to intell igence


Page 108: Deleuze, Gilles. Bergsonism


as the organ of domina t ion and util ization of mat te r . We even

understand that men form societies. It is not that society is solely

or essent ia l ly in te l l igent . From the outse t , human soc i e t i e s

undoubtedly imply a certain intelligent comprehension ol needs

and a cer ta in rational organization of ac t iv i t ies . But they are

also formed, and only subsist through irrational or even absurd

factors, 'fake, lor example , obligation: It has no rational ground.

F.ach particular obl igat ion is convent ional and can border on

the absurd; the only thing that is grounded is the obl igat ion

to have obl igat ions , " t h e whole of obl igat ion"; and it is not

grounded in reason, but in a r equ i rement of nature, in a kind

of "virtual ins t inc t , " that is, on a counterpar t that nature pro­

duces in the reasonable being in order to compensa t e lor the

partiality ol his intel l igence. F.ach line ol differentiation, being

exclusive, seeks to recapture, by its own means, the advantages

of the o the r l ine. T h u s , in thei r separation, inst inct and intel­

l igence are such that the one produces an ersatz of intel l igence,

the other , an equivalent ol ins t inct . Th i s is the "story-tel l ing

func t ion" : virtual i n s t inc t , c r ea to r o l gods, inven to r o f reli­

g ions , that is, ol f ic t i t ious representat ions "wh ich will stand

up to the representat ion ol the real and which will succeed ,

by the intermediary ol in te l l igence i tsel l , in thwarting intel­

lectual work." And as in the case ol obligation, each god is con­

t ingen t , or even absurd, but what is natural, necessary and

grounded is having gods; it is the pantheon ol gods.*" In short,

sociabi l i ty (in the human sense) can only exis t in intel l igent

beings, but it is not grounded on the i r in te l l igence : Social life

is immanen t to in t e l l i gence , it begins with it but does not

derive from it. H e n c e our problem appears to have b e c o m e

more c o m p l i c a t e d instead ol being solved, f o r if we cons ider

intell igence and sociability, both in their complementar i ty and


Page 109: Deleuze, Gilles. Bergsonism


in their difference, nothing yet justifies man's privilege. T h e

soc ie t i e s that he forms are no less c losed than animal species;

they form part of a plan (plan) of nature, as much as animal

species and societ ies; and man goes round in c i rc les in his soci­

e ty jus t as much as the species do in theirs or ants in thei r

d o m a i n . " Nothing here seems to be capable of giving man the

previously m e n t i o n e d excep t iona l opening , as the power of

going beyond his " p l a n e " (plan) and his cond i t ion .

Unless this kind of play of in te l l igence and of society, this

small interval be tween the two , is i t s e l f a decisive factor. T h e

small intracerebral interval has already made in te l l igence pos­

s ib le , and the ac tua l iza t ion of a m e m o r y useful. Moreover ,

thanks to it, the body imitates the whole life of the mind, and

we were able with a leap to place ourselves in the pure past.

We now find ourselves before another intercerebral inten'al between

in te l l igence itself and soc ie ty : Is it not this "hes i ta t ion" of the

intelligence that will be able to imitate the superior "hesitation"

of things in duration, and that will al low man, wi th a leap, to

break the c i r c l e of c losed soc ie t ies? At first sight, the answer

is no. For, if in te l l igence hesi tates and some t imes rebels , it is

primarily in the name ol an egoism that it seeks to preserve

against social r e q u i r e m e n t s . ' 2 And whi le soc ie ty makes itself

obeyed it is thanks to the story-tell ing function, which per­

suades the in te l l igence that it is in its interest to confirm the

social obligation. We therefore seem to be constantly sent back

from one term to another. But everything changes when some­

thing appears in the interval.

W h a t is it that appears in the interval be tween intel l igence

and society (in the same way as the recollection-image appeared

in the cerebral interval appropriate to in te l l igence)? We can­

not reply: It is intuition. In fact, we must on the contrary carry


Page 110: Deleuze, Gilles. Bergsonism


out a genesis ot in tui t ion, that is, de te rmine the way in which

intel l igence itsell was converted or is converted into intuition.

And i f we recal l , accord ing to the laws of di l lerent ia t ion, that

i n t e l l i g e n c e , in separat ing i t s e l f from ins t inc t , never the less

keeps an equivalent of inst inct that would be like the nucleus

of intuition. We are not saying anything of importance, tor this

equivalent ol inst inct tinds i t se l f comple t e ly mobi l ized in the

c losed soc ie ty as such, through the s tory-tel l ing f u n c t i o n . 5 5

Bergson's real answer is comple te ly different: Wha t appears in

the interval is emotion. In this answer, " W e have no c h o i c e . " 5 4

O n l y e m o t i o n differs in nature from bo th i n t e l l i g e n c e and

ins t inc t , from both in te l l igent individual egoism and quasi-

ins t inct ive social pressure. Obviously no one denies that ego­

ism produces emot ions ; and even more so social pressure, with

all the fantasies of the story-telling function. But in both these

cases, emot ion is always connected to a representation on which

it is supposed to depend . We are then placed in a c o m p o s i t e

ol e m o t i o n and ol representat ion, wi thout no t ic ing that it is

potential (en puissance), the nature ol emot ion as pure e lement .

T h e latter in tact precedes all representat ion, itself generating

new ideas. It does not have, s t r ic t ly speaking, an o b j e c t , but

mere ly an essence that spreads itself over various o b j e c t s , ani­

mals, plants and the whole of nature. "Imagine a piece ol music

which expresses love. It is not love lor a particular person

T h e quality of love will depend upon its essence and not upon

its o b j e c t . " 5 1 Although personal, it is not individual; transcen­

dent , it is like the God in us. " W h e n music c r i es , it is human­

ity, it is the whole of nature which cries with it. Truly speaking,

it does not introduce these feelings in us; it introduces us rather

into them, like the passers-by that might l>e nudged in a dance."

In short , e m o t i o n is crea t ive (first , because it expresses the


Page 111: Deleuze, Gilles. Bergsonism


whole of c rea t ion , then because i t creates the work in which

it is expressed; and finally, because it c o m m u n i c a t e s a l i t t le

ol this creat ivi ty to specta tors or hearers) .

T h e l i t t le interval "be tween the pressure ol society and the

resis tance ol i n t e l l i gence" defines a variability appropriate to

human s o c i e t i e s . Now, by means of this interval , some th ing

extraordinary is produced or embodied: creative emot ion . This

no longer has anything to do with the pressures of society, nor

with the disputes of the individual. I t no longer has anything

to do with an individual who contests or even invents, nor with

a socie ty that constrains, that persuades or even tells s t o r i e s . 5 6

It has only made use of the i r c i rcular play in order to break

the c i r c l e , just as Memory uses the c i rcular play of exc i ta t ion

and react ion to embody r eco l l ec t ions in images. And what is

this crea t ive e m o t i o n , il not precisely a c o s m i c Memory, that

actualizes all the levels at the same t ime , that l iberates man

Irom the plane (plan) or the level that is proper to him, in order

to make him a creator , adequate to the whole movemen t ol

c rea t ion? 5 7 This liberation, this embodiment ol cosmic memory

in crea t ive e m o t i o n s , undoubtedly only takes place in privi­

leged souls. It leaps Irom one soul to another , "every now and

then ," crossing closed deserts. But to each m e m b e r of a c losed

society", if he opens h imse l f to it, i t c o m m u n i c a t e s a kind ol

r emin i scence , an e x c i t e m e n t that allows him to lollow. And

from soul t o soul, it traces the design o f an open society, a soci­

ety of c rea tors , where we pass from one genius to another ,

through the intermediary of disciples or spectators or hearers.

It is the genesis of intuition in in te l l igence. II man accedes

to the open creative totality, it is therefore by act ing, by cre­

ating rather than by contempla t ing . In philosophy itself, there

is still t o o much alleged con templa t ion : Everything happens


Page 112: Deleuze, Gilles. Bergsonism


as if intel l igence were already imbued with emot ion , thus with

in tui t ion, but not sufficiently so lor creat ing in conformi ty to

this e m o t i o n . ' 8 Thus the great souls — to a greater ex tent than

philosophers — are those ol artists and myst ics (at least those

of a Christian myst ic ism that Bergson descr ibes as being c o m ­

pletely superabundant activity, act ion, c rea t ion) . ' 1 ' At the l imit ,

i t i s the myst ic w h o plays wi th the whole of c rea t ion , w h o

invents an expression of i t whose adequacy increases with its

dynamism. Servant ol an open and finite God (such are the char­

acterist ics o f the Elan Vital), the mystical soul actively plays the

whole ol the universe, and reproduces the opening ol a W h o l e

in which there is noth ing to see or to c o n t e m p l a t e . Already

motivated by emot ion , the philosopher extracted the lines that

divided up the compos i tes given in exper ience . He prolonged

the ou t l ine to beyond the "turn"; he showed in the dis tance

the virtual point at which they all me t . Everything happens as

if that which remained indeterminate in philosophical intuition

gained a new kind ol de te rmina t ion in mystical intui t ion — as

though the properly philosophical "probabil i ty" extended i tself

into mystical certainty. Undoubtedly philosophy can only con­

sider the mystical soul from the outside and Irom the point of

view of its lines ol p robabi l i ty . 4 0 But it is precisely the exis­

t e n c e ol myst ic ism that gives a higher probabil i ty to this final

transmutation into certainty, and also gives, as it were, an enve­

lope or a l imit to all the aspects of me thod .

* * *

At the outse t we asked: W h a t is the relat ionship be tween the

three fundamental concepts of Duration, Memory and the Elan

Vital} Wha t progress do they indicate in Bergson's philosophy."

It seems to us that Duration essentially defines a virtual mul-

1 1 2

Page 113: Deleuze, Gilles. Bergsonism


t ip l ic i ty (what differs in nature). M e m o r y then appears as the

c o e x i s t e n c e of all the degrees of difference in this mult ipl ic i ty ,

in this virtualitv. T h e clan vital, finally, designates the actual­

ization ol this virtual according to the lines of differentiation that

cor respond to the degrees — up to this precise line ol man

where the F.Ian Vital gains self-consciousness.

Page 114: Deleuze, Gilles. Bergsonism
Page 115: Deleuze, Gilles. Bergsonism


A R e t u r n t o B e r g s o n

A " re turn to Bergson" does not only mean a renewed admira­

tion lor a great philosopher but a renewal or an extension of his

project today, in relation to the transformations of life and soci­

ety, in parallel with the transformations of science. Bergson him­

self considered that he had made metaphysics a rigorous disci­

pline, one capable of being cont inued along new paths which

constant ly appear in the world. It seems to us that the return

to Bergson, understood in this way, rests on three main features.


Bergson saw intuition not as an appeal to the ineffable, a par­

ticipation in a feeling or a lived identification, but as a true

method. Th i s method sets out, firstly, to de te rmine the condi­

tions of problems, that is to say, to expose false problems or

wrongly posed questions, and to discover the variables under

which a given problem must be stated as such. T h e means used

by intuition are, on the one hand, a cut t ing up or division of

reality in a given domain, according to lines of different natures

and, on the other hand, an intersection ol lines which are taken

Irom various domains and which converge. It is this complex

" 5

Page 116: Deleuze, Gilles. Bergsonism


linear operation, consist ing in a cut t ing up according to articu­

lations and an intersecting according to convergences , which

leads to the proper posing of a problem, in such a way that the

solution itself depends on it.

Science and Metaphysics

Bergson did not merely cr i t ic ize sc ience as if it went no further

than space, the solid, the immobi le . Rather, he thought that

the Absolute has two "halves ," to which science and metaphys­

ics correspond. Thought divides into two paths in a single impe­

tus, one toward matter, its bodies and movements, and the other

toward spirit, its qualities and changes. Thus , from antiquity,

just as physics related movement to privileged positions and

moments , metaphysics consti tuted transcendent eternal forms

from which these positions derive. But " m o d e r n " sc ience be­

gins, on the contrary, when movement is related to "any instant

whatever": it demands a new metaphysics which now only takes

into account immanent and constant ly varying durations. For

Bergson, duration b e c o m e s the metaphysical correlate of mod­

ern sc ience . He, o f course , wro te a b o o k , Duration and Simulta­

neity, in which he considered Einstein's Relativity. Th i s book

led to so much misunderstanding because it was thought that

Bergson was seeking to refute or cor rec t Einste in , while in fact

he wanted, by means ol the new feature ol duration, to give the

theory of Relativity the metaphysics it lacked. And in this mas­

terpiece, Matter and Memory, Bergson draws, from a scientific con­

cep t ion of the brain to which he h imse l f made impor t an t

contr ibut ions , the requirements of a new metaphysic of mem­

ory. For Bergson, sc ience is never " reduct ionis t" but, on the

contrary, demands a metaphysics — without which it would

remain abstract , deprived of meaning or intuition. To cont inue


Page 117: Deleuze, Gilles. Bergsonism


Bergson's project today, means for example to constitute a meta­

physical image of thought corresponding to the new lines, open­

ings, traces, leaps, dynamisms, discovered by a molecular biology

of the brain: new l inkings and re-Iinkings in thought .


From Time and Tree Will onward, Bergson defines duration as a

multiplicity, a type of"multiplicity. This is a strange word, since

it makes the multiple no longer an adjective but a genuine noun.

Thus , he exposes the traditional theme of the one and the mul­

tiple as a false problem. T h e origin of the word, Multiplicity or

Variety, is physico-mathematical (deriving from R i e m a n n ) . It

is difficult to bel ieve that Bergson was not aware of the scien­

tific origin of the term and the novelty of its metaphysical use.

Bergson moves toward a distinction be tween two major types

of multiplicit ies, the one discrete or discont inuous, the other

cont inuous, the one spatial and the other temporal , the one

actual, the other virtual. Th i s is a fundamental theme of the

encounter with Einste in . O n c e again, Bergson intends to give

multiplicities the metaphysics which their scientific t reatment

demands. Th i s is perhaps one of the least appreciated aspects

of his thought - the consti tut ion of a logic of multiplici t ies.

To rediscover Bergson is to follow or carry forward his approach

in these three direct ions. It should be noted that these three

themes are also to be found in phenomenology — intuition as

method, philosophy as rigorous sc ience and the new logic as

theory of multiplici t ies. It is true that these notions are under­

stood very differently in the t w o cases . T h e r e is nevertheless a

possible convergence as can be seen in psychiatry where berg­

sonism inspired the works of Minkowski (l.e temps i c ' c u ) and in


Page 118: Deleuze, Gilles. Bergsonism


phenomenology those o f Binswanger (Le casSusan Urban), in his

explorations of space- t imes in psychoses. Bergsonism makes

possible a whole pathology of duration. In an outstanding arti­

c le on "paramnesia" (false recogni t ion) , Bergson invokes meta­

physics to show how a memory is not const i tuted after present

perception, but is str ict ly contemporaneous with it, s ince at

each instant duration divides into two simultaneous tenden­

cies , one of which goes toward the future and the other falls

back into the past. I le also invokes psychology, in order to then

show how a failure of adaptation can make memory invest the

present as such. Scient i f ic hypothesis and metaphysical thesis

are constant ly combined in Bergson in the reconsti tut ion of

comple te exper ience .


Paris, July 1988

Translated by Hugh Tomlinson

1 1 8

Page 119: Deleuze, Gilles. Bergsonism

N o t e s


1. Bergson, Oxford: Oxford University Press, 1985, p. 2.

2. See Gilles Deleuze and Claire Parnet, Dialogues (translated by 1 lugh Tom­

linson and Barbara llabberjam). London: The Athlone Press, 1987, pp. 14-15.

5. "I.ettre a Michel Crcssole," in Michel Crcssole, Deleu/e. Paris: Editions

Universitaires, 1973, p. 111.

4. Dialogues, op. cit. , p. 15.

5. Ibid., pp. vii-viii.

(>. Gillian Rose, Dialectic ofSihilism, Oxford: Basil Black well, 1984, Chapter 6.

7. Gilles Deleuze, Cinema I: The Movement-Image (translated by Hugh

Tomlinson and Barbara llabberjam). London: The Athlone Press, 1986, Chap­

ters | and 4; and Gilles Deleuze, Cinema 2: The Time-Image (translated by

I lugh Tomlinson and Robert Galeta). London: The Athlone Press, 1988, Chap­

ters 3 and 5.

N. / ime ami hrec Will, Matter and Memory, Crcatiw Involution, and Mind-Energy.

For lull references, see p. II.

l». Critique o) Pure Reason, A 8 4 / B 1 I 6 ; see Gilles Deleuze, Kant's Critical

Philosophy (translated by Hugh Tomlinson and Barbara llabberjam). 1 ondon:

The Athlone Press. 1984. p. HIT.


Page 120: Deleuze, Gilles. Bergsonism



1. CM, 53 ( 1 2 7 1 . 2 5 ) .

2. Itttrt a Hoffding, 1916 (cf. tcrits el Paroles,\'o\. 3, p. 4 5 6 ) .

3. On the use ol the word intuition, and on the genesis ol the notion in Tl"

and MM the reader is referred to M. I lusson's book, l.'intcllectualisme tie Rergson,

Presses Universitaires de Trance. 1947. pp. 6-10.

4 . CM, 37-38 (1274-1275, 2 9 - 3 0 ) .

5 . CM, 5 8 - 5 9 ( 1 2 9 3 , 5 1 - 5 2 ) . On the "semi-divine s tate ," cf. CM, 75

(1306 , 6 8 ) .

6. According to Bergson. the category ol" problem has a greater biological

importance than the negative category ol need.

7. CM, 115 (15 36, 105). The arrangement ol examples varies in Bergson's

texts. This is not surprising, because each lalse problem, as \vc shall see,

presents the two aspects in variable proportions. On freedom and intensity

as false problems, c f . C M , 2 8 - 2 9 (1268 , 2 0 ) .

8. CM, 118 (1559, 110). On the critique ol disorder and ol nonbeing, cl.

a l soCE, 2 4 2 - 2 4 3 ( 6 8 3 , 223IT.)and 302 -303 ( 7 3 0 . 278IT.).

9 . CM, 5 9 - 6 0 ( 1 2 9 3 - 1 2 9 4 , 5 2 - 5 3 ) .

10. C f . T F . C h . I.

11. CM, 7 3 - 7 4 ( 1 3 0 4 - 1 3 0 5 , 6 6 ) .

12. Cl. a very important note in CM, 305 -304 (1 306 , 6 8 ) [same reference as

note 5 ] .

13. CE, 167 (62 5. 152).

14. Qualitative differences or the articulations ol the real are constant terms

and themes in Bergson's philosophy: c f , in particular, the Introduction to

CM, passim. It is in this sense that one can speak ol a I'latonism in Bergson

(cl. the method ol division). He loves to quote the text ol Plato on cutting

up and the good cook. Cf. C E , 172 (627 . 157) .

I 5. C F . 5 4 6 ( 7 6 4 . 518).

16. For example, intelligence and instinct form a composite which in its

pure state can only be dissociated into tendencies, cf. CE, 150-151 (610, I 57).


Page 121: Deleuze, Gilles. Bergsonism


17. On the opposition "in fact - in principle," cf. MM. Ch. 1 - notably 73

(213 . 6 8 ) . And on the "presence-representation" distinction, MM, 35

(185 , 32 ) .

18. MM, 4 8 (197, 4 7 ) .

19. MM, 36 (186, 33): "Now, if living beings are within the universe just

'centers of indetermination,' and if the degree of this indetermination is

measured by the number and rank of their function, we can conceive that

their mere presence is equivalent to the suppression of all those parts of

objects in which their functions find no interest."

20 . The line does not need to be entirely homogeneous, it can be a broken

line. Thus affectivity is qualitatively distinct from perception, but not in

the same way as memory: Whereas a pure memory is opposed to pure per­

ception, affectivity is more like an "impurity" which troubles perception:

cf. MM, 58 (207, 6 0 ) . W e will see later how affectivity, memory, etc. , denote

very diverse aspects of subjectivity.

2 1 . MM, 67 (214, 6 9 ) . Translation modified.

22 . MM. 1 8 4 ( 3 2 1 , 2 0 5 ) .

2 3. MM, 185 (321 , 2 0 6 ) . Bergson oltcn seems to criticize the

analysis: Although it reduces ad infinitum the intervals that it considers, it is

still content to recompose movement with covered space: lor example, TF

119-120 ( 7 9 - 8 0 , 8 9 ) . But more profoundly, Bergson requires that metaphys­

ics, lor its part, carry out a revolution which is analogous to that ol calculus

in science: cf. CE, 3 5 7 - 3 7 2 ( 7 7 3 - 7 8 6 , 3 2 9 - 3 4 4 ) . And metaphysics should

even draw inspiration Irom the "generative idea ol our mathematics," in

order to "carrv out qualitative ditlerentiations and integrations": CM, 216-217

(142 3 , 2 1 5 ) . [see also n. 2 4 ]

24 . Cf. CM, 216-217 (1416. 2 0 6 ) . And 2 2 8 (1425 . 218): "Philosophy should

be an effort l o g o beyond the human state." (The previously quoted text , on

tn« turning point of experience, is a commentary on this formula.)

2 5. CM, 1 5 7 - 1 5 9 ( 1 3 7 0 , 148-149) .

26. MR. 2 3 7 ( 1 1 8 6 . 2 6 3 ) .

1 2 1

Page 122: Deleuze, Gilles. Bergsonism


27. CM, 8 7 - 8 8 (1515, 8 0 ) .

28 . MR, 2 5 2 - 2 5 3 (1199-1200, 2 8 0 - 2 8 1 ) .

2 9 . ME, 6-7 (817-818, 4 ) , 35 ( 8 3 5 , 2 7 ) .

30. Cl". MM, 71 (218, 74): "Questions relating to subject and object, to their

distinction and their union, should be put in terms of time rather than space."

31. CM, 3 8 - 3 9 ( 1 2 7 5 , 3 0 ) .

32. C E , 13 ( 5 0 2 , 10). In this context, Bergson grants sugar duration only

insofar as it participates in the whole of the universe. The meaning of this

restriction will become clearer in Chapter 4.

33. CM. 217 (1416-1417, 2 0 6 - 2 0 8 ) .

34 . CM, 65-71 (129-130. 5 8 - 6 4 ) .

35. CE, 2 3 6 - 2 3 7 ( 6 7 9 . 217).Translation modified.

36 . MR. 202 (1156, 225).Translation modified.

37. Cf. CM, 4 2 - 4 3 (I278IT., 34IT.). And CM, 112 (1335 . 104): Intelligence

"touches one of the sides ol the absolute, as our consciousness touches


38. CM, 68 ( 1 3 0 0 , 61) .


1. See A. Robinet's excellent analysis on this point, in Rergson, Seghers,

1965, pp. 2811.

2. Admittedly, as early as lime and l-rce Will Bergson points out the prob­

lem of a genesis of the concept of space, starting from a perception ol exten­

sity, cf. 95-97 ( 6 4 - 6 5 , 7 1 - 7 2 ) .

3. TF, Ch. 2 and Ch. 3, 83-84 (107, 122).The badly analyzed composite or the

confusion ol the two multiplicities precisely defines lalse notions ol intensity.

4. On Riemann's theory ol multiplicities cl. G.B.R. Riemann, Oeuvres

Mathimatiquti (French Translation edited by Ciauthier-Viliars, "Sur les

hypotheses qui servent de Iondcmcnt a la geometric"): and II. Weyl, lemps,

Espace, Matierc. I lusserl too gained inspiration from Riemann's theory ol mul­

tiplicities, although in quite a different way from Bergson.


Page 123: Deleuze, Gilles. Bergsonism


5. "IT. 8 5 - 8 4 ( 5 7 . 6 2 ) .

6 . MM, 71-72 (218-219 , 7 5 - 7 6 ) .

7. CM, 1 5 7 ( 1 3 5 3 . 127) .

8. Cf. MM, 2 0 6 (341 , 231) . "As long as we arc dealing with space, we may

carry the division as long as we please; we change in no way the nature ol

what is divided."

9 . If. 81-82 ( 5 5 - 5 6 , 6 0 - 6 1 ) .

10. TF, 84 (57, 6 2 ) .

1 1. IT. 121 ( 8 1 , 9 0 ) .

12. The objective is, effectively, defined by the parts that are actually and

not virtually perceived: TF, 84-85 (57 , 6 3 ) . This implies that the subjec­

tive, on the other hand, is dcFined by the virtualitv of its parts. Let us return

then to the text: "We apply the term subjective to what seems to be com-

pletelv and adequately known, and the term objective to that which is known

in such a way that a constantly increasing number of new impressions could

be substituted for the idea which we actually have of it": TF 83 (57, 6 2 ) .

Taken literally, these definitions arc strange. By virtue of the context, one

might even wish to reverse them. For is it not the objective (matter) that,

being without virtualitv, has a being similar to its "appearing" and finds itself

therefore adequately known? And is it not the subjective that can always be

divided into two parts ol another nature, which it only contained virtually?

We might almost be inclined to think it a printing error. But the terms Bergson

uses arc justified from another point of view. In the case ol subjective dura­

tion, the divisions are onlv valid insofar as they are effectuated, that is, actu­

alized: "The parts of our duration are one with the successive moments ol

the act which divides i t . . .and if our consciousness can only distinguish in a

given interval a definite number of elementary acts, if it terminates the divi­

sion at a given point, there also terminates the divisibility": MM, 2 0 6 (341 ,

2 52 ). It can there-lore be said that, on each of its levels, the division ade­

quately gives us the indivisible nature of the thing while, in the case ol objec­

tive matter, the division does not even need to be effectuated: We know in

' 2 3

Page 124: Deleuze, Gilles. Bergsonism



advance that it is possible without any change in the nature ol the thing. In

this sense, if it is true that the object contains nothing other than what we

know, it nonetheless always contains more: MM, 147 ( 2 8 9 , 164); it is there-

lore not adequately known.

1 3. CM, 2 0 6 - 2 0 7 ( 1 4 0 8 , 1 9 6 - 1 9 7 ) .

14. The denunciation ol the Hegelian dialectic as lalse movement, abstract

movement, failure to comprehend real movement, is a frequent theme in

Kierkegaard, Marx, and Nietzsche, albeit in verv different contexts.

15. Cf. Plato, Philebus.

16. CM. 207-217 (1409-1416, 197-207) . This text is close to the passage in

I'lato where he condemns the pliancy ol the dialectic. W'e have seen that the

Bergsonian method ol division had a Platonic inspiration. The point ol con­

tact between Bergson and Plato is in fact the search lor a procedure capable

ol determining in each case the "measure," the "what" or the "how many."

It is true that Plato thought a relined dialectic could meet these require­

ments. Bergson, on the other hand, considers the dialectic in general, includ­

ing that ol Plato, to be valid only lor the beginnings ol philosophy (and ol

the history ol philosophy). The dialectic passes by a true method ol divi­

sion, it can do nothing other than carve out the real according to articula­

tions that are wholly formal or verbal. Cf. C M , 95 (1321 . 8 7 ) : "There is

nothing more natural than that philosophy should at first have been content

with this, and that it began by being pure dialectic. It had nothing else at its

disposal. A Plato, an Aristotle, adopt the cutting out of reality that they Iind

already made in language "

17. TF, 110(74 . 8 2 ) .

18. Cf. a very important text in CE, 32lff. (75711., 31011.): "But all move­

ment is articulated inwardly," e tc .

19. TF, 2 2 7 ( 1 4 8 , 170) and 2 0 9 - 2 1 9 ( 1 3 7 , 157).Translation modified.

ClI.MM I K i l l

I. MF.. 8 ( 8 1 8 . 5 ) : CM, 211 (1411, 201); MM, 34 (184, 31) The emphasis is

Page 125: Deleuze, Gilles. Bergsonism


ours in each ol these texts. These two forms ol memory should not be con­

fused with those discussed by Bergson at the beginning of Chapter 2 of MM, 78

( 2 2 5 , 8 3 ) ; this is a completely different principle ol distinction, cf. note 34.

2 . CM, 1 9 3 ( 1 3 9 8 , 183).

3 . Cf. ME. 1 3 - 1 4 ( 8 2 0 , 8 ) .

4. Cf. MM, 58 ( 2 0 6 , 5 9 ) .

5. MM, 7 7 ( 2 2 3 , 81).

6. CM, 87 (1315, 80).Translation modified.

7. MM, 148-149 ( 2 9 0 , 1 6 5 - 1 6 6 ) .

8. Nevertheless, on another occasion, Bergson maintained that there was

only a difference in degree between being and being useful: In fact, percep­

tion is only distinguished from its object because it retains solely that which

is useful to us (c l . MM, Ch. I) .There is more in the object than in percep­

tion, but there is nothing that is ol a different kind. But in this case, the

being is merely that of matter or of the perceived object, thus a present being

whose only distinction from the useful is one of degree.

9 . CM. 8 8 - 8 9 ( 1 3 1 6 , 81) .

10. Jean Hyppolitc gives us a profound analysis of this aspect. He attacks

"psychoiogistic" interpretations of Matter &. Memory: Cf. "Du bergsonisme a

I'existentialisme," Mercure dc France, July, 1949; and "Aspects divers de la

memoire che/ Bergson," Revue internationale de philosophic, October, 1949.

11. MM, 1 3 3 - 1 3 4 ( 2 7 6 - 2 7 7 , 1 4 8 ) .

12. The expression "at once" (d'emblec) is frequently used in Chapters 2 and

i of MM.

13. Cl. MM. 116 (261 , 129): "the hearer places himself at once in the midst

"I the corresponding ideas "

14. MM, 135 ( 2 7 8 , 150).Translation modified.

15. Cf. ME, 157-160 (913-914 . 130-131): "I hold that the formation of recol­

lection is never posterior to the formation of perception; it is contemporaneous with

it For suppose recollection is not created at the same moment as per­

ception: At what moment will it begin to exist? . . . T h e more we rellect, the

' 2 5

Page 126: Deleuze, Gilles. Bergsonism


more impossible it is to imagine any way in which the recollection can arise

if it is not created step by step with the perception itself. . . ."

16. A comparison could also be made here between Bergson and Proust.

Their conception of time is extremely different, but both acknowledge a

kind ol pure past, a being in itself of the past. According to Proust this being

in itself can be lived, experienced by virtue of a coincidence between two

instants ol time. But according to Bergson, pure recollection or pure past

are not a domain of the lived, even in paramnesia: we only experience a


17. The metaphor of the cone is first introduced in MM, 152 ( 2 9 3 , 169).

18. MM, 2 4 1 - 2 4 2 ( 3 7 1 , 2 7 2 ) .

19. On this metaphysical repetition cf. MM, 103-104 ( 2 5 0 , 115); 161-162

( 3 0 2 . 181).

20 . Cf. MM. 103-104 ( 2 4 9 - 2 5 0 , 114). Bergson shows clearly how we neces­

sarily believe that the past follows the present as soon as we establish only a

difference in degree between the two; cf. ME, 160-161 (914, 132): "The per­

ception being defined as a strong state and the recollection as a weak state,

the recollection ol a perception being necessarily then nothing else than

the same perception weakened, it seems to us that memory ought to have to

wait in order to register a perception in the unconscious. Indeed, it must

wait until the whole of it goes to sleep. And so we suppose the recollection

ol a perception cannot be created while the perception is being created nor

can it be developed at the same time." Translation modified.

2 1 . MM. 1 7 0 ( 3 0 9 - 3 1 0 , 1 9 0 ) .

22 . MM, 1 3 4 ( 2 7 7 , 1 4 8 ) .

23 . MM, I 3 0 ( 2 7 4 - 2 7 S , 145) .

24 . MM, 1 6 8 - 1 6 9 ( 3 0 7 - 3 0 8 , 188) (our emphasis).

25 . For example, in the passage that we have just quoted.

26 . In fact, the level must be actualized no less than the recollection that it

bears. CI. MM. 242 I 571, 21?): "'these planes, moreover, are not given as

readv-made things superposed the one on the other. Kathcr they exist virtu-


Page 127: Deleuze, Gilles. Bergsonism


ally, with that existence which is proper to things of the spirit. The intel­

lect, forever moving in the interval which separates them, unceasingly finds

them again or creates them anew "

27. MM, 1 6 8 ( 3 0 8 , 188): "without dividing.. . ."

2 8 . M E , 195-198 ( 9 3 6 - 9 3 8 , 161-163). Hence the metaphor of the pyramid

to represent the dynamic schema: "We will descend again from the summit

ol the pyramid toward the base " It is clear that the pyramid is very dif­

ferent Irom the cone and denotes a completely different movement, with a

different orientation. However, in another text (ME, 116 [ 8 8 6 , 9 5 ] ) , Bergson

evokes the pyramid as the synonym of the cone: the explanation lor this is in

the ambiguity pointed out above, note 2 5 .

2 9 . MM, 104 ( 2 4 9 - 2 5 0 , 1 1 4 - 1 1 5 ) .

30. On these two extremes, cf. MM, 153 ( 2 9 4 , 170) .

3 1 . MM, 120 ( 2 6 5 , 133) . Translation modified. And MM, 99 ( 2 4 5 , 108) :

"the last phase of the realization of a recollection - the phase of action "

Translation modified.

32 . Cf. MM, 9 2 - 9 3 ( 2 3 8 - 2 4 0 , 100-102); 98 ( 2 4 3 - 2 4 4 , 107); 112 ( 2 5 5 - 2 5 6 ,

121-122). Above all the motor scheme should not be confused with the dynamic

schema. Both intervene in actualization but at completely different phases:

I he former is purely sensory-motor, the latter psychological and mnemonic.

3 3. MM, 9 7 ( 2 4 1 , 104) .

34. Cf. MM, 108-109 ( 2 5 2 - 2 5 3 , 118-119).

35. MM, 98 ( 2 4 4 , 107) . There are therefore two forms of recognition, the

one automatic, the other attentive, to which correspond two forms of memory,

the one motor and "quasi instantaneous," the other representative and endur­

ing. We should, at all costs, avoid muddling this distinction, which is made

Irom the standpoint ol the actualization of recollection, with a completely

different distinction, made from the point of view of Memory in itself

(recollection-memory and memory-contraction).

36. On these two types of disturbance, cf. three essential texts, MM, 99

(24S, 108). 110(253, 118), 174(314, 196). In this last text Bergson distinguishes


Page 128: Deleuze, Gilles. Bergsonism



between mechanical and dynamic disturbances.

37. C f . MM, 108 ( 2 5 3 , 119): "The evocation of recollections themselves is

hindered" (translation modified); and also MM, 9 7 - 9 8 ( 2 1 5 . 108).

38. MM, 175 (314 . 196) .

5". ME, 1 7 7 - 1 8 3 ( 9 2 5 - 9 2 8 , 146-150) .

4 0 . ME, 1 3 0 ( 8 9 6 , 107).

C l I A P T I K IV

1. CI. above pp. 2 7 - 2 9 .

2. MM, 79 ( 2 2 5 . 8 5): "We pass, by imperceptible stages, from recollec­

tions strung out along the course of time to the movements which indicate

their nascent or possible action in space " MM, 122 ( 2 6 6 , 155): "We have-

here a continuous movement At no moment is it possible to say with

precision that the idea or the recollection-image ends, that the recollection-

image or the sensation begins." Translation modified. MM, 125-126 ( 2 7 0 ,

140): "To the degree that these recollections take the form of a more com­

plete, more concrete and more conscious representation, they tend to con­

found themselves with the perception which attracts them or of which they

adopt the outline."

3. MM, 151 ( 2 9 2 , 168) .

4. On going beyond the two dualisms: ( I ) quantity-quality. ( 2 ) extended-

nonextended, cl. MM, Chs. I and 4.

5. On the movement belonging to things as much as to the Self, cf. MM.

1 9 8 ( 3 3 1 . 2 1 9 ) ; 2 0 4 ( 3 4 0 . 2 3 0 ) .

6. Rcintroduction of the theme of degrees and intensities: CI. MM, Ch. 4, pas­

sim, and 222 ( 555 , 2 5 0 ) : "Between brute matter and the mind most capable

ol reflection there are all possible intensities ol memory or, what comes to

the same thing, all the degrees ol freedom." CT:. 219 ( 6 6 5 , 201): "Our li-cl-

ing of duration, I should say the actual coinciding ol our self with itsell.

admits of degrees." Ami already in IT, 239 -240 (156, 180): "It is because the

transition is made by imperceptible steps Irom concrete duration, whose ele-

Page 129: Deleuze, Gilles. Bergsonism


merits permeate one another, to symbolical duration whose moments are set

side by side, and consequently from free activity to conscious automatism."

7. Reintroduction of the theme of the negative, both as limitation and opposi­

tion: Cf. C E , 99-100 (571ff , 9 0 ( 1 ) , matter is both limitation of movement

and obstacle to movement, "it is a negation rather than a positive reality."

CE, 2 2 0 ( 6 6 6 , 2 0 2 ) : matter as "inversion," "interversion," "interruption "

These texts arc nevertheless related to those where Bergson challenges all

notion of the negative.

8. Cf. MM: on modifications and perturbations, 201 (337 , 2 2 6 ) ; on irre­

ducible rhythms, 2 0 5 - 2 0 6 ( 3 4 2 , 2 3 2 - 2 3 3 ) ; on the absolute character of dif­

ferences, 193-194 ( 3 3 1 - 3 3 2 , 219) .

9. CM, 217-219 (1461, 2 0 7 - 2 0 9 ) . The next two quotations come from the

same text , which is very important to Bergson's whole philosophy.

10. C f . C E , 1 8 4 ( 6 3 7 , 1 6 8 ) .

11 . CE, 13 ( 5 0 2 , 10): "What else can this mean than that the glass of water,

the sugar and the process of the sugar's melting in the water are abstractions,

and that the Whole within which they have been cut out by mv senses and

understanding progresses in the same manner as a consciousness?" On the

particular characteristic of the living being, and its resemblance to the Whole,

cf. CE, 18-19 (507, 15). But Matter and Memory had already invoked the Whole

as the condition under which we attribute a movement and a duration to

things: MM, 193 ( 3 2 9 , 216); 196 ( 3 3 2 , 2 2 0 ) .

12. DS, 4 5 - 4 6 ( 5 7 - 5 8 ) .

1 3. DS, 46 ( 5 8 - 5 9 ) . Bergson goes so far as to say that impersonal Time has

only one and the same "rhythm." Matter and Memory, on the contrary, affirms

the plurality of rhythms, the personal character of durations (cf. MM, 207

[3-42, 2 3 2 ] : "but neither is it that homogeneous and impersonal duration,

the same lor every thing and everyone.. ."). But there is no contradiction: In

DS the diversity of fluxes will replace that of rhythms, for reasons of termi­

nological precision; and impersonal Time, as we will see, is definitely not a

homogeneous impersonal duration.


Page 130: Deleuze, Gilles. Bergsonism


14. I)S. 52 ( 6 7 ) .

1 5. DS, 47 ( 5 9 ) : "We catch ourselves dividing and multiplying our conscious­

ness " Translation modilied. This reflexive aspect of duration brings it

particularly close to a cogito. On triplicity. cf. DS, 54 ( 7 0 ) : There are in lact

three essential lorms ol continuity: that ol our interior life; that ol voluntary

movement; and that of a movement in space.

16. DS, 52 ( 6 8 ) and 61 (81 (.Translations modified.

17. MM, 2 0 6 ( $41, 2 3 2 ) .

18. DS, 47 (59).Translation modified.

19. On this hypothesis of Relativity which defines the conditions ol a cru­

cial kind of experience: Cf. DS, 71 ( 9 7 ) , 7 7 - 7 8 (114), 101 (164 ) .

20 . DS, 72 ( 9 9 ) . Translation modified. It has often been said that Bergson's

reasoning involves a misunderstanding of Einstein. But Bergson's reasoning

itself has also often been misunderstood. Bergson does not confine himself to

saying: A time that is different from mine is not lived, either by me or by-

others, but involves an image that I give myselt ol others (and reciprocally).

For Bergson fully admits the legitimacy ol such an image in expressing the

various tensions and the relations between durations that he will constantly

recogni/e for his own part. What he criticizes Relativity lor is something com­

pletely dillerent: I he image that I make to myself ol others, or that Peter

makes to himsell ol Paul, is then an image that cannot be lived or thought as

livable without contradiction (by Peter, by Paul, or by Peter as he imagines

Paul). In Bergsonian terms, this is not an image, it is a "symbol." If we lorget

this point, all ol Bergson's reasoning loses its meaning. Hence, Bergson's

concern to recall, at the end of DS, 156 (2 34): "But these physicists are not

imagined as real or able to be so "

2 1 . DS, 76 -82(112-116) .

22 . DS, 8 5 - 8 6 ( 1 2 0 - 1 2 1 ) .

2 5. Bergson therefore distinguishes lour types oi simultaneity in an order ol

growing depth: (1) relativist simultaneity, between distant clocks, DS, 54

(71) and 8211. (1161)'.); ( 2 ) the two simultaneities in the instant, between

' 3 0

Page 131: Deleuze, Gilles. Bergsonism


event and nearby clock; ( 3) and also between this moment and a moment of

our duration, DS, 5 4 - 5 8 ( 7 0 - 7 5 ) ; (4 ) the simultaneity of lluxes, DS, 5 2 - 5 3

( 6 7 - 6 8 ) , 60-61 (81) . Mcrleau-I'onty clearly shows how the theme of simul­

taneity, according to Bergson, confirms a genuine philosophy of "coexistence"

(c l . In Praise of Philosophy, translated by John Wild and James M. Edie,

Evanston: Northwestern University Press, 196 3, pp. 1411.).

24 . DS, 1 34 (199) and 155IT. (23311.).

2 5 . Cf. DS, 134 (199) and 150 ( 2 2 5 ) , an attack on a "space which swallows

time." ol a "time which in turn absorbs space."

26 . Against the idea ofa space that is given to us ready made, cf. CE, 2 2 4 - 2 2 5

( 6 9 , 2 0 6 ) .

27. In this sense, matter and dreams have a natural affinity, both represent­

ing a state of expansion (detente), in us and outside us: CE, 220-221 (665-667,

2 0 2 - 2 0 3 ) .

28 . CE, 221-222 ( 6 6 6 - 6 6 7 , 2 0 3 - 2 0 4 ) and MM, Ch. 4, passim.

29. On space as scheme or schema, cf. MM, 2 0 6 (341. 232 ) ; 209-211 (344 -345 ,

2 3 5 - 2 3 6 ) ; C E , 2 2 1 ( 6 6 7 , 2 0 3 ) .

3 0 . C f . C E , C h . 3.


1. Cl. above pp. 7 5 - 7 6 .

2. This ontological "naturalism" appears clearly in MR: On naturing Nature

and natured Nature cf. 49 (1024, 56 ) . The apparently strange notion of "nature's

plan" appears in MR. 48 (1022 , 54 ) . Despite some of Bergson's expressions

("Nature intended," M R , 55 [1029, 6 3 ] ) , this notion should not be inter­

preted in too finalistic a sense: There are several plans and each, as we shall

see, corresponds to one of the degrees or levels of contraction that all coex­

ist in duration. Therefore, they are "planes" rather than "plans," they refer

to sections, to sections ol the cone rather than to a project or to an aim.

3. According to Bergson, the word "Whole" has a sense, but only on con­

dition that it does not designate anything actual. He constantly recalls that:

Page 132: Deleuze, Gilles. Bergsonism


Whole is not given. This means, not that the idea ol the whole is devoid of

sense, but that it designates a virtualitv, actual parts do not allow themselves

to be totalized.

4. Cf. CE, 99 (571 . 9 0 ) . And MR, 282 (1225 . 313): "the essence ofa vital

tendency is to develop Ian-wise, creating, b) the mere Tact ol its growth,

divergent directions, each ol which will receive a certain proportion of the

impetus." On the primacy, here, of an undivided Totality, of Unity or of

a Simplicity, cf. C E , 99-101 ( 5 7 1 - 5 7 2 , 9 0 - 9 1 ) ; 130-131 ( 5 9 5 . 119) "the

original identity."

5 . C E . 1 0 9 ( 5 7 8 , 9 9 ) .

6. In fact, the products of differentiation are never completely pure in

experience. Moreover, each line "balances" that which is exclusive in it:

for example, the line that ends in intelligence arouses in intelligent beings

an equivalent ol instinct, a "virtual instinct" represented by sforr telling:

cf. MR, 1 0 0 ( 1 0 6 8 . 114).

7. Bergson's great reproach to the philosophies of nature is that they only-

saw differences of degree on a single line in evolution and differentiation:

CE, 1 4 9 ( 6 0 9 , 1 3 6 ) .

8. Philosophically, one might find in a system like Leibniz's a si.nilar hesi­

tation between the two concepts ol the virtual and ol the possible.

9. Cf. CM, "The Possible and the Real."

10. C E , 7 2 - 7 8 ( 5 4 9 - 5 5 4 , 6 4 - 7 0 ) .

1 I. C E , 80 ( 5 5 5 , 7 2 ) : How could an external physical energy, light lor

example, have "converted an impression left by it into a machine capable of

using it"?

12. The idea ol diverging lines or ol ramified series was undoubtedly not

unknown to classifiers Irom the eighteenth century. But what matters to

Bergson is the lact that the divergences of directions can only be interpreted

from the perspective of the actualization of the virtual. In R. Ruyer, today,

we find requirements analogous to those of Bergson: the appeal to an

"inventive, mnemonic and trans-spatial potential," the refusal to interpret

' 3 2

Page 133: Deleuze, Gilles. Bergsonism


evolution in purely actual terms (cf. his Elements <le psycho-biologic. Presses

Universitaires de Prance).

I 3. When Bergson (CE, 184 [637 . 168]) says, "It seems as if life, as soon as

it has become bound up in a species, is cut off from the rest of its own work,

save at one or two points that are of vital concern to the species just arisen.

Is it not plain that life goes to work here exactlv like consciousness, exactly

like memory?" The reader must understand that these points correspond to the

outstanding points that became detached at each level of the cone. Each line

ol dillerentiation or actualization thus constitutes a "plane (plan) ol nature"

that takes up again in its own way a virtual section or level (cf. note 2, above).

14. On this negative vocabulary, cf. CE, Ch. 3.

1 5. This character of life, posing and solution of a problem, appears to Bergson

to be more important than the negative determination of need.

16. CE, 188 ( 6 4 0 , 172); M R , 116 (1082 , 132) " . . . a t each stopping-place a

combination, perfect of its kind."

17. CE, 145 ( 6 0 6 , 132).

18. On the opposition of life and form, CE, 141 (60311., I29IT.): "Like eddies

ol dust raised by the w ind as it passes, the living turn upon themselves, borne

up by the great blast of lite. They are therefore relatively stable and counter-

led immobility so well . . ." On the species as "stopping place" see MR, 198

(1153, 221) . This is the origin of the notion of enclosure, which will take on

such great importance in the study of human society. The point is that, from

a certain point of view, Man is no less turned in on himself, closed in on

himself, circular, than the other animal species: It might be said that he is

"closed." Cf. MR. 2 9 - 3 0 ( 1 0 0 6 , 34) ; 2 4 5 - 2 4 6 (1193. 2 7 3 ) .

19. CE, 4 3 - 4 6 ( 5 2 6 - 5 2 8 , 3 7 - 4 0 ) .

20 . Cf. DS, 137 (203IT.) on the example of the "curved plane" and of the

"three dimensional curve."

2 1 . DS. 63 ( 8 4 ) : There is "a certain hesitation or indetermination inherent

in a certain part of things" that becomes merged with "creative evolution."

2 2 . CE, 4 7 ( 5 2 9 , 4 1 ) .

' 3 3

Page 134: Deleuze, Gilles. Bergsonism


23 . CE, 62 (54111., 5511.) "How do wo assume that accidental causes, pre­

senting themselves in an accidental order, have several times ended in the

same result, the causes being infinitely numerous and the effect infinitely

complicated?" I.. Cuenot has set out all kinds of examples going in the direc­

tion of the Bergsonian theory, c f . Invention et finaliteen biologic.

2 4 . CE, 5 8 ( 5 3 8 , 51).

2 5 . Cf. C E , 198-199 ( 6 4 9 , 182); ME. 8 (8I8IT., 5ff.).

2 6 . MR, 2 0 0 ( 1 1 5 4 , 2 2 3 ) .

27. On the man who tricks nature, extending beyond the "plane" (plan) and

returning to a naturing Nature, cf. MR, 48 -57 (1022-1029, 5 5 - 6 4 ) . On man's

going beyond his own condition, MR, passim, and CM, 2 2 9 (1425-218) .

2 8 . CE, 2 8 8 ( 7 1 9 , 2 6 4 ) .

29 . ME, 1 8 - 2 0 ( 8 2 5 - 8 2 6 , 14-15).

30 . MR, 189-190 (1145. 211). On the story-telling function and the virtual

instinct, 99 (1067ff., II31T.) and 109-110 (1076, 124). On obligation and the

virtual instinct, 2 0 ( 9 9 8 , 2 3 ) .

3 1 . M R . 2 9 - 3 0 ( 1 0 0 6 , 3 4 ) .

32. M R , 8 3 - 8 4 ( 1 0 5 3 , 9 4 ) ; 198-199 (1153, 2 2 2 ) .

5 3. Bergson nevertheless suggests this explanation in certain texts, lor exam­

ple, MR, 200-201 (1155, 2 2 4 ) . But it only has a provisional value.

34. MR, 31 (1008 . 35) . The theory of the creative emotion is all the more

important as it gives aflcctivity a status that it lacked in the preceding works.

In Time anil tree Will, affectivity tended to become intermingled with dura­

tion in general. In Matter anil Mcmorv, on the contrary, it had a much more

precise role, but was impure and rather painlul. On the creative emotion

and its relations with intuition, the reader is relerred to the study ol M.

Ciouhier, in / 'histoirc ct sa philosophic (Vrin, pp. 7611.).

35 . MR, 2 4 3 (1191-1192, 2 7 0 ) and 30-32 ( 1 0 0 7 - 1 0 0 8 , 3 5 - 3 6 ) .

36. It will be noted that art, according to Bergson, also has two sources.

There is a story-telling art, sometimes collective, sometimes individual: MR,

IS4-I86 (1141-1142, 2 0 6 - 2 0 7 ) . And there is an emotive or creative art: MR.

' 3 4

Page 135: Deleuze, Gilles. Bergsonism


241 (1190, 2 6 8 ) . Perhaps all an presents these two aspects, but in variable

proportions. Bergson does not disguise the lact that the story-telling aspect

appears to him to be interior in art; the novel would above all be story-telling,

music on the contrary, emotion and creation.

37. Cf. M R , 24 3 (1192, 2 7 0 ) : " . . . c r e a t e creators."

38. MR, 5 5 - 5 6 ( 1 0 2 9 , 6 3 ) .

39. On the three mysticisms, Greek, Oriental and Christian, cf. MR. 2 0 5 - 2 0 6

(115811.. 229IT.).

40 . CI. MR. 2 34 (1184, 2 6 0 ) . l e t us remember that the notion ol proba­

bility has the greatest importance in the Bergsonian method, and that intu­

ition is no less a method of exteriority than of inferiority.


Page 136: Deleuze, Gilles. Bergsonism
Page 137: Deleuze, Gilles. Bergsonism

I n d e x

ABSENCE, 17 -18 . Absolute, the, 35, 49, 76 , 84. Abstraction, 25, 4 4 , 4 6 , 53, 75, 96 ,

98 , 99 . Achilles'race, 4 7 - 4 8 , 8 1 . Action, 14, 19, 24, 55, 56, 68; order

of, 33; possible, 53; psychological, 56.

Actual, the, 15, 85, 9 3 , 9 6 , 97, 98 , 101 -03 ,104 , 106. Seeobo Real.

Actualization, 14, 4 2 - 4 3 , 52 , 53, 5 6 - 5 7 , 5 8 , 6 2 - 7 1 , 73, 8 2 , 9 4 - 9 5 , 97, 98 , 1 0 3 - 0 7 , 1 0 9 , 111, 113;of past, 5 6 - 5 7 ; psychic, 42 .

Affection, 23, 53; -subjectivity, 53. Affectivity, 25 , 56. Affinity, natural, 27, 33. Alteration, 31, 32 ,47 , 92. Analysis, transcendental, 23. Animal, 94 , 95 , 101, 109, 111. Aphasia, 30, 69. Articulations. 27, 28, 68; natural, 18,

22, 27, 29, 31. See also Real, the, articulation of.

Augmentation, 31. Automaton, 67.

BECOMING , 37, 44 , 45; -conscious, 16.

Being, 17-18 , 19, 20, 35, 44 . 4 6 - 4 7 .

5 5 - 5 6 , 61, 6 2 - 6 3 , 7 6 - 7 7 , 84, 85; diminution of, 23; paradox of, 61; -present, 55; pure, 59.

Berkeley. George, 41. Biology, 94, 95, 97; taxonomy, 103 - 0 4 ,

10S. Body, 26, 30, 41, 69 , 7 0 - 7 1 , 1 0 3 , 109. Brain, 24, 5 2 - 5 5 , 69 ,107; faculty of,

and core function, 24—25; -subjectivity, 52.

CALCULUS , 27. Coalescence, 65 , 66 . Coexistence, 59 - 60, 74, 77. 8 0 - 81,

86, 91, 9 3 , 1 0 0 - 0 1 , 1 0 3 , 111; paradox of, 61; virtual, 60.77, 85, 9 3 - 9 4 .

Composite, 1 8 - 1 9 , 2 2 - 2 3 , 26, 28, 2 9 - 3 0 , 32, 34, 37, 38,47, 5 3 , 7 3 , 8 5 , 8 6 , 9 2 , 9 S - 9 6 , 1 1 2 ; badly analyzed, 17, 18, 20, 22, 28, 54, 58, 6 1 - 6 2 , 7 3 , 86.

Concept, 2 8 , 4 4 - 4 5 , 7 5 , 9 7 . Cone, metaphor of, 5 9 - 6 0 (fig.),

6 4 - 6 5 , 6 6 , 6 7 , 8 8 , 100.

Consciousness, 30, 42 , 45 , 48 , 51, 52, 56, 78, 81, 82, 84 ,106; planes of, 65 , 66; psychological, 63; self-, 52, 106, 113.

Contemporaneity, 58, 59, 71. Continuity, 21, 37, 38, 43 , 52, 57, 87.


Page 138: Deleuze, Gilles. Bergsonism


Contraction. 21, 5 1 - 5 2 , 5 5, 60 , 61, 6 4 - 6 7 , 70, 74, 75, 76, 79 , 8 2 , 8 6 - 8 9 , 93 . 102, 105. 107; degrees/ levels of, 6 0 , 7 4 , 7 5 , 8 5 , 9 3 , 1 0 0 , 1 0 1 ; -memory, 26, 52, 60, 74; ontologi­cal and psychological, 65; -relaxa­tion, 75; -subjectivity, 53.

Convergence, 29, 30, 73. See also Intersection.

Creation, 9 7 , 9 8 , 1 0 1 . 1 0 3 , 105, 106. 108,110-12.

Creative Evolution, 37, 77, 78 , 100.

DATUM, IMMEDIATE, 3 8 , 7 5 . Decomposition, 38, 53, 67, 68 , 9 2 , 96 . Deterioration, 22, 23, 46 , 47, 7 5 - 7 6 ,


Determinism, 107. Difference, 35, 7 5 - 7 6 , 9 2 , 9 3 , 9 7 , 9 8 ,

100; in degree, 20, 21, 2 2 - 2 3 , 25, 3 1 - 3 2 , 34, 3 5 , 3 8 , 4 1 , 4 3 , 4 5 . 4 7 , 58, 73, 74, 75 , 7 6 , 9 1 , 9 2 - 93 , 94 , 96 , 101; of intensity. 2 0 - 2 1 , 75 , 91, 92 , 94; internal, 99; in kind, 14, 1 8 - 2 5 , 2 7 - 3 5 , 3 8 , 4 1 , 4 2 - 4 3 , 46, 47, 54, 55, 5 8 , 6 1 , 73 , 7 5 - 7 6 , 81. 82. 91, 9 2 . 93 , 95 , 96; in number, 35, 41; qualitative, 31.

Differentiation, 2 9 , 3 5 , 4 3 , 9 4 , 9 5 , 9 7 , 101-03 (fig. p. 102), 104. 106, 107. 108,110,113.

Dilation, set Expansion, Relaxation. Dimension, 23. Diminution, 2 3, 31. Discontinuity, 21, 37, 51. Disorder, 17 -18 , 1 9 - 2 0 , 4 6 - 4 7 . Distinction, 77; extrinsic, 3 7 - 3 8 ; of

quality, 35; ol quantity, 84; real, 85, 86. Disturbance, 6 8 - 7 0 , 71, 76; dynamic,

69; mechanical, 68 . Divergence, 2 8 - 3 0 , 4 3 , 5 3 , 7 3 , 9 5 , 9 7 ,

1 0 0 , 1 0 5 - 0 6 . Division, 22. 24, 31, 32, 40 , 4 1 - 4 2 . 47,

6 6 , 7 9 , 8 0 - 8 1 , 9 2 - 9 6 , 9 9 , 1 0 3 , 104, 112; two types of, 9 5 - 9 6 .

Doubt, 19. Dreams, 66 , 107. Dualism, 2 1 - 2 2 . 29. 31, 73 , 75, 76,

91, 93, 94 , 96; genetic. 96; reflexive, 9 6 .

Dualities, 74, 93 . Duration, 13 -14 , 19, 21, 22, 26, 28 .

3 1 - 3 5 , 3 7 - 3 8 . 4 0 . 4 2 , 4 5 - 4 6 , 4 8 - 4 9 , 5 1 - 5 2 . 54, 60 . 75 -113 ; -contraction, 2 3, 107; dispersed, 77; external, 48; in general, 45; intense, 77; internal. 81, 8 5 - 8 4 , 107; multiple, 4 8 - 4 9 , 75 , 7 6 - 7 7 , 78, 8 3 - 8 5 ; ontological, 49; psycho­logical. 34. 37. 4 8 - 4 9 , 76. 77;'pure, 95; simultaneous, 48; single, 78.

Duration and Simultaneity, 39, 78, 85.

EGOISM. 109,110. Einstein, Albert. 40 , 7 9 - 8 0 , 83,

8 4 , 8 5 .

llan vital, 13, 14, 16, 94 , 95 , 101. 104, 106, 107, 112-13.

Emotion. 1 8 , 4 2 , 1 1 0 - 1 2 . Energy, 76 , 102.

Essai sur les elements phncipaux de la representation, 44 .

Essence, 32, 3 4 , 9 4 , 110. Eternity, 23, 55 , 56, 104. Evolution, 98 , 100, 106. Evolutionism, 23, 9 8 - 9 9 . Excitation, 24, 52. 107, 111. Existence, 20, 77. Expansion, 30. 60 , 6 6 - 6 7 , 70 , 74,

7 5 , 7 9 , 8 6 - 8 9 . 9 1 . 9 3 . 9 5 . 1 0 0 , 102, 103, 107. See also Relaxation.

Expend it t i r e , 102. Experience. 13, 22. 2 5 - 2 7 , 50, 34,

37, 5 3 , 7 4 , 8 1 - 8 2 . 9 2 , 9 9 ; conditions of, 20. 23, 25, 26, 27, 28, 30, 37,99; physical, 4 7 - 4 8 ; possible, 23; psychological, 5 3, 34, 37. 38; pure. 92; real, 23. 27. 28; turn in. 2 7 . 2 8 . 7 3 , 9 1 , 9 2 , 9 3 , 9 5 .

l 3 8

Page 139: Deleuze, Gilles. Bergsonism


Extension, 22, 34 , 3 5 , 42 , 74 - 7 5 , 7 9 . 8 6 , 8 7 - 8 8 , 8 9 , 9 4 .

Exteriority, 4 9 , 7 4 - 7 5 , 7 7 - 7 8 , 9 3 , 99 .

1 0 2 , 1 0 3 .

FALSITY , 1 6 , 9 8 .

Pictions. 2 5 . 3 4 . 8 5 . 9 8 . 1 0 8 .

Final ism, 1 0 4 . 1 0 5 .

Flux, 8 0 - 8 5 , 9 1 ; triple, 8 0 - 8 1 . 8 2 . Form. 8 8 , 1 0 3 - 0 4 ; variety of, 4 5 . Fourth dimension. 7 9 , 86 . 1 0 4 - 0 5 . Freedom, 1 5 . 1 6 , 1 7 , 1 9 , 5 1 , 1 0 6 . 107. Freud, Sigmund, 55 — 5 6 . Future, the, 5 2 .


God, 1 0 4 , 1 0 8 . 1 1 0 . 1 1 2 .

HAMELIN, 4 4 . Ilegelianism, 44 . Heterogeneity. 2 1 . 37. 4 3 , 7 4 . 1 0 0 . History, 1 6 . Hoffdi'ng, 13.

Homogeneity, 2 0 , 21, 3 3 , 3 7 , 7 4 . 100 .

IDEALISM, 3 3 , 7 7 .

Illusions, 2 0 , 2 1 , 2 3 , 3 3 - 3 S . 58 . 6 1 , 104.

Image, 1 7 - 1 8 , 2 4 , 4 1 , 5 7 - 5 8 . 6 5 - 6 8 .

7 0 . 7 1 , 81. 9 7 - 9 8 ; recall of. 6 3 ;

virtual, 2 8 , 2 9 . Inadequacy, 44 , 4 6 , 7 5 . Indivisibles, 4 2 . Set aim Division. Inextensity, 2 3 . Instant, 2 5 , 5 1 - 5 2 , 5 3 , 7 4 , 84 , 87, 9 5 .

Instinct, 2 1 , 9 4 , 9 5 , 1 0 1 . 1 0 2 . 1 0 3 , 1 0 8 , 1 1 0 .

Intelligence, 21, 8 8 - 8 9 , 9 4 , 9 5 , 1 0 2 .

104 . 1 0 7 - 1 0 , 1 1 2 ; order of, 3 3 .

Intensity, 1 7 . 1 8 - 1 9 , 7 5 - 7 6 . 9 1 - 9 2 .

Intersection, 2 8 , 2 9 , 30 . 3 5 . 5 3 - S 4 .

Set also Convergence. Interval. 2 4 , 4 6 ; cerebral, 2 4 , 2 5 .

5 2 - 5 3 . 107 . 109 . 1 1 1 .

Intuition. 2 1 , 2 7 , 3 1 - 3 2 , 3 5 . 8 8 . 1 0 2 .

1 0 9 - 1 0 , 1 1 1 - 1 2 ; as method, 1 3 - 1 4 ,

2 2 - 2 4 . 32, J3, 3 8 . 7 3 . 7 7 . Invention. 15 -16 . 35, 108. 111.

KANT, IMMANLH I . 2 0 - 2 1 , 46 . Knowledge. 13, 17. 35.

LANGUACI. 15. 57, 62 . 68; foreign. 62; ontologv ol, 57.

Leap, ontological. 56. 57. 61. 62, 109; paradox ol. 61.

Life. 16. 19, 5 2 . 9 4 - 9 5 . 1 0 1 . 1 0 2 , 1 0 3 - 0 4 , 106. 107; attention to, 68 , 70 , 72, 80 .

MACHINE , 107. Man, 106, 109. 113. Marx. Karl. 16.

Mathematics. 1 5 - 1 6 , 27, 4 1 - 4 2 . Stt also Riemann.

Matter, 21, 22, 2 4 - 2 7 , 30, 34, 35, 41, 43 , 53, 5 4 - 5 5 , 6 0 - 6 1 , 73 , 74, 75 , 77. 78 , 82 . 86, 8 7 - 8 9 , 92, 9 3 . 9 4 , 101-03 . 1 0 7 - 0 8 ; contraction of, 2 5 - 2 6 ; -expansion, 23; order ol, 88.

Matter and Memory, 2 3 - 2 4 , 29, 40 , 41, 5 2 - 5 3 , 72 , 73 , 75 . 76 . 78 . 86, 9 2 , 9 6 , 1 0 0 .

Meaning, 88. Mechanism, 19. 23. 6 7 - 6 8 . 69 . 70,

9 8 . 1 0 4 , 1 0 5 . 1 0 7 . Memory, 13. 21. 22 , 2 5 - 2 6 . 30, 37,

4 3 , 5 1 - 5 2 , 5 3 , 5 5 , 5 7 , 6 3 - 6 4 , 6 5 . 6 7 , 7 0 . 7 3 . 7 7 , 9 2 , 9 3 , 1 0 0 , 102, 106-07 . 109, HI. 112-13; Bergson's theory ol, 43 , 5 5 - 5 6 ; ontological, 57, 59; paradox of, 5 8 - 5 9 ; psychophysiological theories of.' 58. 61; pure. 27. 58, 74, 95; two aspects of, 5 1 - 5 2 , 5 3 - 5 4 .

mens momtntanta, 75. Metaphysics. 15. 20. 2 3. 29. 35. Method. I 311.. 32. 38. 75 . 9 1 - 9 2 . 9 3 .

96 , 112; dialectical. 44-45. Stt also Intuition.

Mind. 26. $7. 62, 88. 109.

' 3 9

Page 140: Deleuze, Gilles. Bergsonism


Minil-I nenjK 30. Modification, 7 6 . Monism, 2 9 . 7 3 , 7 4 . 7 5 . 7 6 , 7 8 , 8 2 ,

9 1 , 9 3 , 9 4 .

Motive, psychological, 1 7 - 1 8 . Movement'. 2 4 , 2 7 , 3 1 , 4 5, 4 7 . 4 8 - 4 9 ,

5 2 , 5 4 . 6 5 , 6 7 - 6 9 , 7 0 , 7 1 , 7 4 , 7 5 ,

7 9 , 8 2 , 8 4 , 9 4 - 9 5 . 1 0 3 - 0 5 , 1 0 6 ,

1 0 7 ; executed, 2 4 , 5 2 ; false, 4 4 ; mechanical, 7 0 — 7 1 ; —perception. 6 7 - 6 9 ; received, 2 4 , 5 2 , 74, 87, 9 2 .

Multiple, the, 3 9 , 4 3 - 4 4 . 4 5 - 4 6 ,

47 , 7 6 , 8 0 . 8 5 . 9 3 : unity of, 4 4 , 4 5 ,

9 5 - 9 4 .

Multiplicity, 1 4 , 3 2 . 3811 . , 47 , 4 9 , 7 8 ,

7 9 - 8 0 , 8 5 ; abstract. 4 5 : actual/ spatial. 8 5 ; continuous qualitative, 38 , 3 9 - 4 0 , 4 2 - 4 3 , 4 7 , 8 0 , 8 1 ;

discontinuous/quantitative, 38, 4 0 , 4 1 , 4 3 . 47 . 8 0 ; discrete. 39 ; two kinds of, 1 9 , 2 1 , 3811 . , 47 , 5 3 ,

7 9 - 8 0 , 8 5 ; virtual/temporal.

8 2 - 8 3 , 8 5 , 1 1 2 - 1 3 .

Mysticism. 1 1 2 .

NATURE. 1 9 , 34 . 8 0 . 9 3 , 1 0 7 , 1 0 8 - 0 9 .

1 1 0 . 1 1 3 .

nature naturanut, 9 3 . nature naturec, 9 3 . Need. 6 2 . 6 8 . 1 0 8 ; order of, 3 3 ;

-subjectivity, 5 2 . Negation, 18 , 1 9 , 4 6 . 5 2 , 7 5 - 7 6 :

generalized. 1 7 . 46 . Negative, the, 18 , 46 , 7 5 - 7 6 , 1 0 1 - 0 3 ;

ol limitation and of opposition, 4 6 - 4 7 .

Nonbcing. 1 7 - 1 8 . 1 9 - 2 0 . 44 . 4 6 - 4 7 .

Nothingness, 2 0 , 2 3. Novelty. 2 0 , 6 1 .

Number, potential, 4 0 - 4 1 , 4 2 - 4 3, 4 5 .

OBJECT, 2 4 - 2 5 , 3 3 , 4 0 - 4 1 , 4 7 , 5 2 ,

5 3 . 6 8 , 7 5 , 7 5 , 7 7 , 7 8 , 1 1 0 ; image

of, 4 1 .

()bjectivitv. 50, 5 5 . 4 0 - 4 1 . 4 5 , 5 3 , 54 .

Obligation. 108 . (observation. 30 . One. the. 3 9 , 4 3 - 4 4 . 4 5 - 4 6 . 4 7 ,

80 , 8 5 , 9 3 , 100 .

Ontology, 3 4 - 5 5 . 49 , 5 6 , 7 6 .

Opposition. 4 4 - 4 5 , 4 6 . 7 5 - 7 6 , 8 2 ,

9 6 , 1 0 1 .

Order, 1 7 - 1 8 , 4 6 - 4 7 .

Organism, 1 6 , 1 0 5 .


Participation, 7 7 - 7 8 , 8 8 .

Past, the, 2 5 , 3 7 , 54IT., 7 0 , 7 1 , 7 3 , 7 4 .

7 5 . 9 1 , 9 2 ; in general, 5 6 - 5 7 , 5 9 ,

6 1 , 6 2 - 6 4 ; image ol, 5 1 ; degrees/ levels of, 5 9 - 6 7 , 7 4 . 7 7 ; preser­vation of, 2 5 , 5 1 , 5 4 - 5 5 , 5 9 ; pure, 5 9 , 7 4 , 7 5 , 9 5 , 1 0 9 ; regions of,

5 6 - 5 7 , 58 . 6 1 - 6 6 ; totality of, 2 7 ,

6 1 , 6 2 .

Pedagogy, 1 5 . Perception. 2 1 . 2 3 - 2 5 , 2 6 - 2 7 , 3 0 ,

5 1 , 5 3 , 58 , 6 3 . 6 7 - 6 8 , 7 3 , 7 4 , 7 5 .

1 0 7 ; actual, 4 1 , 6 7 ; -image, 58 , 66 , 6 7 - 6 8 , 7 1 , 7 3 , 74, 9 5 ; - ob jec t -

matter, 2 6 ; Pure. 2 6 . 2 7 , 2 8 , 54 . 5 5 , 58 ; Real, 2 5 ; -recollection, 2 2 , 2 9 ; virtual, 2 5 . 4 1 .

Perfection, 2 3 , 1 0 3 . Philosophy. 1 3 . 1 4 . 2 7 - 2 8 , 4 4 - 4 5 .

4 6 , 7 5 , 9 4 , 9 9 , 1 1 1 - 1 3 .

Physics, 3 5 . Scealw Einstein. Plant. 9 4 , 9 5 . 1 0 1 . 1 0 2 , 1 1 0 . Plato, 3 2 , 4 4 - 4 5 . 5 9 .

Pluralism, 7 6 , 7 7 - 7 8 , 8 3 - 8 4 , 1 0 4 ;

generali/ed, 7 7 - 7 8 , 8 2 ; limited, 7 7 - 7 8 , 8 2 : quantitative, 7 6 .

Plurality, 1 4 , 24 . Point. 7 9 ; mathematical, 2 5 , 5 3; of

unity, 7 3 - 7 4 , 9 5 ; virtual, 2 8 , 2 9 , 3 0 . 1 1 2 .

Position, 2 3 . Possibility, 1 8 . 1 9 , 4 5 . 9 6 . Possible, the, 1 7 - 1 8 . 2 0 , 2 4 . 4 1 , 47 ,

9 6 - 9 8 ; l.cibni/ian. 7 1 .

I 4 0

Page 141: Deleuze, Gilles. Bergsonism


Precision, I 3, 14, 29, 40 , 94 .

Prelormism, 98 .

Presence, 2 2 - 2 3 , 26.

Present, the, 25, 48 , 5 1 - 5 2 , 54IT., 68, 7 0 - 7 5 , 88 , 91 ,92; pure, 74, 95.

Probabilism, 30.

Problems. ISff., 21. 29; badly stated, 17, 18 -19 , 21; creation/statement of, 14, 15-17, 21 ,31, 35; false, 15-17, 18, 2 0 . 2 3 - 2 4 , 2 6 , 3 3 - 3 5 , 4 3 , 54 , 75 , 98, 104; nonexistent, 17, 1 9 - 2 0 ; true, 15-17 , 33.

Proportion, 23, 31.

Proust, Marcel, 85, 96 .

Psychology, 26, 57, 76. Pure, the, 22, 49 , 52.

QUALITY , 21, 31, 3 2 , 4 8 , 51, 53, 74,

87—88, 92; heterogeneous, 74.

Quantity, 21, 74, 91; homogeneous, 74.

Questions, badly stated, 17, 24.

REAL, THE , 17; 21, 29 , 3 0 , 4 1 , 44 , 47,

9 6 - 9 8 , 108; articulation of, 26 ,

92; disarticulation of, 30. Realism. 3 3,77.

Reality, 22 , 34, 42 , 45 , 97, 100; nonpsychological, 56; psycho­logical, 34, 58.

Realization, 20, 41, 43 , 71, 9 6 - 9 7 .

Reason, 20, 108; sufficient, 2 8 - 2 9 , 86.

Recognition, 67, 68 , 69 . Recollection, 2 1 - 2 7 , 30, 37, 51,

5 3 , 5 4 , 5 6 - 5 7 , 5 8 , 6 1 - 7 3 , 1 0 7 ; -image, 58, 63, 6 5 - 6 8 , 70, 71.

73 . 74, 95 , 109; -memory, 26 , 52. 60, 74; -perception, 22; pure,

26, 55, 56, 62 , 63 , 66 , 68, 69 , 70, 71, 74; -subjectivity, 5 3; virtual,

56, 63, 71. Recomposition, 45 , 57. Relativity, theory of, 39, 79, 8 3 - 8 4 ,

86 . Relaxation, 21, 23, 60. 61, 65 , 74, 75 ,

7 6 , 7 9 , 8 2 , 8 7 - 8 9 , 9 5 : levels,

degrees of, 60 , 7 4 , 7 5 , 8 5 , 86 , 8 8 - 8 9 , 9 1 , 100 . Sec also Expansion.

Religion, 34, 108 . Reminiscence, 6 2 ; 1 1 1 ; Platos

theory of, 5 9 . Repetition. 5 1 . 6 0 - 6 1 , 6 8 , 9 3 ;

physical and psychic, 6 0 — 6 1 , paradox of. 6 1 ; virtual, 6 1 .

Representation, 2 2 , 2 4 , 5 3 , 66 , 87, 1 0 8 , IK).

Repression, 2 1 , 7 2 . Resemblance, 1 0 1 , 1 0 5 - 0 6 .

Response, 2 4 , 107 , III.

Rest, 7 9 . Riemann, G.B.R., 3 9 - 4 0 , 7 9 . Rotation, 6 4 , 6 5 - 6 6 , 6 8 , 6 9 , 7 0 ;

-orientation, 64 . Rules, 1 5 . 1 7 , 2 1 , 2 9 , 3 1 .

SCHEME, DYNAMIC, 6 6 , 6 9 ; motor, 6 7 - 6 8 , 69 , 7 0 .

Science, 1 4 , 2 0 , 2 3 , 3 5 , 4 0 . 86 . Section, discontinuous, 3 7 ; instanta­

neous, 5 4 . Self, 4 4 , 7 5 , 1 0 6 .

Sensations, 1 8 - 1 9 , 5 3 , 7 4 , 7 5 , 87.

Sense, 88 . Simultaneity. 48 , 7 9 , 8 0 - 8 1 , 8 4 - 8 5 ;

ol fluxes, 8 1 , 89 . Simplicity, 4 3 , 4 6 , 9 4 , 9 5 , 9 6 , 1 0 0 . Sleep, 6 6 - 6 7 .

Society, 1 5 , 1 0 8 - 1 1 .

Solution. 1 5 - 1 7 , 2 1 , 2 9 - 3 0 , 1 0 3 ;

false, 2 0 . Soul, 1 1 2 ; immortality ol, 30 . Sources, two kinds of, 2 1 . Space, 1 9 , 2 1 , 2 2 , 2 5 , 3 1 - 3 8 , 4 3 , 4 7 ,

4 9 , 6 0 , 7 5 , 7 9 , 8 6 - 8 8 , 9 2 , 1 0 4 - 0 5 ;

auxiliary-, 38 ; homogeneous, 34 ; order of. 34 ; pure, 88 , 8 9 ; real, 1 0 5 : scientific conception ol, 40; - t ime , 7 9 , 8 5 . 9 5 , 1 0 4 - 0 5 .

Spirit, 3 0 , 3 5 , 9 3 . Story-telling function. 1 0 8 - 1 1 1 .

I 4 l

Page 142: Deleuze, Gilles. Bergsonism


Subject. 42 . 4 8 . 7 J . 8 1 - 8 2 , 8 ) . Subjectivity. 26 . JO. 33. 40 . 4 2 - 4 3 .

5 3; live aspects oi, 52 — 53. Succession. 25 . 45 . 48 . 59, 60 . 81;

internal, 37. Sugar, lump ol. 3 1 - 3 2 , 77. Systems, closed, 18, 43, 77; lixed and

mobile, 79 .

TENDENCY , 21, 2 2 - 2 3 , 28 . 31. 32, 9 2 , 9 3 , 9 9 ; motor, 67, 68 .

Tension, 60 , 74, 76 . 79 . 86, 87. 88, 95: levels of, 77.

Time. 22. 31, 32. 43 . 5 8 - 5 9 . 6 1 - 6 2 , 7 4 , 7 5 . 7 8 , 7911., 85 - 86, 93 , 1 0 4 - 0 5 ; multiplicity of, 76 . 78, 79, 8 3 - 8 4 , 85; relativistic. 7 9 - 8 0 , 8 3 - 8 4 ; real, 14, 78, 79; single, 80 . 8 1 , 8 2 - 8 3 . 8 5 , 9 1 , 9 3 , 1 0 0 ; spatiali/ed. 22, 23, 80, 85 . 86 . 104. Sec also Instant.

hint and Iree Will, 37. 39. 40 , 43 , 48 , 5 3 . 6 0 , 76 , 7 8 , 7 9 , 9 1 , 9 6 .

Totality, 32. 105; virtual. 93 . 95 . KM). Translation. 6 3 - 6 4 . 6 5 - 6 6 , 6 8 , 6 9 , 70,

97; -contraction, 64 , 70 . Truth, 16. 18. 29, 34.

UNCONSCIOUS , 42 , 5 5 - 5 6 , 7 1 - 7 2 ; Freudian, 5 5 - 5 6 : ontological, 71; psychological, 5 5 - 5 6 , 71; virtual. 55.

Unity, ontological. 74. 9 3. 9 5 . 100. Universe-. 78; Whole of. 77. 78. 82,

100. 103, 104. 105, 112. Utility, 27, 55. 64 , 67. 68 . 7 0 - 7 1 . 88 .

99'. 106. 107. 109.

V IBRATION, sec Movement, received. Virtual, the. 15, 4 2 - 4 3 , 56, 57, 60 ,

63 , 81, 82, 8 5 , 9 4 . 9 5 . 9 6 . 9 7 - 9 8 . 100, 103. 104, 105. 106; pure, 62.

Virtuality, 41, 8 2 - 8 3 , 93 , 95 , 97,

9 9 - 1 0 0 , 101, 103, 106,113.

W i n . 19.


Page 143: Deleuze, Gilles. Bergsonism
Page 144: Deleuze, Gilles. Bergsonism

This edition designed by Bruce Mau

Type composed by Archie at Canadian Composition

Printed by Areata Graphics llalliday

on Sebago acid-free paper

Page 145: Deleuze, Gilles. Bergsonism