Daniela Voss - Conditions of Thought - Deleuze and Transcendental Ideas

Conditions of Thought:
Deleuze and Transcendental Ideas
Plateaus - New Directions in Deleuze Studies

not ~ mat~er ?f bringing all sorts of things together under a single concept
but rather of relating each concept to variables rhat explain its mutations.'
Gilles Deleuze , Negotiations
It.'s
Seres Editors
Ian Buchanan , Cardiff University
Claire Colebrook , Penn State University
Editorial Advisory Board
Keirh Ansell Pearson
Ronald Bogue
Constantin V. Boundas
Rosi Braidotti
Eugene Holland
Gregg Lambert
Dorothea Olkowski
Paul Patton
Daniel Smith
James Williams
Titles available in the series
Dorothea
Olkowski b U
lh
CFztJ1Fztal Philosot)hyy
Christian Kerslake , Immanence and the Vertigo of Philosophy: From Kant to

Delze
Jean-Clet Martin , Variations: The Philosophy of Gilles Deleuze , translated by

Constantin V. Boundas and Susan Dyrkton
Simone Bignall , Postcolonial Agency: Critique and COllstructvism
1iguel de Beistegui , Immanence: Deleuze and Philosophy
Jean-Jacques Lecercle , Badiou and Deleuze Read Lterature
Ronald Bogue , Deleuzian Fabulatio 1Z and the Scars of Hstory
Sean Bowden , The Priority o{ Events: Deleuze's Logic of Sense
Craig Lundy , History and Becoming: Deleuze's Philosophy of Creativty
Aidan Tynan , Deleuze's Literary Clinic: Critcism and the Poltics of Symptoms
homas Nail , Returning to Revolution: Deleuze Guattari and Zapatismo
Franois Zourabichvili , Deleuze: A Philosophy of the Event with The
Vocabulary o{ Deleuze edited by Gregg Lambert and Daniel W. Smith ,
translated by Kieran Aarons
Frida Beckman , Between Desire and Pleasure: A Deleuzian Theory of Sexuality
Nadine Boljlvac Untimely A/ cts: Gilles Deleuze and an Ethics of Cinema
Daniela Voss , Conditio 1Z S o{ Thought: Deleuze and Transcendental 1deas
Forthcoming volumes:
LeRon Shults , Icolloclastic Theology: Gilles Deleuze and the Secretioll
{ Atheism
Janae Sholtz , The Inventi011 of a People: Art and the Political z Hidegger
and Dellze
Visit the Plateaus website at www.euppublishing.com/series/plat
CONDIIONS OF HOUGH;
DELEUZEAND
RANSCENDENTALIDEAS
05
Daniela Voss
B639751
<D Dani eI a Voss , 2013

Edinburgh University Press Ltd
22 George Squar Edinburgh EH8 9LF
www.euppublishing.com
Typeset in Sabon by
Servis Filrnsetting Ltd , Sto c: kport, Cheshire, ancl
prinrecl and bouncl in Great Britan by
CPI Group (UK) Ltcl , Croyclon CRO 4YY
A CIP recorcl for this book is available from the British Library
ISBN 978 074867625 5 (harclback)
ISBN 978 0 7486 76262 (webreacly PDF)
ISBN 978 0748676279 (epub)
The right of Daniela Voss
to be i lentifiecl as author of this work
has been assertecl in a c: corclance with
the Copyrighr, Designs ancl Patents Act 1988.
Contents
Acknowledgements
Abbreviations
vi
vii
1
Introduction
1 he Dogmatic Image of Thought
2 he Demand for Transcendental Genetic Conditions
3 ldeas as Problen1s
4 Time and the Split Subject
Conclusion
18
74
142
210
265
Bibliography
lndex
271
281
Acknowledgements
This book is the product of several years work which would have
been impossible without the generous support and advice of all those
who assisted me to bring this endeavour to completion.
1 especially owe rny heartfelt thanks to Paul Patton who was
always there for me to discuss problems and ideas and who encouraged me to overcome difficulties that 1 encountered over the course
of this project. Thanks to him 1 also had the opportunity to pursue
part of my research in the School of l-Ii story and Philosophy at the
UNSW in Sydney. The many kindnesses and friendship 1 experienced
there throughout the duration of my stay leaves me with an in1rnense
debt of gratitude to rnany people too numerous to mention them all.
1 am also very grateful to Gunter Gebauer from the Free University
of Berlin for his ongoing encouragement and enthusiasm for this
project. Furthermore , 1 would like to thank 111y friends and fellow
scholars Si1110n Duff) Craig Lundy and Nick Midgley who have
helped improve this work through their pertinent comments , critical
questions and linguistic advice. My friend Elke deserves a special
mention for providing the cover image for this book.
My great thanks are due to Carol Macdonald and the team at
Edinburgh University Press for their professional way of dealing
with all inquiries and working with me throughout the publication
process. My experience of publishing with EUP has been a very
pleasant one.
Finally , and above all , 1 would like to thank n1y family for ther
loving support and encouragement over all these years.
Parts of Chapter 2 were published !l aimon and Deleuze: The
Viewpoint of Internal Geness and the Concept of Differentials' in
Parrhesa , No. 11 (2011) , pp. 62-74. Furthennore , son1e materials of
Cha pter 1 and 3 i\Till appear in 'Deleuze's Rethinking of the Notion
of Sense' ar Deleuze's Third Synthesis of Tirne' in Deleuze Studi
(2013).1 would like to express rny gratitude for the pern1ission of the
editors of these journals to reprint this 111ateria l.
Vl
Abbreviations
The bibliography contains details of the editions and translations

used for each text.
Deleuze
AO
ATP
Anti-Oedipus: Capitalism and Schizophrenia (vo l. 1)

A Thousand Plateaus: Capitalism and Schizophrenia
B
CC
C1T
D
D1
Bergsonism
Essays Critical and Clinical
Cinema 2: The Time-1mage
Dialogues
Desert Islands and Other Texts (1 953-1974)
Difference and Repetition
Empiricism and Subjectivity: A Essay on Hume's Theory of
Human Nature
Foucault
The Fold: Leibniz and the Baroque
Kant's Critical Philsophy: The Doctrine of the Faculties
Lecture Course on Kant held at Vincennes , 14 March
1978
Lecture Course on Kant held at Vincennes , 21 March
1978
The Logic of Sense
Negotiations 1972-1990
Nietzsche and Philosophy
Proust and Signs
What Is Philosophy?
(vo l. 2)
DR
ES
F
FLB
KCP
LK 1
LK 11
LS
N
NP
PS
WP
Note on the citati ns: The order in which the page numbers are
referenced is as follows: the first page number refers to the English
translation , the second italicised number to the French origina l.
Vll
CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS
Kant
CJ
CPR
Critique of the Power of Jume11t

Critique of Pure Reaso11
Note 011 the citatio11s: Page references to CPR are to the original
German editions: A (from 1781) and B (frorn 1787).
Vlll
1troduction
This project on a Deleuzian transcendental philosophy is born out of

a feeling of astonishment. On the one hand , the theme of the transcendental runs through many of Deleuze's works , in particular those
published between 1962 and 1968. On the other hand , the spirit of
Deleuze's philosophical thought seems so very different from that
of Kantian transcendental philosophy: Deleuze does not bother to
seek a justification or ground for the possibility of experience and its
objects. He does not put together a table of categories nor does he
give any transcendental deduction of a priori conditions. Moreover ,
in Deleuze, there is certainly no transcendental subject, which would
have the task of representing the world according to a priori conditions. So why should he label his own philosophy of the 1960s a
transcendental empiricism (cf. DR 144/187 and 56/79 80)? ln what
way , if at all , is his philosophy transcendental?
Our approach to this problem has been guided by the intuition
that the key had to be found in Deleuze's critique of the so-called
dognlatic lmage of thought , a critique that appears in almost every
book from this early period and that makes up the central part of
Difference and Repetition. The first thing to be noted is that Deleuze
understands philosophy fundamentally as critique , and in Nietzsche
and Philosophy he explicitly demands a rethinking and radicalisation
of Kant's critical project. 1 The Kantian critique sought to describe
and ultimately prevent the illusions of reason that are to be found on
the battleeld' of metaphysics and to lead thought back to its proper
use. 2 By comparison , Deleuze also finds that Western philosophical
thought has fallen prey to illusion but for Deleuze it is the illusions
of representation' (DR 270/346) that must be subjected to critique.
Thought has been subordinated to sonle proper image of itself that
ties it to a logic of representation. Kant's critical philosophy sought
to overcome the traditional nletaphysical divide between the essences
of things or ldeas and their mere appearance. For instance , he
criticised Plato for the enthusiasm that led hinl to pass beyond the
concepts of experience to ideas which seemed to h ln explicable only
1
by means of an intellectual comrrlU nion with the origin of all things'. 3

Kant argued against Plato that necessity and universal truth , which
seem to pertain to the essence of things , actually have their ground in
our own reason and its ldeas. A critical use of reason would abstain
from any attelnpt at direct c Olnmunity with an intelligible world and
focus instead on ur world , which is given under the condition of
transcendental concepts. But this means that for Kant our world is
always a represented world , that is not an in1mediate presence but
a world given within a conceptual schen1e or structure that provides
the limit to knowledge. Now , Deleuze certainly agrees with Kant on
the need to overcome the metaphysical dualisn1 between the essences
of things and their appearance but he rejects Kant's representationalist solution to the problem. Deleuze radicalises Kantian critique by
bringing it to bear on the logic of representation itself. In his view ,
the idea of a representationallogic is still too oundationalist' since
it presupposes an a priori given ground that shapes our thinking.
For Deleuze , a radical critique must lead to a liberation of thought
from the fetters of representation; this means to render thought
ungrounded.
Although both Kant and Deleuze are cOlnmitted to philoso
phy understood as critique and the uncovering and warding off
of illusion , there thus remains a stark contrast between the two:
while Kantian transcendental philosophy airrls to provide an a
priori ground , Deleuze precipitates transcendental philosophy into
a 'groundlessness' (sans-fond) or universal ungrounding' (universel
eCondement) (DR 911123). Deleuze dissolves the representational
don1ain into a sub representational play of intensities or pure differences. While Kant locates the ground of our represented world
in universal reason or the transcendental subject, Deleuze adn1its
no transcendental subject that would represent the world but seeks
a point of view beyond representation ,
t ha
t is an a.-
.sub
s Jective an
unconscious transcendental eld. While Kant's transcendental principles are defined as a priori conditions that subordinate thought to
a priori judgernents , Deleuze , by contrast, seeks a transcendental
principle that explains the genesis of the act of thought in thought.
Given these fundan1en
lntroductin
be impossible , if not for some condition'.4 The necessity involved is

supposed to be rnore than a conceptual or logical necessity. Kant is
referring to a transcendental necessity that establishes a necessary
relation between concepts and objects. For if there were no necessary
relation to an object then the concepts would remain empty , that is
without any signification for us. he transcendental conditional can
thus be expressed as follows. Concepts would have no sense for us
at all , if they could not be constructed in a priori intuition , that is be
incarnated in a forn1al intuition of space and tirrle. Kant elaborates
the construction in formal intuition as a non-conceptual and extralogical necessity for the apparition of sense. One could say that
Kant's transcendental logic is a logic o{ sense. Now , the Deleuzian
concept of the transcendental also indicates a relation of necessity:
something {orces us to think. 5 We would not be thinking if not for
some transcendental condition. Furthern10re , Deleuze considers
the transcendental condition as the constitutive , genetic elernent of
sense. That is , the sense and value of our thoughts follow necessarily from the way in which the transcendental conditions (Ideas or
concepts) are incarnated in empirical tern1s and conditions. This is
why Deleuze insists time and again that we always have as much
truth as we deserve in accordance with the sense of what we say' (DR
154/200).6
The last and probably most important reason for Deleuze to adopt
the notion of the transcendental lies in Kant's definition of time as
a forn1 of interiority or inner sense. In fact , Deleuze cla lns that the
essence of the transcendental is found here. Kant discovers the transcendental when he criticises Descartes for failing to recognise time
as the form of the determinable , that is the inner forrrl through which
lny undetermined existence can be determined. For Descartes , the
forrn of the 1 think' immediately implies my undetermined existence ,
and too hastily he draws the conclusion that rrly existence can be
determined as a thinking subject. Deleuze contends that the entire
Kantian critique amounts to objecting against Descartes that it is
impossible for deterrrl nation to bear directly upon the undetermined'
(DR 85/116). Descartes fails to recognise that my own being is given
to the mind , or the form of the '1 think' , only under the inner form
of tIne. In other words , tin1e is the f
two aspects of human nature in the forrn of an internal Difference

which establishes an a priori relation between thought and being'
(DR 86 /1 16). This internal difference must be understood as what
constitutes the transcendental distinction. Establishing a transcendental distinction means interiorising it [difference] within being and
thought' (DR 86/117). For Deleuz the Kantian subject is essentially
a split subject, fractured by the pure and empty form of time. hat is
why , for him , the transcendental is not prirnarily rooted in the notion
of a transcendental epistemic structure conditioning an empirical
content (any such interpretation only re-inscribes hylomorphism ,
that is an external difference between form and matter) , but in the
notion of a fractured 1. Before Kant , noncritical philosophies both
of rationalist or empiricist legacy readily assumed as their point of
departure the identity of the subject , or the unity of a consciousness .7
Now with Kant philosophy has to deal with a split subject and time
as an inner form or form of auto ffection. 'If the greatest initiative
of transcendental philosophy was to introduce the form of time into
thought as such , then this pure and en1pty form in turn signifies
indissolubly the death of God , the fractured 1 and the passive sel f'
(DR 87/117). Frorn this perspective , it becomes comprehensible
why Nietzsche's philosophy , in particular his thought of the eternal
retur becomes such an irnportant reference point for Deleuze.
It also explains why Deleuze's account of the split subject and his
theory of the syntheses of time n1ake up a vital part of his own
transcendental philosophy.
It is perhaps not surprising that Deleuze's transcendental philosophy

will ultimately undermine the Kantian version of transcendentalisrn.
For Deleuze's approach to the history of philosophy displays an
extraordinary creativity of thought and boldness to read traditional
philosophical texts in a new n1anner , sometimes even against their
original sense (in a contre-sens' so to speak). Deleuze clairns that
critical engagelTIent with the history of philosophy is of necessity
creative: criticis111 has to lead to the construction of new concepts
and thereby contribute to the creation of concepts understood as
the essential task of philosophy (cf. 83/80).Concepts' Deleuze
says , 'are not waiting for us ready-made , like heavenly bodies' (WP
5/11) , they need to be created. Deleuze reveals his comrnitment to a
creative conceptual constructivism throughout his writings on other
philosophers (for example , his early books on Hun1e , Nietzsche ,
Kan Bergson and Spinoza , as well as his book on Leibniz frOlTI
4
!ntroduction
1988). hese rnonographs do not represent neat commentaries on
other thinkers , but rather a sort of symbiotic style of writing: Deleuze
enters into what Paul Patton has called a ual and paradoxical relation' with these philosophers , thereby undermining the traditional
author-function. 8 As Deleuze says , he always needed intercessors
(i ntercesseurs) who would lend their \vords and through which
one could say what one has to say (N 125/171). This technique
of ventriloquising' another thinker allows Deleuze to express his
own thoughts through the words of others , resulting sometimes
in an indiscernibility between the different voices. Zourabichvili
suggests calling this technique 'free indirect discourse' (discours
indirect libre) probably referring to Deleuze's definition of ree
indirect discourse' in A Thousand Plateaus. 9 In the case of Deleuze's
co-authorship with Flix Guattari it is indeed almost impossible
to discern the different voices; the two of them truly function as
a collective assemblage' (agencement collecti{) (A TP 80/101) and
it is not necessary to emphasise that these works have a doubled ,
collective source. However , with regard to Deleuze's writings on
philosophers who already have a distinct identity in the history of
philosophy , it is necessary to point out that his accounts of Plato ,
Nietzsche and Kant transform or metamorphose these thinkers into
a Deleuzian Plato , a Deleuzian Nietzsche , a Deleuzian Kan t. In the
course of this book , we will at tirnes call attention to this typical
Deleuzian symbiotic style of doing philosophy , in order to prevent
n1isunderstandings or accusations that Deleuze misinterprets' the
philosopher under consideration. We also have to keep this in mind
when we consider Deleuze's notion of the transcendenta l. It is surely
Kantian in origin , but Deleuze will transform the Kantian notion of
the transcendental in order to remedy what he sees as a lack in that
notion (i.e. the failure to address the problern of genesis) and thereby
turn it into a weapon against Kant's transcendental philosophy itself.
In fact , Deleuze not only turns against Kant but against a whole
tradition of Western thought obsessed with the idea of a 'beginning'
in philosophy that is the idea of a foundation or of grounding principles. His allies in this combat against foundationalism' are various
post-Kantian philosophers such as 1\
particular Difference and Repetition (1968). We will also refer to

Nietzsche and Philosophy (1962) , Kan t' s Critical Philosophy (1963) ,
Proust and Signs (1964) , Bergsonism (1966) and The Logic of Sense
(1969) , as well as to son1e articles and lecture courses mainly from
the 1960s and 1970s. Only occasionally will we refer to Deleuze's
later uvre and only in cases where the reference helps to bring out
certain ideas that rernain unclear or concealed in his early writings.
As Deleuze becomes rrlore involved with psychoanalysis and politics
subsequent to the reTolutionary events of May 1968 in France and the
encounter with Guattari , his philosophy and style of doing philosophy change. His thought is captured by new problems that require
a rnodified use of familiar concepts and the creation of entirely new
concepts.herefore it would have a confusing and distracting effect
if we included some of the serrl nal concepts and thoughts of the
later works. By no means do we want to claim that Difference and
Repetition is Deleuze's one and only masterpiece and the climax of
his philosophical thought. This being said , the project of transcendental philosophy undertaken in Difference and Repetition deserves
to be examined closely in its own right , not only for the contributions
it n1akes to a criticism of Kantian transcendental philosophy but also
for the elaboration of a highly original , differential and genetic model
of the transcendenta l.
When we started on this project in 2007 there was a conspicuous
lacuna in the Deleuze scholarship with regard to the nexus Deleuze
and Kant. The only monograph on the topic available at that tirrle
was Marc R lli's dissertation on Gilles Deleuze and the philosophy
of transcendental empiricism , published in 2003. 1n this work R lli
undertakes a very detailed exploration of the respective traditions of
empiricism and transcendental philosophy with a particular focus on
Deleuze's early book on Hume , Empiricism and Subjectivity (1953).
He also devotes a great deal of attention to Husserl and Heidegger ,
pointing to the importance of Heidegger's Kant and the Problent
of Metaphysics for Deleuze's three syntheses of time. R lli's study
is very clearly written , he analyses Deleuze's criticisn1 of Kant with
great accuracy and presents Deleuzian transcendental empiricism as
a coherent philosophical system.
explains the genesis of experience by self-organising pro
lntroductin
sense in which Deleuze's philosophy is transcendental , R lli focuses

on the second part of the term 'transcendental empiricisIT1' , which
is why he gives so much space to Deleuze's reading of Hume. Now
it is certainly true that Hume plays a role in Deleuze's criticism of
representation. In Empiricism and Subjectivity Deleuze puts great
emphasis on the external character of relations: Hume's principles
of association are not subordinated to any representational totality ,
logic or schen1e. However , Hume provides no insight into the genesis
of real experience and as such , the reference to Hume is not relerant
to Deleuze's project of rethinking the transcendental as a genetic
principle. In fact , it is to rationalists such as Leibniz and Maimon
that Deleuze turns in order to extract the concept of internal differential relations which function as the genetic element of the production of the rea l. In Dirence and Repetition , Deleuze underlines
the importance of internal relations , that is relations of difference
where the connecting factor is difference qua difference ,10 and even
calls it a mistake' of empiricism 'to leave external what is separated'
(DR 170/221).11 R lI i does not elaborate the notion of differential
relations , instead giving preference to the concept of intensity. Not
surprisingly , he mentions Salolnon Maimon only in passing 12 and
even criticises Daniel Sn1ith for having overestInated the importance
of Maimon for Deleuze in his (unpublished) doctoral thesis on
Deleuze's philosophy of difference. 13 Our study seeks to show , to the
contrary , that Maimon's influence and the theory of internal differ
ential relations are of central in1portance to Deleuze's rethinking of
the transcendenta l. For this reason we wi lI focus on the contribution
of Maimon , Nietzsche and Bergson to Deleuze's thought , rather than
on that of Hun1e or phen0111enology.
Only in recent years have there been further publications on
Deleuze and Kant , and on Deleuze's transcendental empiricism. Levi
Bryant's book Dirence and Givenness: Deleuze's Transcendental
Empiricism and the Ontology of lmmanence appeared in 2008.
It is a very fruitful examination of Deleuze's relation to Kant and
provides many interesting insights. For Bryan the crucial point
of divergence between Deleuze and Kant bears on the character of
intuition: While Kant takes intuition to be something given , Deleuze
claims that intuition is essentially productive. It itself creates the
objects of intuition by ITleanS of differential mechanisn1s. Bryant
concludes that insofar as differentials function as the productive
rules for the qualitative givens of being , Deleuze's position is best
7
thought of as a hyper-rationalism rather than an errlpiricisn1\14

Bryant clearly recognises the in1portance of internal differential relations for Deleuze's philosophy (although he refers to MaInon only
occasionally).15 However , we would argue that Bryant's choice of
terms is not always helpful; for instance , it is rather lnisleading when
he states that Deleuze is led to rationalize intuitions thernselves ' l 6
or that experience is for Deleuze intelligible in principle'17 and that
Deleuze's philosophy locates intelligibility at the level of the aesthetic or sensible itsel f' .18 We believe that these traditional philosophical concepts relating to rationalism and empiricism cannot easily
be applied to Deleuze's thought. Both rationalism and ernpiricism
take as their point of departure the presupposition of a pre-existing
consciousness endowed with ready-made faculties and then go on to
examine its possible relation to an objective world. For Deleuze , on
the contrary , the subject and the object are not determined entities
given in advance but the effect of differential processes that take
place in the virtual and extend thernselves into processes of actualisation. Hence choosing the viewpoint of a subjective consciousness
and analysing experience in terms of intellectualised intuitions is a
rather ill-advised strategy for it neglects Deleuze's endeavour to seek
a viewpoint of intrinsic genesis beyond a philosophy of mind and a
theory of representation.
ln his attempt to forrnulate Deleuze's philosophy as an ontology ,
Bryant pays much attention to Bergson's influence on Deleuze. ln
fact , Bryant presents Deleuze as 'jurnping out of the critical philosophy' into an analysis of the being of time as a pure transcendental
field. 19 It is shown that critical thought ultimately reaches the point
where it becomes indiscernible from speculative thought. 20 There is
certainly sorne truth to this claim. Although Deleuze's transcendental
philosophy in no way means a return to a pre-critical metaphysics
that presupposes an inaccessible transcendent realm in principle ,
he does outline a metaphysics of immanence. As Deleuze contends
we have the rneans to penetrate the realrn of the sub-representative
(DI 115/161) , to explore virtual ldeas (DR 194/25 1), to live and to
experience the pure past (cf. B 122/55 , note 16/1) and even to create
fragments of the pure past in
Introduction
early philosophy is not so rnuch the question of 'what is therebut
rather what are the real conditions of (experience , thought , consciousness , subjectivity , objects , and so on)This approach searching for the real conditions of genesis is best described as Deleuze's
transcendental viewpoint.
Another major work in this field is Christian Kerslake's Immlence
and the Vertigo of Philosophy: From Kant to Deleuze published
in 2009. Unlike Levi Bryant , Kerslake proposes a rnore aesthetic
and existential reading of the transcendental conditions of thought.
he transcendental ldeas that provoke thought and rnake it enter
into psychic repetition' are real problematic experiences beneath
the mystifications of everyday ( ideological') reality , such as birth
death , sexual difference and so on. 21 he ldeas or real problems'
are necessarily unconscious but can be realised in various ways:
they may be treated from the perspective of knowledge claims , but
they can also be given indirect presentation in art , theatre , literature ,
music and cinerrla. Kerslake reads Deleuze's transcendental empiriclsn1 as a direct continuation of the Kantian turn' as a project
of 'lnetacritique' begun by the post-Kantians Maimon , Schelling ,
Novalis and Hlderlin. 24 In a footnote , Kerslake makes it clear that
his study is only concerned with Deleuze's philosophy in the works
of 1953 to 1968. 25 In fact , the n1ajor reference point for his study
is Deleuze's very early and newly discovered lecture series 'What
Is Grounding (Qu'est-ce que fonder?) , held at the Lyce Louis le
Grand fron1 1956 to 1957. 26 In this lecture course Deleuze "enacts a
repetition of the Kantian enterprise" , working through the premises
of Kantian , post-Kantian , and Heideggerian existential approaches
to "self-grounding" in philosophy' .27 On this basis , Kerlake argues
that Deleuze's real questions and problems emerge from within the
post-Kantian tradition of philosophy'. 28
lt is certainly a great achievement to have set out the relation of
Deleuze's project to that of the post-Kantians , but the danger is that
Deleuze's philosophical questions and problen1s are not evaluated
in their own right. Indeed , Kerslake neglects the originality and
radical openness of Deleuze's investigations by identifying them with
the post-Kantian project. 'The main claim of this book is that the
philosophical work of Gilles Deleuze represents the latest flowering
of the project , begun in the irnmediate iV ake of Kant's Critique of
Pure Reason , to complete consistently the Copernican revolution"
in philosophy.'29 is strong claim leads Kerslake in some cases to
9
a bizarre reading of Deleuze , as if Deleuze's philosophy only served

to arnend transcendental philosophy and to solve Kantian and postKantian problems. According to Kerslake , Deleuze rebaptises the
faculty of reason as the faculty of "thought'" ,30 but Kerslake thereby
ignores the fundarnental differences between Deleuze's conception
of non--representational , nomadic and creative thought and the
role of reason in Kant's epistemology which comes down to securing
the coherent errlployment of the understanding and the production
of systematic theoretical knowledge. In Kant's practical philosophy
reason appears as the sole legislative power. Is it by analogy with this
Kantian autononlY of reason that Kerslake finds the conception of an
autonomy of thought in Deleuze? Kerslake contends that if thought
'distrusts "comlnon sense" , and instead seeks autonomy , it must then
learn how to regulate itself'31 and it is for this reason that 'Deleuze
develops this model of a free , selgrounding generation of the relations of a systeln of faculties 32 We would argue , by contrast, that
Deleuze's genetic model of the transcendent exercise of the faculties
by no means relies on autonomy and self-regulation , but rather on
heteronomy.33 Deleuze repeatedly insists that we are forced to think
and that it is a violent experience which even takes from us the power
to say '1'.
Kerslake's attenlpt to align Deleuze's project with post-Kantian
philosophy , in particular Gernlan Idealisln , also leads hirrl to discover a new subject in Deleuze's philosophy. As is well known ,
Fichte , Schelling and Hegel all searched for an unconditioned , selfgrounding principle of subjectivity that would provide a ground for
our conscious experience. Kerslake presents Deleuze's philosophy as
an extension of Schelling's metaphysical enlpiricisln:
We wiU present Deleuze's philosophy of immanence as caught between
two poles , represented by the late Fichte on the one hand , and the
Absolutes' of Wronski and the late Schelling on the other. Insofar
as his metacritical , systematic approach remains valid , we will argue
that Deleuze's nal resubjectification' of Life [in Immanence: A Life']
signals his arrival at the same point as Wronski and the later Schelling,
who ended up positing the existence of subjectivity within a primordi
ally living . . . actual being' , a being that is preceded by no other and is
therefore the oldest of all beings'. [Schelling , The Ages of the World]3
Kerlake's claim that there is a 'resubjectification' of life in Deleuze's
last essay 'Imrrlanence: A Life' , first published in 1995 seems
pecular in the light of Deleuze's own claim that he talks about
10
lntroduction
an indefinite , impersonal life , precisely a life accompanied by the
indefinite article. 35
Kerslake's study is valuable for the information it provides on
the metaphysical origins of Kantianisrn , on the evolution of Kant's
critical philosophy starting from his Inaugural Dissertation , and on
the projects and endeavours of various post-Kantian philosophers.
However , in this mass of detail the reader risks losing track of its
relevance to Deleuze: Kerslake's identification of Deleuze as the
inheritor of the project of post-Kantian philosophy loses sight of
what radically differentiates Deleuze from this philosophy.
he
most recent study in the field of Deleuze's transcendental empiricism was released in November 2009. Anne Sauvagnargues' book
Deleuze: L'Empirisme transcendantal analyses Deleuze's transcendental philosophy better and more deeply than any prior commentato r. Sauvagnargues both covers Deleuze's many philosophical and
literary sources and offers a comprehensive and convincing account
of Deleuze's transcendental empiricism. Our book has benefited considerably from this excellent study , which ought to be an important
reference point for Deleuzian scholars. However , our own project
differs from Sauvagnargues' book in that it examines luany topics in
much greater detai l. For instance , our analysis of Salon10n Main10n's
philosophy provides insights into his critique of Kant , his own
philosophical system and background infonuation about the early
phase of Leibniz's and Newton's differential calculus , all of which
is covered by Sauvagnargues but not explained in any depth. On the
other hand , Sauvagnargues covers a much wider range of Deleuze's
sources: apart from Kant , Proust, Bergson , Nietzsche and Mairnon ,
she also deals with Spinoza and Simondon , and in particular devotes
luore time to exploring the rrlaterial or bodily aspects of transcendental empiricism. Since our project focuses rather on the transcendental
conditions of thought and the critique of the illusionary effects of the
dogmatic Image of thought , the aspect of material nature does not
play a large role. However , anyone iVho wishes to have a broader
picture of Deleuzian transcendental empiricism is recommended to
consult Sauvagnargues' seminal book.
Throughout our st l1 dy , Daniel Srrl th's essays on Deleuze in relation

to Kantian Ideas , Leibnizian principles and Maimonian differentials
provided a very helpful source. 36 he recent collection of essays
Deleuze's Philsophical Lineage , edited by Graham Jones and Jon
11
CONDITIONS OF
HOUGHT:
DELEUZE AND TRANSCENDENTAL IDEAS
Roe and published by Edinburgh University Press in 2009 , also

proved to be a helpful compendium , since it presents a series of thinl
ers (no t: only philosophers) that have been influential for Deleuze
but whose thought is not necessarily widely known. 37 Particularly
important for this study has been Simon Duffy's article on the mathematician Albert Lautman (1908 -4 4) who provided Deleuze with a
mathematical and dialectical model of problematic Ideas.
his
present study will examine the precise nature of Deleuzian transcendental conditions , in particular the transcendental conditions of
thought , i.e. that which compels thought to transgress the established
boundaries of represen t: ation. We wiU argue the following. (1) For
Deleuze transcendental conditions need to be geneti c. They have to
account for the genesis of sensations and the variations of sense. It
should be noted that Deleuze does not seek to give an account of
the genesis of every particular given , here and now , but rather of
a compound of sensations or a sign which can also mean a literary
work , a painting , a thought. 38 (2) The conditions must not be larger
than what they condition. They are deterrnined at the same time
as they determine the conditioned. Borrowing a Nietzschean tern1 ,
transcendental conditions are plastic in opposition to the invariant
and abstract transcendental conditions of Kantian legacy. (3) he
transcendental for Deleuze does not designate a rst condition , a
beginning or ground but rather a groundlessness' or 'universal
ungrounding'. The ground is devoured so to speak by an 'original
depth' which is to be conceived as a pure intensive 'spatium' (DR
230/296) or a differential multiplicity. (4) he multiplicity of differential relations is immanent to this world as its virtual half or
dirnension. (5) The Deleuzian transcendental conditions are further
defined as Ideas or problems that we encounter in our unconscious
and sub-representational mode of existence. They demand solutions , but the way in which they are actualised and answered in
empirical terms and circumstances can never exhaust their ideal and
problematic character. (6) Ideas or problems do no t: appeal to given ,
ready-made faculties that are already regulated wi t: hin a harmonious accord. Rather , they transfonn those faculties forcing then1 to
transgress their limits in a transcendent and paradoxical exercise. (7)
The va 1' iations of Deleuzian t 1' anscendental conditions which have
been specified as diffe 1' ential Ideas 0 1' problen1s , n lU st be thought of
as unfolding outside historicallinear tinle. Deleuze envisages a thi 1' d
synthesis of time in trms of Nietzsche's eternal 1'eturn and 1'elates
12
Introduction
it to a dissolved subject or fractured 'I' which is shown to be the

correlate of the transcendental conditions of thought.
The method of this study involves a close reading and interpretation of Deleuze's texts from the 1960s, with occasional references
to his later works. It will be noticed that we rarely make use of
Deleuze's The Logic of Sense although it was published in 1969 only
one year after Difference and Repetition , and Deleuze was apparently
working on their ideas simultaneously. However , it seems to us that
they stand as two separate sources of his work , and each should be
treated in its own right. While Difference and Repetition is centred
on rethinking Kant's transcendental , The Logic of Sense draws on the
Stoic ontological distinction between bodies and incorporeal senseevents. Not only the terminology but also the whole philosophical
framework of these two books differ considerably from each othe r.
The ain1 of this study is to give a systematic reconstruction of what
can count as a Deleuzian transcendental philosophy. The decision
to focus on Nietzsche , Maimon and Bergson and to include thinkers
such as the mathematicians Albert Lautman , Karl Weierstrass and
Bernhard Rielnann has been dictated to us by the very nature of the
task of determining a transcendental which is conceived as a genetic
and differential multiplicity with varying relations and located within
an unconscious , virtual realn1.
Chapter 1 will exan1ine what Deleuze understands by the expresslon iogmatic Image of thought' and why he rejects it. One of the
n1ain reasons for his rejection is the adherence of the dogn1atic Image
of thought to the model of representation , which operates through
judgements on the basis of the identity of concepts , thereby excluding
all the non-conceptual or pre-linguistic differences and repetitions
that cannot be identified. he n10del of representation denies these
differences and repetitions a concept of their own and dismisses
the genetic and creative force that allows them to produce signs or
senses that cannot be captured by logic. By means of a criticism of
the dogn1atic Image of thoug Deleuze opens the path to a detailed
exploration of the transcendental as consisting of genetic , differential
forces beneath the order of representation.
Chapter 2 will focus on Deleuze's den1and for transcendental
genetic conditions. This dernand arises from a
13
combination of forces that determines the sense and val l1 e of what

we feel , say or think. The Kantian transcendental conditions , by contrast , are Cl1 t off from their genealogical origin and presented as p l1 re
a priori givens. This is why they are too large (general or abstract)
with regard to what they condition. On the other hand , Kantian transcendental conditions are also still too sin1ilar to the psychological
conditions governing the empirical cases of recognition. Kant simply
traces the transcendental from the ernpir al and projects the newly
discovered transcendental backwards as the transcendental gro l1 nd
of recognition. We will arg l1 e that in search of transcendental conditions that are not only genetic b l1 t also l1 nconditioned then1selves ,
that is heterogeneo l1 s to the series of conditioned conditions , Deleuze
finds what he requires in Maimon's concept of differentials of consciousness'. For MaimoI the sub-conscious n1anifold of genetic di
ferential elements serves as an explanatory principle for the manner
in which real experience arises.
Chapter 3 i\1ill take l1 p the theme of problen1atic Ideas as it is laid
out in Kant's Critique of Pure Reason and Critique of the Power of
Judgment. We will explore the way in which Ideas count as problems
'without sol l1 tion' (CPR A 328/B 384) and how they are capable
of animating Ol1 r cognitive fac l1 lties to transcend their bO l1 ndaries
and thereby el l1 de determining j l1 dgen1ents. In this context, Kant's
model of the experience of the sublime will play an important role.
The last part of this chapter will investigate how Deleuze develops
Kant's notion of Ideas into his own theory of Ideas , and analyse the
str l1cture , internal operations and the movement of act l1 alisation of
Dele l1 zian Ideas. In the elaboration of his lialectics of Ideas' Dele l1 ze
is inspired by the philosopher and mathematician Albert La l1 tman
whose ideas will be addressed in this chapter.
After having specified the transcendental conditions of tho l1 ght as
genetic and differential Ideas endowed with a problematic character
and located within a virtual realrn , we will then in Chapter 4 exarnine
the way in which the transcendental conditions are related to a
split subjec t.'e wiU follow Dele l1 ze's reading of Kant according to
which Kant's invention of the pure and en1pty forrn of time not only
breaks with the ancient circ111ar n10del of time , b l1 t also introd l1ces a
fract
14
Jntroduction
opening up the subject and referring it to the realm of Ideas. We will

illustrate this relation by looking at how Deleuze makes use of Pierre
Klossowski's account of Nietzsche as a thinker of the eternal return.
On the basis of these four chapters , we hope to offer a comprehensive account of Deleuze's transformation of transcendental philosophy as it is usually conceived. By doing this , we airn to convince
the reader of the originality and critical power of Deleuze's own
conception of the transcendenta l.
Notes
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
C f. NP 1 8990/102 9 1/ 104 and 106112 1.

Kant , Critique of Pure Reason , B xv; hereafter cited as CPR.
Ka Critique of the Power of Judgment , 5: 363; hereafter cited as CJ.
Franl All or Nothing , p. 204.
See DR 139/182 , 275/353 and also PS 97/189.
See also DR 159/206, NP 1/2 , 1041118 , 110/125 and B 16/5. In
Chapter 1 in the discussion of the seventh postulate of the dogmatic
Image of thought , we wiII show in what way the determination of a
problem and its conditions releases a necessary force that urges us all
the way down to the very end of the problem's necessary implications.
It should be noted that Deleuze explicitly exempts Hume's empiricist
philosophy from this charge. In his early book on Hume , Empiricism
and Subjectivity , Deleuze attempts to show that subjectivity in Hume
does not have the characteristics of a pre-existing subject. On the
contrary , empirical subjectivity is determined as an ef)ct of principles
transcending and affecting the mind: 'it is in effect an impression of
refl:tion. The mind having been affected by the principles , turns now
into a subjec t' (ES 26/8).
Patton , Deleuze: A Critical Reader , p. 3.
Zourabichvili , Une philosophie de l'vnement' , p. 14. C f. Deleuze's
defnition of 'free indirect discourse': there are no c1 ear , distinctive
contour what com
le
S rst is not an insertion of variousl
y indiv
iduat
ed
statenlents or an interlocking of diff~ent subjects of enlUnciation but
a collective assemblage (agencement collectif) resulting in the deter
mination of relative subjectifcation proceedings , or assignations of
individuality and their shifting distributions within discourse. Indirect
discourse is not explained by the distinction between subjects; rather ,
it is the assemblage (agencement) as it freely appears in this discourse ,
that explains all the voices present iVithin a single voice' (ATP 80/101).
In fact , the emphasis on internal difference can also be found in some
other of Deleuze's early writings , in particular in the context of Bergson's
concept of duration or non-numerical , qualitative multiplicity; see B
15
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
2 1.
22.
23.
24.
25.
26.
27.
28.
29.
30.
3 1.
32.
33.
34.
35.
42/36 and DI 32-5 1/43-72. See also PS 4 1/54: But what is an absolute , ultimate difference? Not an empirical difference between two
things or two objects , always extrinsic. [. . .] In this regard , Proust is
Leibnizian.'
This comment is made when Deleuze criticises Kant for leaving concepts and intuition external to one another: in this sense there is sti Il
too much empiricism in the Critique' (DR 170/221). Deleuze blames
Kant's transcendental philosophy for not being transcendental enough.
Rlli Philosophie des transzendentalen Empirismus , p. 10 and p. 34.
Ibid. , p. 70. Unfortunately , Daniel Smith's dissertation The Concept
of 'Difference' in the Philosophy of Gilles Deleuze , accepted by the
University of Chicago (Ill inois) in March 1997, is not available to us.
Bryant , Difference and Givenness , p. 1 1.
Ibid. , pp. 8 9 , 46 , 194 , 242.
Ibid. , p. 28.
Ibid. , p. 63.
Ibid. , p. 12.
Ibid. , p. 18 1.
Ibid. , p. 183.
Kerslake , Imma ence and the Vertigo of Philosophy , p. 9 1.
Ibid. , p. 4.
Ibid. , p. 5.
Ibid. , p. 7.
Ibid. , p. 42 , note 8.
The manuscript of the lecture course , recorded by Pierre Lefebvre , is
available online at <http://www.webdeleuze.com/php/sommaire.htmb.
Kerslake
17
Ibid. , p. 3.
Ibid. , p. 5.
Ibid. , p. 83.
Ibid. , p. 84.
Ibid. , p. 8 1.
Cf. Sauvagnargues , Deleuze: L'Empirisme transcendantal, p. 84.
Kerslake , Immanence and the Vertigo of Philosoph) pp.212-13.
Using the example of Charles Dickens' story Our Mutual Friend ,
Deleuze shows that the life of the individual gives way to an impersonal and yet singular life that releases a pure event freed from the
accidents of internal and externallife , that is , from the subjectivity and
objectivity of what happens'. See Deleuze Immanence: A Li fe' , p. 28.
36. To name only a few of his essays that we 1' e particularly important
fo 1' us , these a 1'e (1) Deleuze's Theory of Sensation: Overcoming the
Kantian Duality (2) Deleuze Kant , and the Theory of Immanent
Ideas' , (3) Genesis and Difference: Deleuze , Maimon , and the PostKantian Reading of Leibniz'.
16
lntroduction
37. See , for instance , the following essays in this volume: (1) Simon Duffy ,
Albert Lautman' , pp. 356-79; (2) Graham Jones , Solomon Maimon' ,
pp. 104-29; (3) 1elissa Mc1ahon Immanuel Kant' , pp. 87-103; (4)
Daniel Smith , G. W. F. Leibniz' , pp. 44-66.
38. Levi Bryant calls them 'topologic or morphological essences' ,
describing such an essence as a possible world , a system of appearances , a way of being'. As examples he refers to the style of an author
(the worlds of Proust , Joyce or Kafka) , the unique patterns of seashells
or the migration patterns of birds (cf. Bryant, Dierence and Gvenness ,
p. 66). In our view , his examples seem to be too close to a structuralist
viewpoint. A painting by Francis Bacon , for instance , captures intensive
forces or dynamisms that do not belong to the world of appearances
but to an intensive depth , which is made visible in the vibration of
colours and deformation of bodies. Thus Bacon adds a vertical layer ,
that is a dynamic , intensive and temporal dimension to the merely
horizontal structure of the good form' of conventional representation.
Deleuze makes use of the term sign' to refer to this dynamic or genetic
structure. We prefer the term sign' to Bryant's term of morphological
essence in order to elude the classical philosophical notions of essence
and appearance , already laden with meaning.
17
The Dogmatic Image of l'hought
The image of thought is a recurrent thenle in Deleuze's philosophy:

it appears already in his writings on Nietzsche (1962) and Proust
(1964) , and is then fully laid out in his magnum opus Difference and
Repetition (1968).
At first sight , it might seem peculiar to talk about an 'irnage' of
thought , for it seems that the realm of images or the imaginary s
opposed to the realm of thought or the intelligible. Is the expresslon image of thought' not a contradiction in terms? Must we
not distinguish carefully between the inlaginary and thought , or in
fact between the imaginary and the real? In one of his early essays
Deleuze remarks:
We are used to , almost conditioned to a certain distinction or correlation
between the real and the imaginary. A Il of our thought maintains a diaIectical play between these two notions. Even when classical philosophy
speaks of pure inteIl igence or understanding , it is still a matter of a faculty
de f ned by its aptitude to grasp the depths of the real (le rel en son fond) ,
the real in truth' , the real as such , in opposition to , but also in relation to
the power of imagination. (DI 171/239)1
Thus in the classical tradition of epistemology, the understanding
is understood as the faculty of thought capable of grasping the real
or the true essence , while the faculty of imagination grasps nothing
on its own because it produces only an endless series of varying
images or representations. The imagination is powerless with regard
to the recognition of true essences. This classical binary rnodel of
thought and imagination is still preserved in the psychoanalytic tradition. For instance , Deleuze criticises Freudianisnl for differentiating
between he reality principle with its power to disappoint , [and]
the pleasure principle with its hallucinatory power of satisfaction'
(DI 17 1/240) , since it reinscribes the binary model between the real
and the imaginary. Deleuze further asserts that the lnethods of Jung
and Bachelard also endorse this mode l. Psychoanalysis is consigned
to the analysis of the imaginary , which 'is defined by games of
18
The Dogmatic Image of Thought

mirroring , of duplication , of reversed identification and projection ,
always in the mode of the doub (DI 172/241 ).2 Sufce here the
example of the distinction between the real father and the variety of
father-images. On the whole , the psychoanalytic tradition uses the
term image or imaginary to refer to deceptive representations that
disguise and conceal the thought of the real and thereby disable the
subject through fixations which are produced by the unconscious
mechanisms of imagination.
he binary model of the real and the imaginary even extends itself
into socio-political theory in the distinction between real men and
their real relations versus ideologies and imaginary relations. In fact ,
the notion of ideology suggests the existence of some brute reality
which is concealed , distorted or oppressed by the grip of an in1age.
For this reason , Deleuze takes a highly critical stance toward the
notion of ideology understood as some repressive idea or image to
which we submit. 3 In S llI Delellze rejects the simple binary model of
the real and the imaginary. This brings llS back to the initial q llestion
as to how to llnderstand the meaning of the terrn image of thOllght'.
Delellze does not take something real as a starting point , such as a
natural capacity to think which is then obstrllcted and distorted by
a deceptive irrlage. Rather , for Delellze , there is an intertwining of
thOllght and image. Phrased more precisely , thought itself is produced by an image that acts like a n1achine which is coding thOllghts
in accordance with some normative form. As Claire Parnet explains
in Dialogues , co-allthored with Delellze:
Images' here doesn't refer to ideology but to a whole organisation which
effectively trains thought to operate according to the norms of an established order or power , and moreover , installs in it an apparatus of power ,
sets it up as an apparatus of power itsel f. (D 23/31)
ThllS for Deleuze , a critiqlle of the In1age of thOllght cannot simply

mean to llnveil the trlle essence of thought by wiping Ollt SOlne
false image. Instead , the Image of thOllght has to be considered as a
prodllctive lnachine or apparatlls of power: we have to examine its
n1echanisms that prodllce thought according to certain axes or rllles
and n1akit serve SOlne preconceived ends. Although in his early
work Delellze has not yet found the concepts of n1achine' , 'coding'
and apparatlls of power' , he does speak of the In1age of thought as
an established order (i.e. the orcler of representation1) and analyses
it in
t er
~ms
think and what the ul
tim
1ate goals of thought are. These postulates
19
need not operate visibly; rnore often they rernain llnconsciollS

forming implicit subjective presllppositions , which are essentially
pre-philosophical or non-philosophical in natllre. his means that
they are already at work when we start to think , althollgh we 111ay
not always be aware of it.
Why is it that Deleuze speaks of a single Image of thought? To be
sure , throughollt the history of philosophy there have been 'variant
forms' (DR 13 1/172) of the image of thOllght. Thus the Greek
lnage of thought manifest in Plato certainly differs from the modern
Cartesian image of thOllght. Again , rationalists hold to a different
form of image of thought than empiricists. Despite the various forms
of images of thought pertaining to each philosopher or style of
philosophy , there are , according to Deleuz some features that are
repeated and can be regarded as variations within a single movement.
They form 'a single 1mage in general' (DR 1321172) - 'i111age' written
with a maj llscllle - which Deleuze specifies as a logma tic , orthodox
or moral image' (DR 13 1/172). This 1mage is particlllarly detrimental , inaslnllch as it obstrllcts the genesis of a thought that ellldes the
requirements of representation and that is capable of calling forth an
absolllte , wholly llnpredictable and new , fllture. '1n effect, thought
is covered over [se recouvre] by an "image" n1ade up of postlllates
which distort [dnaturent] both its operation and its genesis' (DR
265/341 ).
One of the main featllres of the dogmatic 1mage of thought is the
privileging of identity over difference. Mo1
ifference
d
i s distorted by the 'fou1r illusions of r.epresen
1t
tior}' (DR
a
270/346
): identity , resemblance , opposition and analogy. 1n particlllar , identity is preserved in the identity of the thinking sllbject,
which serves as the principle for the production of abstract concepts
or llniversals subs Ulning the diversity of the sensible llnder the identity of the concept. Thus differences in the sensible are cance l1 ed in
fa !Ollr of resemblances and covered over by the identity of qualities
or the qllantity of extension. Opposite predicates divide the diversity
of the sensible in a binary manner into lnlltllally exclusive properties and by attributing only one property of a pair ofopposites to
each thing , detennine this thing in accordance with the ideal of
cornplete deterrnination. Furthermore , there are certain determinable concepts (categories or genera of Being) , which distribllte Being
in analogy with the relaton between large genera and their species.
In Sllmlnary , not only s the world of the sensible denatured by
20
The Dogn1atic Image o{ Thought

being covered with a conceptual net of identical concepts , but also
the nature of difference itself is distorted. Difference is treated as a
default of identity and lacks a concept in its own right. Difference
is either difference without a concept, concerning only external
relations between individuals (being outside one another) , or it
is reduced to a merely conceptual difference (specific difference ,
generic difference) , which allows the identity of a concept to subsist.
As Deleuze says , there is conceptual difference , but no concept of
difference (c f. DR 27/40-1).
The task wiU be to establish a concept of difference and to restore
difference to being and thought. Difference in being operates as the
individuating force of nature , while difference in thought implies
the fracture of the thinking subject , which is the necessary genetic
condition for the 'genitality of thinking' (DR 266 /3 42). Deleuze
insists that thinking is not an innate faculty given by nature and
conditioned by ready-made concepts rooted in our mind. Thinking
rnust be engendered in thought , but to engender thinking in thought
does not lnean to follow a safe and secure method leading up to
truth. Rather , thinking arises through the encounter with something
that ~ rces it to think. Thinking occurs by chance and through
constramt.
Deleuze's Critique of the Dogmatic lmage of Thought

Deleuze's rejection of the dogmatic In1age of thought is motivated
by its deeply conservative and moral character. It is conservative
because it distributes our experience according to 'sedentary' and
fixed determinations. he language of territories' is particularly
notable in Kant. hus Kant wants to secure a territory for the exercise of our faculties of cognition and defend it against the invasion
of the sceptics , 'a kind of nomads who abhor all permanent cultiva
tion of the soil' (CP R A ix). Kant also distances himself fron1 the
battlefield' of n1etaphysics: on this battlefield no combatant has
ever gained the least bit of ground , nor has any been able to base
any lasting possession on is victory' (CPR B xv). Kant relegates
the acquisition of knowledge and truth to a specific legislating
faculty (the faculty of understanding) , which prescribes to each of
the remaining faculties a well-defined function in conformity with
the overall interest. Deleuze calls this harn10nious collaboration
of the faculties under the legislation of the understanding the logical
'common sense\
21
According to Kant , however , the greatest threat to the legitirrlate

llse of Ollr faclllties comes not from external forces bllt frorn reason
itself. This is becallse reason finds within itself Ideas which cannot be
applied to anything in ollr experience , bllt which reason is tempted
to llse for the determination of metaphysical objects (the sOlll , the
world and God) in order to satisfy our metaphysical need' (CPR B
21). herefore reason entices the understanding 1'0 an illegitimate
llse of its a priori concepts and thllS gives rise to transcendental
illllsions. According to Kant , internal illllsions are natural and unavoidable (CPR A 298/B 354-5) , bllt their illusionary effect could be
defused through a critiqlle of reason itself. This n1eans that reason
takes over the chairmanship within the project of an imlnanent
critiqlle. It is the site of a trlle court of justice' (CPR A 7511B 779);
it is simllltaneollsly both the judge and the accllsed. Kant's language
invoking the court' , the jlldge' , legislation' and laws' explici t1 y
points to the moral character inherent in the dogmatic In1age of
though t. According to Deleuze , the entire doctrine of j lldgen1ent,
which implies the institution of a tribllnal and the clain1 to power , i
the power to judge , 'l11erges [se confond] with the psychology of the
priest' (CC 127/159). A moral or theological fon11 is inscribed even
in j udgen1ents of knowledge , inasn111ch as their llniversal predicates
lay claim to validity at all times and places. ]udgel11ents of knowledge enter into a relation with infinity and eternity , as they consider
the existing reality in the nal11e of higher concepts and vallles (sub
specie aeternitatis , so to speak).4 This is why Kant never actllally
invented 'a trlle critique of judgrnent' (CC 126 /1 58). For Kant , the
exercise of the cognitive faculties is subject to universal laws , which
should guarantee their proper llse' and prevent the faculties from
transgressing their limits and evolving freely. Variance , deviation ,
transgression - all these processes which are vital for creativity and
novelty in thought are judged and punished since they undermine
the privileged identities of concepts. The logical common sense only
produces an orthodox thought , which conrms iVhat already exists ,
i.e. prejudged identities , prevalent opinions (doxa) and established
values. 'The form of recognition has never sanctioned anything but
the recognisable and the recog
22

Kant claims to be beholden to the requirements of truth and reason; but
beneath these requirements of reason are forces that aren't so reasonable
at all: the state , religion , all the current values. 5
he
conse 1' vative and mo 1'al cha 1'acte 1' of the dogmatic Image of
thought can be demonst 1'ated not only with 1'ega 1'd to Kant , but also
with 1'ega 1'd to othe 1' state philosophe1' s' f1' om Plato to Hege l. Against
this backg1' ound , Deleuze calls fo 1' a mo 1'e 1'adical c 1'itique than that
of Kant and a new image of thought - 0 1' 1' athe 1', a libe 1'ation of
thought f1'om those images which imp 1' ison it' (DR xvi-xvii).
Philosophy is inseparable from :ritique'. Only , there are two ways of
going about it. On the one hand , you criticize 'false applications': false
morality , false knowledge , false religions , etc. This is how Kant , for
instance , thinks of his famous Critique': ideal knowledge , true morality , and faith come out perfectly intact. On the other hand , you have this
other family of philosophers who subject true morality , true faith , and
ideal knowledge to comprehensive criticism , in the pursuit of something
else , as a function of a new image of though t. As long as we're content
with criticizing the false we're not bothering anyone (true critique is the
criticism of true forms , not false contents. You don't criticize capitalism
or imperialism by denouncing their mistakes'). (DI 138/191)6
A 1'adical c 1'itique , acco 1'ding to Deleuze , does not simply affect the
false contents of thought i.e. erro 1' s 0 1' misguided opln lO ns. The
whole 'fo 1' m' of thought has to be ove 1'th 1' own , which ce1' tainly
implies a ve 1' y distu 1' bing and unsetding p 1' ocess. Deleuze derrlands
that the philosophical logos has to be 1'eplaced by a 'nomad nomos'
(DR 36/54); the 'sedenta 1'Y' and xed dete 1' minations irnposed
on thought by 1'ep 1'esentation have to be dissolved in favou 1' of a
completely othe 1' dist 1'ibution which must be called nomadic , a
nornad nomo without p 1'ope 1' ty , enclosu 1'e 0 1' measu 1'e' (DR 36/54).
Howeve 1', Deleuze's nomads a 1'e not a people of sceptics. His philoso
phy is not a scepticislTI but an attempt fo 1' a 1'enewed t 1' anscendental
philosophy , which he himself narned t 1' anscendental empi 1'icism'
(DR 1441187 , 56/79-80).

ONHE NECESSITY OF THOUGHT
Like Kan Deleuze is interested in the necessity of thought , that is

in a thought that imposes itself and leaves no choice for the thinke r.
his necessity has comlTIonly been calld 1' uth the pa 1' adigm of
which has long been lTIathematics with its method of deduction.
23
Kant seeks to ground our knowledge of the exterior world on

principles that are the result of a completely a priori deduction.
o this end , he interiorises the relation between thought and the
exterior world and he does so by making the truth of our statements of knowledge formally depend on transcendental subjective
principles abstracted from our empirical acts of recognition (cf. D R
135/176 7). However , in this way the necessity of thought eludes
us , because we have done nothing but closed thought upon itself in
a fatal circularityJ Deleuze will not cease accusing Kant of having
traced the so-called transcendental structures from the empirical
acts of a psychological consciousness. He particularly points to
Kant's three syntheses of apprehension , reproduction and recognition in the ranscendental Deduction of the pure concepts of the
understanding of the first edition of the Critique of Pure Reason.
Contrary to Heidegger and Hegel Deleuze brings the third synthesis of recognition into focus , in which the manifold of intuition is
subsumed under the concept of the identity of the object. Kant finds
that our thought of the relation of all cognition to its object carries
something of necessity with it in short our cognitions 'must have
that unity that constitutes the concept of an object' (CPR A 104-5).
For Kant , the identity of the concept of an object as well as the
unity of consciousness (the 'transcendental unity of apperception')
are transcendental requirements. In Deleuze's view , these transcendental conditions are directly traced from the e111pirical conditions
of the synthesis of recognition: 'In order to hide this all too obvious
procedure , Kant suppressed this text in the second edition. Although
it is better hidden , the tracing 111ethod , with all its "psychologism
nevertheless subsists' (DR 1351177). he fatal circularity of this
philosophical proo f' consists in Ka 's attempt to ground the
necessity of thought on transcendental principles , which he had preriously abstracted from our empirical objective thought , that is from
n1ere doxa. 8 Kant's transcendental conditions only allow for possible experience , which is always already delineated and prejudged by
the a priori concepts of the understanding.
In fact , concepts only ever designate possibilities. They lack the claws
of absolute necessity in other words , of an original violence inflicted
upontho
24
TheD tic 111ge fThought
If Deleuze calls his philosophy transcendental empiricism' , then

he uses the term transcendental in an entirely modified meaning.
he necessity of thought cannot be encountered by closing thought
upon itself. Rather , thought must be opened up to the outside world
that has to be conceived as a true exterior. The terrn empiricism' in
Deleuze's transcendental empiricism' means restoring exteriority to
the world , as well as the concrete diversity and plenitude of the sensible. Whatever forces us to think con1es from this outside. It imposes
itself upon us and intrudes as involuntary thought. According to
Deleuze , our thought is not an innate faculty characterised by categorial spontaneity\Instead , thought is lnarked by a passivity or
affectivity , which brings it in direct contact with the outside. 9 Its
principle is thus not autonomy , but rather heteronomy.
But what precisely forces us to think? What produces the absolute
necessity of an act of thought or a passion to think? Already in his
early book on Proust, Deleuze states that it is the encounter with a
'sign' that forcibly engenders though t.
What forces llS to think is the sign. The sign is the object of an encollnter ,
bllt it is precisely the contingency of the encollnter that gllarantees the
necessity of what leads llS to think. The act of thinking does not proceed
from a simple natllral possibility; on the contrary, it is the only trlle creation. Creation is the genesis of the act of thinking within thOllght itsel f.
This genesis implicates something that does violence to thOllght , which
wrests it from its natllral stllpor and its merely abstract possibilities.
(PS 97/189-90)
A sign is material and intensive in nature. In Proust and Signs ,

Deleuze distinguishes different types of signs , all of which have a
more or less violent effect on the person sensitive to them. The violent
impression triggers an involuntary exercise of the faculties of cognition , no longer bound by the laws of common sense. Deleuze develops this theory of signs in Difference and Repetition , in particular by
elaborating its link to Kantian Ideas. Deleuze defines sensible signs
as the bearers of Ideas or problems (D R 140/182). In his appropriation of Kant , Ideas or problems become the transcendental , genetic
conditions of thought , which carry our faculties of cognition to their
'nth power' or transcendent exercise'. 'Transcendence' in this case
does not indicate a passage to a transcendent world , to n1etaphysics
but a confrontation with the lin1it , an openng to the outside world
and an immediate relation with exterior forces. The exteriority of
thought' is distinguished from the form of interiority of thought
25

which closes thought upon itself. 1o Deleuze aims to substitute for the
Kantian transcendental conditions of knowledge qua representation
transcendental , genetic conditions of the emergence of thought and
the production of the real (DR 154/200 , 170/221 ).11
AN ALTERNATIVE IMAGE OF THOUGHT?
It is important to note that the term 'iInage of thought' does not
necessarily connote sOlnething noxious or detrimental to thought.
It first of all means that philosophy always relates to some kind of
pre-philosophical or non philosophical presuppositions. In his book
Nietzsche and Philosophy , Deleuze spells out that thinking always
depends on forces that take hold on thought.
he
theory of thought depends on a typology of forces. And once again

a typology begins with a topology. Thinking depends on certain coordinates. We have the truths that we deserve depending on the place we are
carrying our existence to , the hour we watch over and the element that we
frequen t. [. . .] Every truth is truth of an element, of a time and a place.
(NP 1101125 , translation modied D. V.)
This means that we have to take the lives of thinkers into consideration , or rather - in order to get rid of 'personalist' references (NP
xi) - the socio-historical milieu , the composition of forces , in which
thought is born. We have to ask: 1ho speaks 1here 'hen?'
6 ow?' 'In which case?' 'Fr Oln what point of view?'
Traditionally , the legitimate philosophical question has been
Plato's question: 'hat is . . .?' (ti estin?) , which asks for the essence
of things , that is for what is valid at all times and places , for instance
1hat is justice?' What is courage?' What is beautyDeleuze suggests other forn1s of question which relate thought to the psychoso
cial and historical conditions taking hold of the thinker hic et nunc.
A thought that pretends to ascend to the Idea or essence of things
by following the What is?' question in dialogue with others and
subrrlitting only to the requiren1ents of reason , presupposes precisely
a dogn1atic Irnage of thought. According to this Image , thought has
a natural affinity to truth. One aSSU111eS that thought either possesses
truth in tern1S of innate Ideas or formally contains truth in tern1S
of a priori concepts and Ideas of reason that serve as guiding rules.
One believes that if thought is based on a n1ethod and regulated
by self-imposed rules of reason it cannot go wrong. It is of a good
and upright l1 ature. Any deception that can befall man is due to
26
The Dogmatic In1age of Thought
forces external to reason; they come fron1 outside , for instance from
the body , the senses or our passions , or they result from a poorly
educated faculty of judgement. According to the dogmatic Image of
thought , all these forces that divert thought from its proper use count
as error\Error is the negative of thought , but since error amounts
only to an empirical fact , it can easily be averted. One has only to
submit oneself to a methodological discipline , that is to the guidance
of reason and understanding. In the case of an earnest commitment
to reason and understanding , that is under the condition of the good
will of the thinker , thought will naturally follow the path to truth.
In his book on Nietzsche and also in his book on Proust, Deleuze
opposes this dogmatic Image of thought , which has its roots in the
ancient Greek luage of the philosopher as the friend or lover of
wisdom , with a new image of thought. he general features of this
new image can be sumluarised in several interrelated postulates: (1)
The philosopher is not the friend or lover of wisdom , but rather
the jealous lover , the lover to be found in Proust's In Search of
Lost Time , who receives the signs emitted by his beloved and who
senses the unknown , dark regions in which dwell the effective forces
that act upon thought and force him to create (cf. PS 97-8/190).
Philosophy n1eans an apprenticeship to signs:One must be endowed
for the signs , ready to encounter thelu , one n1ust open oneself to their
violence' (PS 10 1/194). (2) There is no natural affinity of thought to
truth that could actively be played out by a truthful thinker endowed
with a good wil l. On the contrary , thought is characterised by its
capa;ity for being affected , that is its receptivity to sensible signs
or intensive forces that c Olumunicate a violence and wrest thought
from its natural stupo r. (3) The act of thought is not determined by
luethods which would allow us to follow pre-existent lines of thought
a roadrnap of reason. Rather , the genesis of the act of thought
occurs as an involuntary adventure , under the conditions of chance'
and 'constraint' (PS 16/23). The thinker has to undergo a process of
formation of thought , a violent training ( culture' or 'paidei') which
brings the whole unconscious of the thinker into play (NP 108/124).
(4) What the thinker needs to extract from the encounter with signs
is the pro lem or Idea that it poses to the thinke r. Philosophy thus
I
27
Finally , the philosopher has an essential relation to tirne: he is a

thinker of the future' attracting forces of the future , forces yet to
COITle (NP xiii). He is always untimely in the sense that he acts against
the reactive forces of the present world , which are a hindrance to life
and its active plastic force , that is the force of metamorphoses and
transfonnations. 12 This implies that the philosopher subject cannot
be identified with a substanti wellconstituted and identical subject
(the Cartesian Cogito) unaffected by the condition of time.
This list of postulates is not supposed to be exhaustive. It is lneant
to give a first Inpression of the new image of thought which Deleuze
proposes and which points in the direction of a genetic , problematic
alternative that calls for apprenticeship. We now need to address an
inevitable objection that arises in connection with Deleuze's claim
in Difference and Repetition that he seeks a thought without image.
ln this book Deleuze explicitly sets himself the task of discovering
thought without Image , even at the cost of the greatest destructions and the greatest delnoralisations' (DR 132/1 73). He compares
this revolution in thought to that already accomplished by nonrepresentative art: he theory of thought is like painting: it needs
that revolution which took art fron1 representation to abstraction.
his is the ain1 of a theory of thought without image' (DR 276/354).13
How can we make sense of this paradox? Has Deleuze simply
changed his lnind between writing the books on Nietzsche and
Proust and writing Dirence and Repetition? This seems improbable given that in Logic of Sense , Deleuze insists that thought necessarily presupposes a non-philosophic intuitive ground on which it
develops (LS 127/1 52). Again , roughly twenty years later in What Is
Philosophy? , Deleuze COlnes back to the idea of a relation between
thought and a non-philosophical iITlage of thought that is its presupposition. Here Deleuze even considers whether it is not the case
that every great philosopher draws up a new image of thought (WP
5 1/52) , or even whether a single philosopher cannot sustain several
philosophies or lnages of thought during his life. Jean-Clet Martin
indeed interprets Deleuze to be now claiming that a new image of
thought emerges whenever thought encounters a problem and is
forced to create concepts and deterrnine a case of solution. 14 This
confu
28
'h e
Dogmatic Image o{ Thought
demand for a thought without image is due to the fact that in

Di{{erence and Repetition Deleuze rst and foremost challenges the
particular dogmatic Irnage of thought. his does not rnean , however ,
that Deleuze renounces every presupposition of thought. On the
contrary , even in Di{{erence and Repetition , Deleuze stipulates
transcendental conditions of thought , which , as one could argue ,
already form a new image of thought. Yet , in spite of the demand for
a thought without image , Deleuze does not fall prey to an internal
contradiction. This is because this new image of thought is precisely
an in1age of a thought without image' , since it compels the philosopher to engender thought always anew , and thus commits him to a
philosophical commencement and recommencement' (DR 167/217).
By determining the transcendental ot larger' than what it conditions , i.e. as not consisting of abstract universal principles but
of principles that are 'plastic' and changing, Deleuze obliges each
philosopher to determine for hin1self the transcendental conditions of
the problen1 he deals with and to generate concepts in relation to the
problem and its conditions. Every treatment of a problem and every
act of creation of concepts thus require the erection of a new image of
though that is a plane of composition , on which the problern and the
concepts are laid out. Hence this in1age of a thought without image'
is an image , which does not sirnply presuppose itself and the natural
exercise of the faculty of thought (cf. DR 1391182) , but which forces
the philosopher to engender the act of thinking within thought.
Therefore , despite a certain tension on a superficial level , we do
not find a contradiction between the demand for a thought without
image in Di{{erence and Repetition and the frequent appeals to a new
image of thought in Deleuze's other works .1 5 Rather , Difference and
Repetition functions to prepare the way for those later texts in which
Deleuze once again directly addresses the question of a new image
of thought. his is also why Deleuze can consistently declare in the
Preface to the English edition of Difference and Repetition:
It is therefore the third chapter ['The Image of Thought'] which now
seems to me the most necessary and the most concrete , and which
serves to introduce subsequent books up to and including the research
undertaken with Guattari where we invoked a vegetal model of thought:
the rhizome in opposition to the tree , a rhizome-thought instead of an
arborescent though t. (DR xvii)
The rhizon1e and the tree are two conceptions that Deleuze often
uses in order to indicate two very different ways of thinking. 16 lhile
29
the rhizome grows frorn the rniddle and proliferates in an endless

number of ways , the tree is rooted in a primary unity orn there
drawing lines which go from one point to another lines which
might bifurcate but only according to the principles of a binary logic.
While the rhizome operates in accordance with the conj unction 'and
. . . and . . . and' (et . . . et . . . et) , the model of the tree is based on the
verb 'to be' (tre) , attempting to implant the logic of representation
in being. While the vegetal rnodel of thought or rhizoIne implies variations of the axes and orientations along wich thought develops ,
the model of the tree subordinates being to an exhaustive set of fixed
categories. While the model of the rhizorne always constructs decentred , pluralist systems , the rnodel of the tree erects its system around
a centre , that is a central unity (even if this unity is synthetic) , and
builds from there a hierarchy of terrns and pre-established relations.
The vegetal model of thought or rhizome is a product of the combined forces of Deleuze and Guattari , and in A Thousand Plateaus
they not only explicate it but also put it into practice. In effect,
the problem of the image of thought has been a rnajor concern of
Deleuze throughout his work , including the books he co-authored
with Guattari. 17 However , the struggle against the particular dogmatic Image of thought is most clearly laid out in Dierence and
Repetition. It is in the third chapter that Deleuze explains in every
detail what is at stake in the assumption of the dogrnatic Image.
While in Nietzsche and Philosophy , Deleuze sumrnarises the dogn1atic Image of thought in three essential theses , in Dierence and
Repetition he lists eight postulates. In the remaining part of this
chapter , we will go through each of these eight postulates in turn ,
and in so doing deal with some further important issues that could
not be covered in this introductory section. Thus we will have before
us a comprehensive account of what is at stal in Deleuze's criticism
of the dogmatic Image of thought when we turn to the exarnination
of the precise nature of Deleuze's notion of the transcendental in
subsequent chapters.
Postulates of the Doglnatic Jmage of Thought

COGITATIO NATURA UNIVERSALIS
he
first postulate which Deleuze identifies in his cr t icsmof the dogmatic Image of thought is what e calls Cogtatio natura universalis ,
that is the assumption of a universally distributed natural capacity
30
The Dogmatic Image ol Thought

for thinking , together with a good Nill on the part of the rhinker
and an llpright nature of thought. lt is assumed that in order to start
thinking it suf6ces to exercise one's natural capacity , which is good
in itself, and ro be guided by a thirst for knowledge and truth. As
such , even the simple man' can attain higher truths on the premise
that he scrutinises the grounds of whatever knowledge he believes he
has. Otherwise any knowledge j ust counts as opinion , that is mere
doxa. A very illuminating text in this respect is Descartes' dialogue
The Search Aer Truth , in which he opposes the character Eudoxos ,
a simple , uneducated man who dislikes learning from books and
rather trusts in his own natural power of thought , to Epistemon , an
educated man , who draws all his knowledge from books. Eudoxos
practises the method of universal doubt , which Descartes has
become famous fo r. The 6rst step is to discard all prej udices and
inherited propositions of knowledge. Thought is to be released from
its content, that is from the n1esh of concepts that depend on one
anothe r. What needs to be found is one originary concept , from
which all other concepts can be derived. For Descartes , this is the
Cogito , which consists of the three components '1 doubt' , '1 think'
and 1 aln'. Yet , in order to arrive at this ultimate truth , Descartes
has to appeal to an intuitive certainty of one's own existence , that
is to a pre-philosophical or non-philosophical experience of one's
empirical and concrete being. Hence he presupposes that everybody
knows what it means to think , what it n1eans to say '1' and how it
iels to exist. According to Delellze 'everybody knows , no one can
deny , is the form of representation and the discourse of the representative' (DR 1301170). he one endowed with a Cogitatio natura
universalis is elevated to being the representative of rational discourse. Outside this discourse of representatives there is no audible
or comprehensible speech , only noise'. This implies that the person
who denies knowing what everybody knows , who fails to think
and talk reasonably in line with rational discourse , and who poses
disturbing questions challenging the current ralue/belief systern , is
not represented in the discourse of representatives. Instead , such a
person of ill will is excluded from representation.he discourse of
the representatives casts out all of those who do not confonn to the
way everybody thi
31
of representation. In other words , someone who is not integrated ,

'who neither allows himself to be represented nor wishes to represent
anything' (DR 130/171) , carries the potential for a revolutionary
struggle , a struggle for liberation:
The forces of repression always need a Self that can be assigned , they
need determinate individuals on which to exercise their power. When we
become the least bit fluid , when we slip away from the assignable Self,
when there is no longer any person on whom God can exercise his power
or by whom He can be replaced , the police lose it. (DI 138/191)1 )3
Thus , for Deleuze , it is not a particular class of society or an exclu
sive , revolutionary elite that produce disorganising or deterritorialising' effects. On the contrary , Deleuze cites the 'young people today'
that practise 'many and various forrns of non-integration , the differ
ent forms of refusal' who are perhaps the bearer of a rel01utionary
struggle (DI 138/191).19 In any case , it is the excluded people who
manifest best the necessary luodesty and ill wi not managing to
know what everybody knows and denying what everybody is supposed to recognise. Deleuze believes that the presupposition of a
Cogitatio natura universalis is a hindrance not only to philosophy
but also to political thought.
One might aSS U1ue that the philosopher is best quali ed to resist
the presupposition of a Cogitatio natura universalis. Is he not the
one who proceeds with the greatest disinterest' , in pursuit of the
pure fonu of knowledge and truth? Is e not the one who refuses to
affirn1 the particular contents of what everybody knows (doxa)? But
in fact the philosopher still holds rm to the orm of representation
or recognition in genera l' (DR 13 1/171): the fonu of what is generally recognised and represented as universally valid. However , such
a thought will only lead to an affirmation of the represented order
and values , and exclude the elements which can initiate a radical new
beginning.
'COMMON SENSE' AND 'GOOD SENSE'
The second postulate can be summarised as the assurrlption of

cOlumon sense and good sense. Deleuze uses 'co l1lluon sense' and
'good sense' as specific technical terrns. By 'common sense' he does
not sirnply mean a con1mon hun1an understanding confirrrl ng a set
of opinions that people have in common and agree upon. Instead ,
Deleuze uses the term 'c Olurnon sense' (0 1' sensus conununis) in the
32

way Kant defines it in the Critique of the Power ofJudgment (CJ ~20
and ~40).
1n the Aristotelian tradition CO lTlmOn sense' is defined as an inner
mental faclllty that recognises in the given manifold of sensory data ,
sllpplied by the external senses , the identity of one common object.
Kant differs from this tradition in that he does not assllme one
inner faclllty or special sense' bllt rather a communal sense , i.e. , a
faclllty for j lldging that in its reflection takes accollnt (a priori) of
everyone's else's way of representing in thOllght' (C} 5: 293). his
con1mllnal sense is the effect of an a priori accord of Ollr cognitive
faclllties (imagination , llnderstanding and reason) , which mllst be
presllpposed as the sllbjective condition for all 'commllnicability'.
That is to say , this a priori accord of Ollr faclllties makes experiential
knowledge , the representation of the lnoral law or the feeling of
pleasllre universal and commllnicable in principle. here are at least
three different forms of common sense or ifferent proportions' (C]
5: 238) of the faclllties depending on the interest of reason they serve.
These are respectively: the logical common sense which is directed
toward the recognition of objects , the mOl 1 common sense which
a lns at the representation of a pllre form , i.e. the morallaw , and the
aesthetic common sense which isthe effect of the free play of Ollr
cognitive faclllties' (C] 5: 238). In each case , the specific proportion'
or the relations that pertain between the faclllties are of a different
kind. For example , in the logical common sense the llnderstanding
is the legislating faclllty , which provides pure concepts or categories
for determining the general form of objectivity. It forces the faclllty
of imagination to conform its empirical synthesis to the intellectual
synthesis of the forrnalllnspecified object in genera l. Reason , on the
other hand , thinks the formal completeness or totality of the pure
concepts of the understanding throllgh transcendent 1deas , and posits
a focal point or focus imaginarius olltside experience , which prOCllres
for the llse of the pllre concepts 'the greatest llnity alongside the
greatest extension' (CPR A 644/B 672). 511Ch is according to Kant ,
the relation between our active cognitive faclllties in the 'logical
co unon sense'.
Delellze's llse of the terrn 'good sense' refers back to Descartes who
defined it as a capacity for thOllght , which he co
33

piece of wax despite a11 the SenSllOllS alterations it has llndergone and
a11 the imagined changes of its shape and magnitllde. Now it n1ight
seem that cornmon sense' and good sense' mean the same thing: a
collaboration of faclllties with the ain1 to relate their data to one and
the sarne object. Bllt Delellze distingllishes them in that 'comrrlon
sense' gllarantees the transcendentalllnity of an llnspecied object in
general , while good sense' accollnts for the empiricalllnity of this or
that particlllar determined object.
For while common sense is the norm of identity from the point of view
of the pure Self and the form of the unspecified object which corresponds
to it , good sense is the norm of distribution from the point of view of
the empirical selves and the objects qualied as this or that kind of thing
(which is why it is considered to be universally distributed). Good sense
determines the contribution of the faculties in each case , while common
sense contributes the form of the Same. (DR 133 -4/1 75)
Cornl11on sense and good sense complement one another: 'together
they constitllte the two halves of the doxa' (DR 134/1 7S) that is
the transcendental form and empirical llnity of the objects of Ollr
experiential propositions. Althollgh it is trlle that philosophers have
always opposed particlllar doxa Delellze insists that they nevertheless always n1aintained the forn1 of the doxa. hey held firn1 to the
concept of identity , that is the formal identity of the sllbject and the
object, and they invoked the idea of harmony that is deployed in
the co11aboration of the faclllties in the lnodel of recognition. That is
to say , for the recognition of an object the faclllties are sllpposed to
enter into a harmoniolls relationship , to fonn harrnoniolls proportions. Delellze critcises Kant for not having offered in his first two
Critiques any explanation for the way in which this harmoniolls a
priori accord of faclllties arises. Kant rather introdllces comrllon
sense' like a sllpreme nalist and theological principle' (KCP 20/3S).
With respect to the harmony between the two faclllties , sensibility and llnderstanding , in an experiential cognition , Kant says he
can provide as to their origin no fllrther grollnd than Ollr divine
creator' .21
THE 10DEL
OF RECOGNITION
The rnodel of recognition is the third postlllate of thdogrnatic Image

of thOllght. he role of recognition has already been addressed in
the discllssion of the previolls two postlllates , which only testi es
34
to the interrelatedness of the postulates with one anothe r. Deleuze

nevertheless makes the n10del of recognition a postulate of its own ,
and highlights two irrlportant points of criticism.
First , he criticises the modelling of thought upon everyday acts of
recognition , as if the destiny of thought were at stake in these acts
and as if to recognise were to think (cf. DR 135/176). Philosophers
have chosen the lnost banal and puerile examples for demonstrating
the faculty of judgernent in our acts of recognition: this is a table' ,
this is a piece of wax' , Good morning , Theaetetus'. According
to Deleuze , Plato's aporetic dialogue Theaetetus is the first great
theory of common sense , of recognition , representation and error as
their correlate' (DR 149/1 94). From the very first , error is explained
by a confusion of the data supplied by the inferior senses. If, for
instance , 1 confusedly n1ap the present object of my sight onto
allother object of my memory - as in the case of Good lnorning ,
Theaetetus' when it is Theodorus who passes by 1 make a faulty
judgement.
Deleuze also raises a second objection against the model of recognition. It is not only the banality and insignificance of our daily acts
of recognition that discredit it as a model of thinking. The model of
recognition is not as harmless as it might seem. Recognition rearms
established values: conventional , social and moral values.
In so far as the practical nality of recognition lies in the established
values' , then on this model the whole image of thought as Cogitatio
natura bears witness to a disturbing complacency. (DR 1351177)
he
real danger of the n10del of recognition lies therefore in its

complacency , in its unquestioned support of established values and
institutions. Philosophers believed that on the basis of reason and
understanding one could establish a model of recognition valid for all
times and places , and did not see that even orm criteria of recognition evolved in a historical and socio-political context. By maintaining the form of recognition (i.e. the fonn of knowledge and truth) ,
philosophers have ensured that thought "rediscovers" the State ,
rediscovers "the Church" and rediscovers all the current ralues' (DR
136/177). However , with considerable debt to Nietzsche , Deleuze
defends a different and critical conception of philosophy:
We require a genesis of reason itself, and also a genesis of the understar
35
Kant's irnmanent critique of reason by reason itself does not satisfy

the demands of a radical critique. Kant denounces false claims to
knowledge and also a false morality but he leaves the domains of
knowledge , morality and religion intact. The higher interests of
reason (true knowledge , true morality , true religion) remain sacred.
herefore his critique is not as revolutionary as he maintains. We are
told that we are free when we obey the demands of reason , since it
is we who are giving the orders. In this way , reason persuades us to
continue being docile (NP 92/106). Kant simply installed the priest
and legislator within us (NP 9311 06) and thus the power of the
state and the church remain unchallenged. By contrast , a philosophy understood as a radical critique in the Nietzschean Deleuzian
sense questions the established powers and current values. In effect,
philosophy should upset the established order:
Philosophy does not serve the State or the Church , who have other
concerns. It serves no established power.he use of philosophy is to
sadden. A philosophy that saddens no one , that annoys no one , is not a
philosophy. (NP 106/120)
HE POSTULATE OF REPRESENTATION
he
fourth postulate , the postulate of representation is a key target

in Difference and Repetition. Deleuze retraces the model of representation through the philosophies of Plato , Aristotle , the neo-Platonists ,
Descartes , Leibniz , Kant and Hege l. In what follows we will focus
primarily on Deleuze's position with respect to Plato and Aristotle.
There are two reasons for this priority: Deleuze demonstrates that for
Plato the world of representation is essentially a moral vision of the
world' (DR 127/166). Later in history , this rnoral origin of the world
of representation will be more or less forgotten (DR 265/341) , but
since it is precisely the moral n10tivation that inspires Deleuze's critiqu his interpretation of Plato is of central in1portance. However ,
the Platonic COS1110S stilllacks the categories of representation , which
will be fully deployed only with Aristotle. Aristotle provides the clas
sificatory concepts according to which Being is distributed in deter
minable forms (genera of Being or categories) and further divided
up by fixed determinations (specic differen ces).he Aristotelian
philosophy of categories takes judgement as its operative too l.
Judgement fulfils both requirements of representation: distribution
which it ensures by the partiti n of concepts; and hierarchisation ,
36
which it ensures by the measuring of subjects' (DR 33/50). That is ,

the order of representation privileges those subjects which it considers endowed with a good sense' and thus better capable of exercising
judgernent than others. (For instance , in the days of Aristotle , these
were free adult men , aristocrats and citizens of the Greek polis'. )
The logic of representation and the function of judgement are both
complicit in the dogmatic Image of thought.
Let us begin with Deleuze's discussion of Plato. The Platonic
theory of Ideas introduces a distinction between the model and the
copy , between the Idea and the proper representations of the Idea.
The Idea is de ned as being identical with itself. That is to say , the
form of the Same applies to Ideas and to Ideas onl) because the Idea
of justice is nothing other than just, the Idea of courage nothing other
than courageous , and so on. The copies , on the other hand , are those
instances which claim to participate in the Idea , that is which claim
to have acquired the quality of being just, or being courageous , and
so on. According to Deleuze , the problem at the heart of Plato's philosophy is to find a method for distinguishing the true claimants from
the false pretenders. Plato's method of division is therefore a method
of selection. 22 he selected clairrlants or proper representations are
those which bear an internal resemblance to the Ideas , i.e. the ultimate things in therrlselves or models. This means that the true copies
are in the essential respects the same as the model.heir difference
is only secondary , or in other words: it is a mere conceptual difference a difference that remains subordinated to the prior concepts of
identity , resemblance and similitude. Apart from the true copies or
proper representations there are also those which fail the test of the
copy and the requirements of the mode l. These are the false pretenders or sirnulacra , which only simulate participation in the quality
of the Idea. They produce an external effect of resemblance (DR
128 /1 67) but in fact they are based on dissirnilitude , perversion and
disparateness. Their element is 'free , internal differences' which are
not subject to the structure of representation. That is to say , in the
world of simulacra , the San1e and the Similar do not pre-exist; the
concept of difference is not inscribed in a prior concept of identity.
Plato realises that simulacra pose a real threa
37
simulacra and seeks to banish or exorcise the latter. 23 As Deleuze

concludes: the will to eliminate simulacra or phantasrns has no
motivation apart from the moral' (DR 265 /3 41). The representative
theology' (DR 265/341) tolerates no one who threatens the affirmed
hierarchies of the model and the copy , the original Idea and the
derived representation the ground and the grounded. In sum , the
order of representation is characterised by the determination of hierarchies , fixed distributions and the exclusion of simulacra , that is the
subordination of difference to the Identical , the Salne or the Sirnilar.
According to Deleuze , philosophy or the theory of thought needs
a revolution , which willlead to the abandonn1ent of representation.
In fact , modern works of art have already completed that revolution
which took the tradition of realist or representational art to abstraction. 'Aesthetic modernity provides Deleuze with one example of a
world in which difference has free reign. He suggests that modernity is defined by the power of the simulacrum.'24 In overturning
Platonism , that is in liberating simulacra from the Platonic order of
representation , the Inposed hierarchy of the original and the copy,
the true and the false , or the good and the bad , Deleuze sees an
escape for the theory of thought. ut while there is still an escape
route in Plato , in Aristotle the order of representation is deployed
n1uch more rigorously.
With Aristotle , the logic of representation gains a new methodological approach and new conceptual tools. Aristotle rejects Plato's dialectic method of division , which operates by selection among rivals ,
by testing and authenticating the claims of participation in Ideas.
The method of division is a bad and illegitimate syllogism according
to Aristotle , since it lacks the required middle term (DR 59/83; LS
254/293). hat is to say , for determining the relation between the
determinable (Idea or concept in general) and determined objects
(individual claimants) , some rnediation is needed. In Aristotle ,
deterrnination proceeds through the mediating concept of species.
Aristotle's method is one of continuous specification from the highest
genera down to the lowest indivisible inmae species.
Deleuze's critique of the logic of representation applies to Aristotle
much more than to Plato. According to Deleuze , the logic of representation does not allow for an absolute concept of di erence of
difference in itself. In the Platonic cosrnos , there remains at least
the displaced realm of simulacra , where we find no prior identity ,
no internal resemblance , where difference has become the ultimate
38
The Dogmatic Image o /, Thought

element in which the identity of any concept in general is dissolved.
In Aristotle by contrast the conception of difference is only relative ,
which is to say that difference always operates in relation to the supposed identity of a concept. This can best be seen by the example of
sp:i fic differences': species are said to differ only within the identity
of the genus. Given the genus anirna l', for instance , we can distinguish 'terrestrial animal' and 'aquatic anin1al'.25 These are specific
differences in the form of contraries (not contradictions , since these
two detenninations do not necessarily include a logical contradiction
but they nevertheless divide the genus into different species through
their real opposition).26 Specific differences are predicates , though
predicates of a peculiar kind: the specific difference is attributed
to the species thereby constituting it, and at the same time specific
difference preserves that which it differentiates (DR 31147).27 his
rneans that the specific difference is inscribed within the identity of
a generic concept, in this case the genus anirnal'. For Aristotle , specific difference is maximal (megist) and perfect (teleios). It is of all
kinds of differences the best example. Generic difference , on the one
hand , is too large , as it holds between categories that have nothing in
common; dirence between individual substances or little species ,
on the other hand is too small , since the srnallest indivisible objects
or little species (atomon , adiaphoron 0 1' eidos) do not ente 1' into 1'elationships of contrariety. Their determination proceeds not so much
by the opposition of predicates (cont1' aries) but by the perception of
resemblance which presupposes continuity of sensible intuition in
the world of representation.
At rst sight it seems that generic differences , which hold between
categories , can be assin1ilated to specific differences because they
are also said in relation to a unique term. However , this unique
term is not a genus , that is a collective , explicit and distinct unity.
The categories are related to Being , which is a confused unity , since
it has no distinguishing marks , no content of its own. Rather than
being divided , Being is dis'ibuted to the terrns - that is the genera ,
the differentia of a genus , the sub-genera , the specic differences ,
the species , the individual differences and the individual substances
39
the fact that , although there are various senses of Being , they are
all related to a single central sense. Aristotle rnaintains a quaslidentity' of Being by judging Being in analogy with the identity of a
genus. Yet Aristotle explicitly states that Being is not a genus. When
reconstructed , the argument goes as follows:
1. If sOITlething is a genus , then it cannot attribute itself to its specific
differences [Aristotle , Topics , VI , 6 , 144a]
2. Being can be said of its differen t a (because we can say that generic
or categorial differences al just as we can say that specific
differences are).
3. Being is not a genus.
This argument , however , implicidy presupposes that the generic
or ca tegorial differences of Being are sin1ilar to specic differences ,
which ulti lTlately allows Aristotle to conclude that Being is not a
genus. According to Deleuze , Aristotle borrows an argument from
the nature of specific difference in order to conclude that 'generic
differences are of another nature' (DR 32/49). However one should
note that Aristotle does not actually conclude that generic differences
are of an utterly different nature than specific differences. It is rather
Deleuze who wants to argue for the different nature of generic differences. What argUITlent does he put forward to support his claim?
Obviously , he cannot argue that (1) if Being were a genus , its differences would be assin1ilable to specic differences (DR 34/51) , (2)
Being is not a genus; therefore generic differences are not assimilable
to specific differences. Deleuze seen1S to suggest this line of argument
(cf. DR 34/51) , but in any case , this argument would be invalid
(p q , -,p , therefore - q). hus Deleuze has to find other means to
argue for the different nature of generic and specific differences. l-Ie
makes the following distinction:
1. The pros hen relation of the categories 1' 0 Being is a r e1 ation of

interiority: it is on its own account that each [category] has unity
and being , by virtue of its own nature' (DR 309/50 , note 5/1; see
also DR 33/49).
2. A genus is determinable only by specific differences nl without
(DR 34/51; italics are mine , D. V.).
From (1) and (2) it follows that 1' he relation of generic differences
to Being differs fron1 the relation of specific differences to a genus.
Deleuze's intentions are clear. He wants to show that , in Aristotle ,
there are
40
'h e
Dogmatic Image of Thought
two Logoi' , differing in nature but intermingled with one another: the
logos of Species , [...] which rests upon the condition of the identity
or Ufllrocity of concepts in general taken as genera; and the logos of
Genera , [. . .] which is free of that condition and operates both in the
equivocity of Being and in the diversity of the most general concepts.
(DR 32-3/49)
Given these two Logoi , Deleuze wonders whether we must not recognise here a kind of fracture introduced into thougl one which
will not cease to widen in another atmosphere (non-Aristotelia'
(DR 33/49). In order to maintain the coherence of his philosophy of
representation , Aristorle bases it on relations of analogy: the rnodel
of representation requires that the identity of the concept of Being,
although it is only a confused and distributive unity , is analogous
to the identity of the genus. Deleuze concedes that Aristorle himself
did not speak of analogy with respect to Being (DR 308/50 , note
5/1) , but Deleuze finds support for his claim in the scholastics , who
translated the Aristotelian pros hen as analogy of proportionality'
(DR 309/S0 , note 5/1).
With the example of Aristorle , Deleuze attempts to show that neither
specific difference nor generic difference delivers an absolute concept
of difference. Instead difference is fully subjected to the requirements
of representation: it is subordinated to
identity , in the form of the undetermined concept [Being]; analogy , in the
relation between ultimate determinable concepts [genera or categories];
opposition , in the relation between determinations within concepts [specific differences as contrary predicates]; resemblance , in the determined
object [individual substance] of the concept itself. (DR 29/44-5)
Deleuze calls the quartet of identity , analogy , opposition and resemblance the necessarily quadripartite character of representation'
(DR 34-5/S2) , whereby representation is specified here as organlc
representation'. By organic representation' Deleuze refers to a
world of representation , in which Being is distributed and divided
up according to fixed and proportional determinations. Deleuze
con1pares this world to a 'sedentary space' (DR 36/S4) with limited
territories and defined properties. These sedentary structures of
representation correspond to our faculty of judgernent and its concon1itants of common sense and good sense. Another fundamental
feature of Aristorle's philosophy of representation is the privilege he
gives to specifc difference. Difference operates lnainly in the middle
41
regions of genus and species , that is equidistant from the boundaries

of the large and the small (DR 38/56). It rneasures and divides up the
average forms (DR 42/61) , while excluding the extreme. Differences
which are too large (generic differences) are made analogous with
specifc differences (despite their different nature) , while differences
which are too small (individual differences) are neglected as differences without a concept. Organic representation' is characterised by
keeping difference within the limits of the large and the small , and by
inscribing it within concepts in genera l.
However , Deleuze identifes a second type of representation ,
which he calls orgiastic' or infnite representation' (cf. DR 42/61).
With Leibniz on the one hand and Hegel on the other , representation
is rendered infinite: difference is drawn into an infnite movernent to
the 11l0ment at which difference either vanishes (Leibniz) or reaches
its absolute maximum in contradiction (Hegel). The two lirnits
of organic representation , i.e. the small and the large , are pushed
towards the infnite , the infnitely small or the infnitely large. The
introduction of the in nite into representation , Deleuze explains ,
renders the determined object independent of the genus as the ultimate determinable concept and of the species as the determination
within the concept, by relating it instead to a ground (DR 43/62-3).
This ground is c0 1l1pletely indeterrninate: differences remain unidentifable and the distinction of forms is dissolved. he ground in orglastic representation' equals 'a completely undifferenciated abyss , a
universal lack of difference , an indifferent black nothingness' (DR
276/354). In the case of Leibniz , the ground is the infnite continuity
of the universe , which is itself contained in fnite particular Selves
considered as essences (monads). In the case of Hegel , particular
Selves are only detern1inations of a pure Self, that is a Self' enveloped
by the in nite self-movelllent of a universal ground. In both cases ,
however , the deterrnined is referred to an infnite and indeterminate
ground , which must be said to engender fnite determinations or to
make difference. This means , though , that the concept of difference
has not become absolute but on the contrary is included in this irm
nite and indeterminate ground. he logic of in nite representation
depends on a s uf
While this foundation is not the identical itself, it s nevertheless a way of
taking the principle of identity particullrly seriously , giving it an in t nite
value and rendering it coextensive with the whole , and in this manner
allowing it to reign over existence itself. (DR 49/70)
42

innite re'esentation does not free
itselfjrom the principle of identity as a presupposition of representation.
(DR 49/70)
The point is that in the last resort
Deleuze rejects both types of representation , since both finite and

infinite representation (organic and orgiastic representation) suffer
from the same defect , that of subordinating difference to identity
(whether this is the identity of a common genus or the identity of an
infinite and founding principle).
THE CONCEPT OF ERROR
The fifth postulate of the dogmatic Image of thought is the postulate

of the negative , or of error. As we have seen , the dogmatic Image
supposes that thought is naturally oriented towards truth. It further
subjects thought to con1mon sense and good sense , as well as to
the model of recognition and the requirements of representation.
Is there anything that can divert thought from its proper path?
There is of course the possibility of errofor instance a false act of
recognition , which results frorn a failure of good sense. That is to
say , in the determination of a particular object the collaboration of
our inferior faculties rnight be confused , insofar as they erroneously
relate son1ething perceived and something remernbered to one and
the same object , although the contents should have been distributed
to different objects. However , in this case the form of the common
sense still remains integral and intact (DR 149/193). The error
amounts to a simple fact that can easily be corrected. Errors that we
cornmit due to a failure of good sense , absent-mindedness or a poorly
educated faculty of judgement cannot endanger the in-principle good
nature of thought. Philosophers have been aware that there are more
serious threats besides error. Plato's notion of ignorance or forgetting , the Stoic notion of stultitia (a hybrid of stupidity and madness) ,
Spinoza's notion of superstition , Kantian transcendental illusion or
Schopenhauer's notion of vulgarity - all these concepts testify to
other dangers: amnesia , delirium stupidity madness , malevolence
or baseness. Yet these dangers have n10stly been treated as external
forces , threatening thought fron1 outside. The effect of these forces
on thought is still regarded as error. 'Error is the infinite n10velnent
that gathers together the whole of the negative' (WP 52/53). Only
Kant realised that the real threat for thought COlnes from within ,
from pure reason itself. Transcendental illusion arises if reason is left
43
'in the state of nature' (CPR A 7511B 779): then reason makes assertions which go beyond the conditions of all possible experience and
claims their objective truth. However , the in-principle good nature of
reason is not called into question. Kant even goes as far as putting the
blame entirely on the faculty of j udgement s uch as to underline the
good nature of our faculty of thought. 29
Contrary to the view that thought possesses a good nature and an
a nity to truth , Deleuze argues that none such presupposition can
be made in principle. Moreover even a discourse , which is entirely
rnade up of truths , can be a symptom of stupidity and a base way of
thinking.
Stupidity is not error or a tissue of errors. here are imbecile thoughts ,
imbecile discourses , that are made up entirely of truths; but these truths
are base , they are those of a base , heavy and leaden sou l. (NP 1051120)
hus
stupidity and baseness express themselves even in true discourse. But what precisely does Deleuze mean by base truths? A
base truth is nothing but a plain and exact recognition , a proposition that is detached from its context of living thought and from
the problem it is supposed to answe r. For instance , in his book on
Hume , Empiricism and Subjectivity , Deleuze criticises the discourse
that accuses Hurne of a psychological aton1ism of perception:
'''Hume has pulverized the given." ut what does one think has
been explained by this? Does one believe sornething important has
been said?' (ES 105 6/118 19). A proposition only makes sense if it
is related to a problem or set of problerns that haunted the philosopher.he highest art of thought is to pose problems and to develop
these problems all the way down , to the very end , of their necessary
implications (cf. ES 106/1 19). It is at this point that a base way of
thinking' is separated from a 'noble way of thinking' , that we can distinguish stupid and irrelevant remarks from the critical and creative
activity of posing problems. However , even the posing of problems
can be threatened by stupidity , namely in the case of badly posed or
inexistent problems. 30
Teachers already know that errors or falsehoods are rarely found in
homework (except in those exercises where a fixed result must be produced , or propositions must be translated one by one). Rather , what is
more frequently found and worse - are nonsensical sentences , remarks
without interest or importance , banalities mistaken for profundities ,
ordinary points' confused with singular points badly posed or distorted
problems - all heavy with dangers yet the fate of us al l. (DR 153/198)
44
The Dogn1atic Image o{ Thought

According to Deleuze , stupidity and baseness are far more devastating than error can ever be. he danger that pervades thought from
inside is not error or falsehood , but stupidity. Stupid thoughts affect
us all; they are structures of thought as such' (DR 15 1/ 196). hey
are not simply external forces that act on thought accidently , such
as traits of character or an imbecile culture of society. Stupidity
and baseness belong to thought by right. This means that the relation of thought and stupidity poses a transcendental problem' (DR
15 1/196).
The important question is: what is the transcendental significance
of stupidity? Deleuze clairns that stupidity is not only the greatest
weakness of thought , but also the source of its highest power in that
which forces it to think' (DR 275/353). The reason for this is that he
sees stupidity and its allies first and foremost as {orces. Force is not
understood as a blind and brutal drive which seeks to destroy and
dominate. Forces are rather impersonal flows of energies or intensities , and their effects totally depend on the way they are compollnded
within a particlllar field. As sllch they can either inhibit and prevent
critical thOllght 0 on the contrary , increase the critical power. Take
for instance Delellze's example of the two types of idiots in philosophy
and literatllre: On the one hand , there is Descartes' idiot (Eudoxos)
who doubts particular oplnlns and prejlldices bllt maintains the
frm of rational discollrse; on the other hand , there is Dostoyevsky's
idiot , who tentatively calls the fonn of rational discollrse itself into
qllestion. This new Rllssian idiot rejects the 'trllths of reason' or the
trllths of history' and exercises a critical aggressivity 31
The old idiot wanted indubitable truths at which he could arrive by
himself: in the meantime he would doubt everything , even that 3 + 2 = 5;
he would doubt every truth of Nature. The new idiot has no wish for
indubitable truths; he will never be resigned' to the fact that 3 + 2 = 5
and v ills the absurd - this is not the same image of thought.he old idiot
wanted truth , but the new idiot wants to turn the absurd into the highest
power of thought - in other words , to create. he old idiot wanted to
be accountable only to reason , but the new idiot , closer to Job than to
Socrates , wants account to be taken of every victim of History' these
are not the same concepts. The new idiot will never accept the truths of
History. The old idiot wanted , by himself, to account for what was or
was not comprehensible , what was or was not rational , what was lost
or saved; but the new idiot wants the lost , the incomprehensible , and the
absurd to be restored to him. This is most certainly not the same persona;
a mutation has taken place. And yet a slender thread links the two idiots
45
as if the first had to lose reason so that the second rediscovers what the
other , in winning it , had lost in advance: Descartes goes mad in Russia?
(WP62 3/61 )
In this quotation on the old and the new idiot (Descartes' idiot and
Dostoyevsky's idiot) , Deleuze distinguishes two different conceptual
personae (personnages conceptuels) , two different Inages of thOllght
and two divergent series of concepts. A conceptllal persona , for
Deleuze , is the subject of enunciation of a philosophy; it is the one
who speaks and invents concepts once the philosopher-subject ceased
to say '1' and turned into his conceptual persona. This is not to say
that the conceptllal persona functioned as a representative of the philosophe r. On the contrary , the philosopher literally becomes his conceptual persona( e). hus is the case with Nietzsche who at the end
of his life even signed his letters with Dionysus' or the Crucified'.
But Deleuze cites many more exarnples of conceptual personae sllch
as Plato's Socrates , Kant's figure of the judge or , precisely , Descartes'
idiot. 32
Descartes' idiot (Elldoxos) is the 'private thinker' , in opposition
to the public teacher or learned man (Epistemon) who operates with
scholastic definitions and established truths. he private thinker
rejects all knowledge that is passed on to him , pretending to know
nothing , and strives to think for himself by rneans of strict methods
procured by reason. It is the private thinker , i.e. Descartes' idiot ,
who launches the Cogito as a rst principle. However , the Cartesian
Cogito does not yet think: 'it only has the possibility of thinking ,
and remains stupid at the heart of that possibility' (DR 276/354).
What it lacks is the necessity of thought , the contact with the exterior , since it is closed upon itself in an indefinite regress: 1 think that
1 think that 1 think . . .33 As Deleuze says , 'thought remains stupid
so long as nothing forces it to think' (DR 275/353). Dostoyevsky's
idiot is a new type of private thinker , though private thinker' is not
a satisfactory expression , because it exaggerates interiority , when it
is a qllestion of outside thought'. 34 This new idiot rejects n1ethod'
and the Cogitatio natura universalis; he constantly produces 'counterthoughts' (contre-penses) questioning the importance of abstract ,
rational truths and the reign of so-called 'truths of history' interpreted by the power of an established orde r. 35 The new idiot wants
the absurd , the lost and the incomprehensible to be restored to hin1'
(WP 63/61) - is this not the quest for a iVorld of simulacra
46

It seems that the new idiot indeed relates to another image of
thought , which differs frorn the dogmatic Image. The 'slender thread'
(WP 63/61) , which links the old and the new idiot , is the programme
of a 'private thinker' launching himself into critical thought. But the
critical thought of Descartes' idiot remains captured in the dogmatic
Image , whereas Dostoyevsky's idiot is pushed to a more radical
critique , turning the absurd into the highest power of thought' (WP
62/61) and thereby creating new values.
Coming back to the question of the transcendental nature of stupidity , we see now that stupidity can be taken in two senses. On the one
hand , it amounts to the greatest weakness of thought when it entangles us in a perpetual confusion with regard to the important and the
unimportant , the ordinary and the singular' (DR 190/24S). We take
banalities , simple acts of recognition and opinions for truths , and are
incapable of evaluating the truths and values that are proclairned by
the powers of an established orde r. On the other hand , stupidity as
the highest power of thought is the refusal of knowing what everybody knows. In this sense it becomes a transcendental condition for
critical thought. he conceptual persona of the idiot is a thinker , who
has no wish to know what everybody knows , who struggles against
opmlon (doxa) (Descartes' idiot) and even against the rational
form of discourse (Dostoyevsky's idiot). With Dostoyevsky's idiot ,
n1adness or deliriun1 becomes a critical forc which precisely turns
against stupidity in the first sense , i.e. the reign of base truths and
petty ralues. Perhaps one could say that stupidity in the first sense
also has a transcendental significance , inasmuch as it incites critical
thought:
Is it not also the existence of stupidity which forces it to think , precisely
the fact that it does not think so long as nothing forces it to do so? Recall
Heidegger's statement: 'What gives us most cause for thought is the fact
that we do not yet think.' (DR 275 /3 53)36
Stupidity or Heidegger's 'idle chatter the thoughts of what 'they'
(nlan) think provoke the acts of the philosophical idiot , i.e. of critical
thought.
THE POSTULATE OF DESIGNATION
the sixth postulate , the postulate of designation , Deleuze

challenges both the traditional conceptions of sense and truth and
Tith
47
the relation that has been identified between them. In the light of
Frege's account of sense and truth , it becornes apparent how we
commonly distinguish sense and truth as two separate dimensions of
the proposition where sense subsists only in a formal relation to the
othe r. By contrast, Kant's account of sense works quite differently
insofar as sense becomes the superior transcendental condition of
possibility of truth. However , Deleuze argues that both accounts
define sense as merely the form of possibility of truth and as such
sense remains an extrinsic condition , not capable of founding truth
in terms of an intrinsic genesis. Deleuze demands a new conception
of sense which is distinct from that of Frege and Kant. According to
Deleuze , sense has to be conceived as both the e ct and the intrinsic
genetic elenlent of an extra-propositional , differential structure (the
problem). His point of view has affinities with the structuralist tradition but it is certainly not the same. Let us first begin with Frege's
account of sense and truth.
In his fanlous essay 'On Sense and Reference' (1892) , Gottlob
Frege introduces the following phraseology: 'A proper name (word ,
sign , sign combination , expression) expresses its sense , stands for or
designates its reference.'37 He goes on to examine what the sense of
a sign and what its reference is. According to hIn , the reference of
a sign is the definite object that the sign designate while the sense
of a sign contains the 'nlO de of presentation' of the object. There
can be different modes of presentation for the same object. As such
there are also di erent signs for the san1e object , that is signs with
different senses but with the same reference. Frege's most famous
example is that of the morning star' and the 'evening star' , which
are two signs with different senses yet referring to the same planet
Venus. Now , Frege treats propositions in a sin1ilar way. Propositions
have two dirnensions , the dimension of expression and the dimension of designation or reference. The sense that a proposition
expresses is the thought contained in the proposition. Thought' is
understood as the objective content , which is capable of being the
common property of several thinkers' , no matter what language
they speak and regardless of the desires and beliefs they individually
connect with this thought. 38 The reference of a proposition is generally deterrnined by inquiring after the r
48
The Dogmatic Image
o/" Thought
example: Odysseus was set ashore at It haca while sound asleep.'39

Poetry or fiction provides plenty of these examples , that is propositions that contain proper names without a reference. However , as
Frege states , it is only in works of art that we are satisfied with the
dimension of sense and do not seek to advance to the q uestion of
reference , but in other cases we press on: It is the striving for truth
that drives us always to advance from the sense to the reference.
he dimension of reference is the locus where the question of the
truth and falsity of a proposition , that is its truth-value , is decided.
This is why Frege says that the reference of a proposition , if it has
one , is either the True or the False' .41 Given this linkage between
the truth-value and the reference of a proposition , there results a
strangely detached relation between sense and the question of truth
and falsity. As Frege remarks: The truth-value of a sentence lUUSt
remain unchanged when a part of the sentence is replaced by an
expression haring the sarrle reference 42 That is to say , although
the sense of a proposition luight be altered by the substitution of
one expression by another , the truth-ralue of the whole proposition remains the sarrle , provided that the substituted expression has
the same reference as the one it replaces. In other words , truth and
lsity remain to some extent unaffected by sense. On the other hand
a proposition with no sense cannot be true and in this respect the
requirement of sense is a necessary condition for a proposition to be
true. Sense is a condition of truth' insofar as it contains a number of
logical conditions , which define a gralumatically well-forn1ed expression. But sense is by no means a sufficient ground for truth: there
are propositions , which make perfect sense from the point of view of
their fonu of expression , but that are nonetheless false. An example
would be the proposition 'A decahedron is a regular geometrical
figure.' It is impossible to find a referent for the whole proposition
(decahedrons are non-regular) , but still the proposition contains a
thought that makes sense and could n1istakenly be regarded as true.
Thus , in grasping the sense of a proposition , one cannot be assured
of its truth-value. The question of truth and falsity is decided only
with regard to the d iIuension of reference , independently from the
din1ension of sense.
Deleuze's objection to these conceptions of sense a
49
The relation between the condition and the conditioned is too

loose because both terrns of the relation remain more or less independent from one another. lhile sense is indifferent to what it
conditions , the conditioned itself, that is truth , remains unaffected by
the condition , which is supposed to render it possible. Sense is only
the formal condition of possibility of truth (LS 18/29). That is to say ,
it can only determine the logical conditions under which a proposition would be true. It cannot exclude the case of false propositions
that rnake sense. As Deleuze says: sense retains an extension larger
than that which is conditioned , sense does not ground truth without
also allowing the possibility of error' (DR 153/199). Thus Frege's
conception of sense cannot materially account for the truth-value of
a proposition. In order to determine the truth-value of a proposition ,
we have to turn to the dirnension of reference. There is in fact no way
to pass from sense to reference , i.e. the truth-value. One can certainly
deduce frorn one proposition further propositions by means of grammatical transformation rules or semantic implications of concepts ,
distributed within the original proposition. For instance , given the
proposition that a person X is a widower , we can conclude from this
that X is a man , that X was rnarried , and that the wife of X has died.
However , in inferring these further propositions we remain on the
sarne level , the level of signification , and never cross over to the level
of denotation or truth. In effect, signification can never exercise its
role of last foundation , since it presupposes an irreducible denotation' (LS 18/29). Hence , although sense is discovered as a condition
of truth , it amounts only to a formal condition of possibility for a
proposition to be true.
he second part of Deleuze's objection actually follows from the
first. In making sense the condition of truth , an essential step was
made to establish a ground for a critique' of truth. I-Iowever , the
critical project is doomed to fail not only because the ground remains
larger than the grounded (first objection) , but also because the
ground is thought in the image of the grounded The condition
resenzbles the conditioned from the point of view of its logical form.
hat is to say , the conditioned or those propositions that we hold
to be true (for example , scientific propositions describing objective
states of affair
50

strategy becomes even rnore apparent in the case of Kant , which we
will now turn to.
As is well known , Kant draws the clue to the discovery of all
pure concepts of the understanding' (CPR A 76/B 102) from the
Aristotelian table of logical forms of judgement. He recognises the
importance of the logical forn1s as the forrnal condition of possibility
for a proposition to be true. Thus Kant clairns that the formal aspect
of all truth consists in agreement with the laws of the understanding'
(CPR A 294/B 350). This is to say that certain laws of the understanding , for example the principle of non-contradiction , rnust hold
for all cognitions posed in propositional form. In this way , general
logic provides a necessary formal criterion of truth. 44 However , Kant
goes much further than general logic can ever go. He explains that
general logic only considers the logical form of our cognitions and
their relation with one another , hence it abstracts frorn all content,
that is from any relation of our cognitions to the object (CPR A 55/B
79). 'ranscendentallogic by contrast , has to do with pure concepts
of the understanding insofar as they are related a priori to objects.
Therefore transcendental logic provides a further criterion of truth:
namely the requirernent that pure concepts can be constructed in the
formal intuition of space and time and us be related to an object;
for without the relation to a possible object of experience , a cognition will completely lose its content and hence all truth. 45 This means ,
for example , that from the point of view of transcendental logic the
proposition A decahedron is a regular geometrical figure' cannot be
said to be false , but rather it makes no sense. Since it is impossible
to construct a regular decahedron in space , the proposition cannot
be related to a possible object of experience , hence it has no content ,
that is no sense. While the proposition can be said to have a sense
from the point of view of general logic , transcendental logic dispels
this proposition altogether. The san1e happens with a proposition
like the following: 'The width of Naridson's house inside exceeds
the width of the house as measured from the outside by a quarter of
an inch. 6 From the point of view of formal logic , there is nothing
wrong with this proposition. However , we are dealing with a transcendental in1possi ility: a house the interior of which is greater than
its exterior is an i
51
determination of parts. Thus the proposition cannot be related to a

possible object of experience and therefore has no content , no sense.
The same kind of transcendental irnpossibility applies in cases where
son1ething appears without any preceding cause , for instance the
sudden apparition of an ange l. Such a thing would also be no possible object of experience. Hence in the Kantian transcendentallogic
nothing escapes the transcendental conditions of sense. Anything
that falls outside the transcendental scheme or structure , that is that
which cannot be constructed in space and time , and ordered according to a priori conceptual rules , is not a possible object for us. To put
it positively: whatever proposition we hold to be true , the referent of
this proposition has to satisfy the transcendental conditions of sense.
ln this way , sense is made a superior condition of truth. lndeed , one
could say that Kant's transcendental philosophy is a logic of sense:
he stipulates the transcendental conditions for something to have a
sense , that is to be an object for us.
Kant's invention of transcendentallogic has a revolutionary effect:
truth is no longer simply a matter of adequation with an external
state of affairs , as it is suggested from the point of view of an empirical consciousness. To put it differently , the Kantian invention of
the transcendental C01 adicts the simple assumption that the locus
of truth for a proposition is the dimension of designation or reference. The idea that (1) a proposition is true , if and only if what is
said or expressed applies to the designated object or state of affairs
it refers to , and that (2) a successful reference makes a proposition
true , while an unsuccessful reference makes a proposition false ,
grounds a conception of truth which presupposes the existence of a
reality exterior to sense. Kant's revolutionary move is to make truth
dependent on sense , which is to say that a true cognition necessarily
points beyond itself to an object or state of affairs that can no longer
be posited in reality exterior to sense. However , for rendering sense
a superior condition of truth , Kant has to pay a high price: narrrely
that of interiorising the relation between cognitions and the manner
in which they relate to objects or states of affairs. This means that
for Kant the outside world is not truly exterior: it remains relative to
the a priori conditions of the transcende1
52

only to the extent that thought itself provides the ground. As Kant
states ,we can cognize of things a priori only what we ourselves have
put into them' (CPR B xviii). This circularity in Kant's account is of
course intended. hought is led in a circle and closed on itself. The
question arises , however , as to whether this foundation that Kant
provides is really convincing and fulfls its purpose. 'One is perpetu
ally referred from the conditioned to the condition , and also from
the condition to the conditioned' (LS 19/30). This is to say that the
condition is nothing but the form of possibility of the conditioned ,
and the fonn of possibility is fabricated retroactively in the image
of the conditioned. Consequently, the general forrn inscribed in the
proposition of knowledge is irrefutable to the extent that it responds
exactly to the a priori conditions , which render the proposition possible. 47 In fact , Kant's revolutionary achieverrlent comes down to
having invented a new form of possibility , that is of replacing the
forrnal possibility with transcendental possibility (cf. LS 18/30). But
it remains that Kant's invention of the transcendental condition is
an odd procedure since it involves rising from the conditioned to the condition , in order to think of the condition as the simple possibility of the
conditioned. Here one rises to a foundation , but that which is founded
remains what it was , independently of the operation which founded it and
unaffected by it. (LS 18-19 /3 0)
Deleuze's twofold objection , as to the condition being larger than
the conditioned and being thought in the image of the conditioned ,
is rooted in a critique that concerns much more than a linguistic
analysis of sense and truth-ralues of a proposition. Deleuze is concerned with all those philosophers who claimed to have discovered a
true beginning in philosophy , a ground for truth and necessity , but
who in fact argued in a circle and closed thought on tself. hought
presupposes truth and necessity by projecting a frst principle (Plato's
Ideas , Being , suffcient reason , the Cogito , the transcendental 1 and
its forms , a frst proposition of consciousness , etc.) back to a fcttious original point from where philosophy can begin. According
to Deleuze , the 'true beginning' in philosophy has to refer truth to a
ground that is truly groundless (un vritable sans-fond , DR 154/200) ,
an absolute 'outside' that resists being interiorised. Thought enters
into a relation with pure difference , or in other words , with a di
ferential structure consisting of divergent series of heterogeneous ,
nonsensical elements. In this differential structure inheres an intrinsic
genetic power to produce sense and to 'ground' truth as the limit
53
object of a genetic series of different senses (cf. DR 154/200). Sense

is therefore not simply a formal condition of truth , its conceptual
possibility fabricated retrospectively in the image of the conditioned ,
but a material and genetic principle of truth that is heterogeneous to
what it conditions. Sense as the condition of truth
ought to have an element of its own , distinct from the form of the conditioned. It ought to have something unconditioned capable of assuring a
real genesis of denotation and of the other dimensions of the proposition.
Thus the condition of truth would be defined no longer as the form of
conceptual possibili but rather as an ideational material or stratum ,'
that is to say , no longer as signification , but rather as sense. (LS 19 /3 0)
For distinguishing these two understandings of sense , Deleuze refers
to the former as signification' and to the latter as sense'. That is to
say , signification is the formal condition of truth , providing a number
of conditions under which a proposition considered frorn the point
of view of its logical form and conceptual in1plications "would
be" true' (LS 14/25). By contrast , sense , defined in Deleuze's new
W provides the material condition , something unconditioned and
heterogeneous , which accounts for the real genesis of truth.
It should be noted , however , that Deleuze is not rejecting a whole
tradition of linguistic analysis and philosophy of language. He is
rather trying to complement it by adding a new dimension to the
proposition , and calling this new din1ension 'sense'.he reason for
this is that the din1ension of signication is not sufficient to ground
truth , and the dimension of designation or reference can ground
truth only in rare cases , namely when the proposition is assumed to
be ready-made and isolated from the context of living thought (DR
154/200). As Deleuze says , 'there is only a single case where the
designated stands alone and remains external to sense: precisely the
case of those singular propositions arbitrarily detached from their
context and employed as exan1ples' (DR 154/200). However , as
soon as we place a proposition in the context of living thought , that
is in relation to a problem , we will see how sense is engendered in
the particular determination of the problerrl and its conditions and
how this sense already in1plicates a truth which cannot be detached
from the genesis of sense. For this reason , Deleuze claims that we
alwavs have as much truth as we deserve in accordance with the
sense of what we say. Sense is the genesis or the production of the
true , and truth is only the empirical result of sense' (DR 154/200).
This is 01 of Deleuze's central claims , which he repeats time and
54
The Dogmatic Image o{ Thought

again on different occasions lt leads us to the i111portant concept of
the problen1 , which is the therne of the seventh postulate of the dogmatic Image of thought. For Deleuze , the 'proble111' is a differential
structure endowed with an intrinsic genetic power to generate sense.
Although the problem can be incarnated in propositional form and
in the empirical world , it belongs to an extrapropositional and subrepresentative realm. It is only in relation to a certain problem that a
qllestion becomes possible and a proposition acquires sense. 49
PROBLEMS AND SOLUTIONS
The seventh postulate of the dogmatic Image of thought deals with

our conception of problems and solutions. Deleuze's main point of
critique is that philosophers since Aristotle have misconstrued the
nature of problems. They have fallen prey to a double illusion: a
natllral and a philosophical illusion.
According to the natural illusion , philosophers are led to believe
that problems can be traced from empirical propositions through
a simple change in the phrasing. For instance , it is assurned that a
declarative sentence in which a predicate , y, is attributed to a subject ,
x , can be turned into a problem by means of the interrogative form
'Is y an attribute of x or not?' In this way , Aristotle claims that 'out of
every proposition you will n1ake a problem if you change the turn of
phrase'.50 Thus the co 1'responding problem to the proposition 'Man
is a two-footed terrestrial anin1al' would be 'Is two-footed terrestrial
animal the definition of 111an 0 1' not?' The number of p 1'oblems is
therefore thought to equal the number of propositions that respond
to them.he natural illusion makes us believe that to set and to solve
a p 1' oblem is the SalTIe procedure as interrogation and response. And
since for every question , be it part of an examination , a quiz garne ,
a c1'ossword puzzle , a poll 0 1' government referendum , there is an
answer ready at hand , we are certain that there is always an answer
to a problem , that the solution p 1'e-exists and is only waiting to be
discovered by us. lt is cO l1l1TIonly assumed that a problen1 or question is nothing lTIOre than the expression of a subjective uncertainty ,
a temporary lack of knowledge which can be mended by a piece
of information. In finding the answer , the question or problem is
thought to disappear with the solution.
According to Deleuze , such a concept of problem keeps us in a
state of irnmaturity and dependence on those who actually pose
the questions and set the problems. In fact , problelTIs are not given
55
ready-made; they have to be created. We remain slaves , 'as long as

we do not control the problerns themselves , so long as we do not
possess a right to the problems , to a participation in and management of the problems' (DR 158/205-6). hus the most important
task is to construct the problem , to fully determine its conditions and
to choose the terms in order to specify its sense. Once the problem
is completely determined , a solution necessarily follows. However ,
there always remains the possibility of other determinations and
other solutions. A problem is not dissolved or exhausted either with
a single proposition or with a series of propositions. It does not disappear with a determinate response or a general solution comprising
a series of particular responses. Although the problern never exists
separately from its cases of solutions - that is to say , the problem
insists and persists in these solutions' (DR 163/212) it nonethe
less has a being of its own (DR 269/345; LS 123/148). The being of
the problematic is that of a positive virtual structure or complex of
differential relations endowed with the genetic power to give rise to
concolTIitant solutions. his virtual structure is never entirely captured in any given determination of its conditions and cases of solution. As Deleuze says there is a difference in kind between problerns
and propositions , an essential hiatus' (D R 162/210). Deleuze defines
the problen1 as being
at once both transcendent and immanent in relation to its solutions.
Transcendent, because it consists in a system of idealliaisons or differer
tial relations between genetic elements. Immanent, because these liaisons
or relations are incarnated in the actual relations which do not resemble
them and are defined by the field of solution. (DR 163/212)
Deleuze equates problems with Ideas , fo l1 owing the Kantian definition of transcendental Ideas as essentially problematic. According to
Deleuze , Kant was without doubt the first to consider the problematic not as a fleeting , subjective moment in the mind of the thinking
subject , but as an essential trait of the transcendental Idea , which
functions as the ideal focus (focus imaginarius) or transcendental
horizon for every empirical inquiry into the problem and its conditions (LS 54/70). Deleuze characterises problelTIs as ideal 'objectivities' , as Ideas-problems , which are made up of differential elements
and relations (DR 267/343) , as a structure constitutive of sense (LS
120/145).
However even in the case where philosophers have come to recognise the objectivity or irreducibility of problems and have brought
56
The Dogn1atic Image of Thought

to bear the question of truth or falsity on problems thernselves , they
have not succeeded in giving a satisfactory account of the truth of
problems. According to Deleuze , they have fallen prey to a second
illusion, i.e. the 'philosophical illusion' (DR 159/207). his implies
that a problem is defined as true if and only if it is modelled upon
the form of possibility of propositions. In other words , the truth of
a problem depends upon its logical possibility of finding a solution'
(DR 160/208). Aristotle , for instance , posed problems in terms of
syllogisms and evaluated their 'truth' with respect to their logical
'solvability'. According to Deleuze , the Aristotelian dialectic , conceived as the science or art of syllogisms , is a bad realisation of dialectic. First of all , the subject and premises of syllogisms are chosen
with regard to 'commonplaces' , that is mere opinions accepted by
the majority of men. Furthermore , the logical form of the propositions is modelled upon the general form of empirical propositions.
But Deleuze insists that as long as the criterion for the evaluation of
problems is the extrinsic and variable form of possibility 'we confuse
sense with signification , we conceive of the condition only in the
image of the conditioned' (LS 1221147). What philosophy needs
is the invention of a new dialectics , that is 'dialectics as a superior
calculus or combinatory' of problems (DR 159/207).
Deleuze suggests that a good dialectic frst and foremost has
to decide whether we deal with a badly posed or a well-formed
problem. With a considerable debt to Bergson , Deleuze demands that
the test of truth or falsity has to be applied to problems themselves
(cf. B 15/3). The true and the false primarily qualify problems rather
than their corresponding solutions. The question to ask here is: what
is the criterion on which the truth or falsity of a problem depends?
According to Deleuze , the criterion has to refer to an internal characteristic: it is the relation of the problem with its conditions which
defines sense as the truth of the problerrl as such (LS 1211145). A false
problem would be one whose conditions remain either insufficiently
determined or overdetermined. But how do we pose a true problem?
We have to detern1ine its conditions and the distribution of singular
and ordinary points (the important and the unin1portant points) (DR
189/245). Adn1ittedly , this relnains a rather vague in
57
CONDITIONS OF THOUGHT: DELEUZE AND
RANSCENDENTALIDEAS
between rnonads to rne 1'e phenomena. But by sirnply insisting Look ,

that's not the way things a1'e' , one has not achieved a 1'efutation of
Leibniz's systel11. In fact , one has not even unde 1' stood the sense of
the Leibnizian theo 1'Y of monads , fo 1' it finds its sense only in the
subjacent p 1' oblem that inspi 1'es it.
Oncewe forget' the problem , we have before us no more than an abstract
general solution , and since there is no longer anything to support that
generality , there is nothing to prevent the solution from fragmenting into
the particular propositions which constitute its cases. (DR 162/211)
Mo 1'eove1', one understands nothing if one does not see that Leibniz
opens up a p 1' oblematic field which forces him to go all the way following the necessa 1'y in lications of the questions and p 1'oblenls that
he encounte 1' s. As Deleuze says in one of his lectu 1'es on Leibniz: he
cannot stop.53 Leibniz's constant deduction of p 1'inciples and n1ad
c1'eation of concepts a 1'e not a 1' bit1'a 1'Y effo 1' ts; they a 1'e necessitated
by a continuous specification of the p 1' oblematic field. Let us sea 1'ch
fo 1' one of the initial p 1'oblems 0 1' questions. It is ce 1' tainly not the
metaphysical question: 'What is the1'e What a 1'e the most aSlc
components of realityThese What isquestions lack the g1'ip of
necessity. They lead us nowhe 1'e.
We will now ext1'act what appears to us a t 1' ue p 1'oblem f1'o111 his
notes that he w 1'ote p 1'io 1' to his lette 1' to A1' nauld f1' on1 July 1686:
1 mean [. . .] that there is always something to be conceived in the subject
which serves to explain why this predicate or event pertains to it , or why

this has happened rather than not. 54
Leibniz nds himself in a dilemma. He wants to believe that eve 1'Ything that has happened and that will eve 1' happen to a subject has
a 1'eason (p 1'inciple of sufficient 1'eason). Because of theological conce1'ns it cannot be the case that God inte1' fe 1' es ad libitum in the o 1' de 1'
of the unive 1'se. Fo 1' Leibniz , God is not like a h Ul11an being whose
acts of 1'esolution depend on pa 1' ticula 1' circun1stances. F1' om the ve 1'y
fi 1' st moment , God has fo 1'eseen and 1'egulated the enti1'e sequence
of things fo 1' ete 1'nity , he does not act apa 1' t f1' om this order of the
unive 1' se that he has chosen. That is to say , nothing irregular happens
in the world. But , on the other hand , whateve 1' happens to a subject
cannot follow with int1' insic necessity from its natu 1'e 0 1' essence.
Only fr0111 mathematical concepts can it be said that the predicates
contained in the concept necessa 1' ily follow by a simple analysis of
the concept. Geol11etric demonstrations fo1' instance , yield necessa 1'y
58
The Dogmatic Image
1' Thought
propositions whose contraries irnply a contradiction and are therefore impossible. Such ruths of essence' , governed by the principle
of non-contradiction , fall into the sphere of mathematics. Yet this
doctrine of universal necessity cannot bear on truths of existence
for instance contingent truths that concern our life. For theological
and moral reasons , Leibniz rejects (Spinozist) fatalism and wants to
rnaintain a sense of the freedom of the subject. He achieves this by
identifying a third option: yes , there is an intrinsic and necessary connection between the subject and all its past , present and future states.
Indeed , the individual notion of each person contains all that has
ever happened , is happening and will ever happen to him. However ,
in order to safeguard the notion of freedom , this necessity can only
be ex hypothesi , that is the contrary of whatever happens lnust be
possible. 55 Leibniz's response to the problern drives him to develop
a general solution , which enc Olnpasses the theory of complete indi.
vidual concepts , the theory of infinitely many possible worlds in the
mind of God , and the notion of compossibility.
Now we have to ask for the conditions under which the problen1atic field is determined. he lynchpin of the whole problematic
eld is the principle of reason: il est sine ratio (nothing is without
reason). his means that nothing is or happens for which we could
not find a reason why it is rather than is not , why it is so rather
than otherwise. This is the popular formula of the logical principle
that every truth is analytic. In every true proposition the predicate is
included in the subject. Therefore every truth can be demonstrated
a priori by means of an analysis of the subject term. In his serninal
essay On Leibniz's Metaphysics' (1902) , Louis Couturat claims that
'it is the entire Monadology which Leibniz thus progressively derives
from the principle of reason\56 The metaphysical import of the
principle' is tremendous:
He [Leibniz] derived from it the principle of indiscernibles and that other

principle [. . .] that there are no purely extrinsic characteristics (dnominations)'; then , step by step , the notion of the monad [. . .]; further , the
pre-established harmony [. . .]; finally , the ideality of space and time and
hence of movement and of bodies [. . .], and the immortality not only of
souls , but of all substances. 57
Taking the principle of reason (every truth is analytic) as the point
of departure Leibniz could count on the unreserved approval of his
contelnporaries , who , like hin1 , received an education in Aristotelian
and scholastic logic. However the decisive step occurs , when Leibniz
59
applied this principle equally to necessary truths of essence and

contingent truths of existence. This means that Leibniz considered
contingent truths not as synthetic , but as analytic. The sub-problem
that then emerges concerns the question of how a predicate can be
contained in the subject with the proposition being necessary only
ex hypothesi (i.e. the contrary being possible). A solution to this
problem finally came into sight thanks to Leibniz's involvement with
mathematics , in particular the infinitesin1al calculus. A new and
unexpected light finally arose in a quarter where 1 least hoped for itnan1ely , out of mathematical considerations of the nature of the in
nite.'58 Leibniz was now able to understand the nature of contingent
truths: they differ frorn necessary truths as the infinite differs from
the finite , or as irrational nurnbers differ from rational nulbers. This
means that contingent truths are indeed analytic , but the analysis of
their terms is in nite and God alone can complete this infinite analySIS. his is how Leibniz was able to secure both the notion of freedom
and the notion of necessity. According to Leibniz , the true idea of
free will does not exclude detenninism on the contrary , every free
decision or act has its reasons in the individual nature of the subject.
Yet , it is free to the extent that it is unpredictable and escapes any
generallaw. Only God can foresee each decision or act with certainty
and infallibility. Hence , the second most lnportant condition of the
Leibnizian problernatic field is the condition of infinity or in nite
analysis. As Leibniz repeats on several occasions: contingency is
rooted in infinity.59
Having shown the in1portance of logic and mathelnatics for
Leibniz's metaphysical theses , we will not follow Leibniz any further in
the elaboration of his metaphysical systen1. What has become apparent is that we cannot understand anything of Leibniz's notion of the
monad if we detach it from its context. Reading the Monadology is
not sufficie because the Monadology begins with the notion of the
rnonad , but the rnonad itself derives frorn a continuous specification
of a problenlatic field. As Couturat remarks , the order followed by
the Monadology is really the inverse of the order that is both logical
and genetic'.60 Extracting the problem fron1 a philosopher's thought
and detern1ining the conditions and singular points that stl
60

for contingent tr l1 ths , the speci f1 c Leibnizian problem disappears
and his sol l1 tion makes no sense any more. This happened with the
Kantian determination of contingent tr l1 ths as synthetic. Sl1 ddenly ,
the existence of p l1 rely extrinsic characteristics became possible.
Th l1 s two things could be the same with regard to their concepts
but different with respect to their spatio-ternporal relations. Not
every external relation between things had to be expressed by an
internal modi f1 cation of the substance or essence of a thing. Hence
Leibniz's principle of indiscernibles , that for every concept there is
one and only one thing , could not be upheld. Equally , the notion of
the windowless lTIonad' and the pre-established harmony between
monads no longer made any sense. Sl1 bjects c0111d now interact not
only apparently but really on a physicalleve l.
HE POSTULATE OF KNOWLEDGE
Let l1 S proceed with the eighth post111ate , the post111ate of knowledge.

While in the preceding paragraph we provided an example of a
philosophical problem or problematic f1 eld with its conditions and
concomitant sol l1 tions , Dele l1 ze makes it clear that problems exist
in vario l1 s different elds - there are mathematical problems , problems in biology and physics , social , political and moral problems ,
problems in the arts , problerns in Ol1 r relationships and in everyday
life. Deleuze's very specifc definition of the concept of problem' is
accompanied by a whole new idea of apprenticeship' or learning'
through the enCO l1 nter with problems. In the dogrnatic Image of
thought 'learning' is considered as jl1 st an intermediary passage from
ignorance to knowledge l1 nder the g l1 idance of methods. he desired
res l1 lt of the process is the acquisition of knowledge which , when it
is achieved , settles the whole question or problem. For Dele l1 ze , on
the other hand , learning' is not to be equated with an empirical ,
psychological process , but with a transcendental and l1 nconscious
activity. 'Learning de f1ned in this new way , involves two essential
aspects: (1) the exploration of the problen1atic Idea , that is the relation of its differential elements and singular points; (2) the transcenden unregulatd exercise of the faculties triggered by the encounter
with the Idea-problem (cf. DR 164/213 , 194/251). Deleuze calls the
dereg111ation and trespassing of the faculties beyond their proper
domain their transcender exercise in opposition to their empirical
exercise. 61 'Transcendent' is not taken to refer to objects o l1 tside the
empirical world , but rather to something within the world , which
61
however , cannot be grasped by the empirical operation of the facul

ties being employed together in a :ommon sense'. In its transcendent
exercise each faculty apprehends something which concerns it exclusively , which it alone is able to grasp , yet which is ungraspable from
the point of view of its empirical exercise (DR 143/186). Let us look
at these two essential aspects of learning' in greater detai l.
In order to illustrate what it means to explore a problematic Idea ,
Deleuze takes up Leibniz's example of the sea and considers the case
of learning to swim. The sea with its movement of waves incarnates
the Idea with its differential relations and corresponding singularities , that is physical points at which , for instance , the wave breaks or
a dangerous current is formed. Our body has to adjust itself within
the systern of waves and troughs of the sea. It has to find a way of
complicity between its own distinctive points and the singular points
of the objective Idea which make up the problematic field (c f. DR
165/214 192/248). Our conscious acts are just the global outcome ,
derived from the differential relations of infinitely many minute
physical points or excitations and infinitely many rninute perceptions
and actions of our biological bodies.
he process of learning is in fact an unconscious , 'involuntary
adventure' (DR 165/215): entering into the differential relations of a
problematic Idea is like being thrown into the ocean. No method can
guide us safely through the differential , problematic field. Before we
even start to think and act consciousl) we ha ve already undergone a
'process of forn1ation of thought a violent 'training which brings the
whole unconscious of the thinker into play' (NP 108 /1 24).
hought
does not need a method but a paidei a formation , a culture.

Method in general is a means by which we avoid going to a particular
place , or by which we maintain the option of escaping from it (the thread
of the labyrinth). (NP 110/1 26)62
Taking recourse to the Nietzschean concepts of paidei and

states that thinking is not just a natural exercise
based on a good will and the application of method. Thinking is
rather an extraordinary event , which explodes in thought itself under
a constraint , a violent training. That which forces us to think is not
an identifiable , recognisable object. In opposition to the process of
recognition , which presupposes a cornn10n sense or the collaboration
of all the faculties in relation to a supposed same object, the process
of learning rather requires a disjointed and unregulated exercise of
the facultie such that each faculty COlTIlTIUnicates the violence it
'culture Deleuze
62

receives to the other faculties carrying them to their transcendent
limit or 'n powe r. Arthur Rimbaud clairrls that the poet becomes
clairvoyant only by undergoing a deregulation of the senses' (dra
glement de tous les sens) , by experiencing all forrrls of love , of suffer
lf
a
to
is
maintain nothing but t}le essence 63 According toMarcel Proust it

in the dal
fk
regions
t ha
t we discover tru
lths
t ha
t matter tous.
The truths that intelligence grasps directly in the open light of day have
sornething less profound , less necessary about thern than those that life
has cornmunicated to us in spite of' ourselves in an irnpression , a rnaterial
irnpression because it has reached us through our senses , but whose spirit
we can extract. 64
For Deleuze , it is the fundamental encounter with an Idea-problelu

which sets our faculties into luotion and raises each faculty to the
level of its transcendent exercise. But how do we encounter an Ideaproblem which is , according to Deleuze , a transcendental instance'
(DR 164/213)? We encounter it through 'signs' , within a symbolic
field' in which the problen1 expresses itself. As Deleuze says in his
book Proust and Signs , the signs mobilize , constrain a faculty: intelligenc luemory or imagination. This faculty , in its turn , mobilizes
thought , forces it to conceive essences' (PS 98-9/191). He continues:
Everything that teaches us something ernits signs; every act of learning is
an interpretation of signs or hieroglyphs. Proust's work is based not on
the exposition of mernory , but on the apprenticeship to signs. (PS 4/9)
The narrator in Proust's In Search ol Lost Time is confronted
with rnany different signs: worldly signs (i.e. signs that society emits) ,
signs of love , material signs (i.e. sensuous impressions or qualities)
and essential signs of art. Although the narrator always feels that he
wastes' his time in society or with love affairs instead of fulfilling
his dream and starting a professional career as a writer , he gradually
con1es to realise that , in effect , he has been an apprentice all the time.
It is through the violent encounter with signs that he is forced to
think and forced to create. For instance , his love for Albertine and the
jealousy she awakens within hin1 force him to interpret all the enigmatic and deceptive signs she emits , to iIuagine all possible kinds of
infidelity , danger or loss. He wants to penetrate the unknown world
that revolves around her and fron1 which he is excluded. But also the
landscape in which they first lU the sea and beach in Ballc - IS
converted to signs and enveloped in her face and body.
whole
63
sensual appearance - the freshness of her face , her red cheeks , the
little beauty patch and her soft fragrance - Seerrl like rnaterial signs ,
which need to be deciphered. Like a work of art , Albertine triggers
an inexhaustib unlimited process of interpretation and thought,
finally leading the narrator to an aesthetic production which he
would have never been capable of without this love. He learns that
the signs of love refer to a truth that is not restricted to the individual
person , Albertine , but to all the past and future loves. The signs
of love reveal a secret: the power of difference and repetition. The
present love only repeats our past loves , and there is no end in this
regress because there is no original first love. In a futural sense of the
word , the present love also repeats' its end , the inevitable break-up ,
with every heated argument caused by jealousy.
Every love affair is 'a dispute of evidence' (PS 117/1 43) and the
jealous lover nds himself in a delirium of signs' (PS 122/1 50). His
task is to decipher the 'spiritual element' , the essence' ortruth'
out of the con1plex sensation. There is certainly a Platonic elernent
in Proust. And Deleuze pays homage to Plato by granting that his
theory of apprenticeship is an exception within the broader picture of
the In1age of thought. While the Image of thought n10dels 'learning'
on an elnpirical psychological process , and disparages it as a rnerely
transitory stage in the acquisition of knowledge , Plato de ned 'learning' as a truly transcendental movement of the soul , irreducible as
lnllC h to knowledge as to non-knowledge' (DR 166/215-16). The
transcendental conditions of thought are not drawn fron1 innate , a
priori forrrls but found in a process of reminiscence. T lne is introduced into thought , though not in the form of the empirical time of
a thinking subject. The object to be remembered , the nzemorandum ,
is not a contingent empirical content that was a former present and
has become our past , but a transcendental Idea which has never been
present but always past. The realm of Ideas eludes our empirical
time and consciousness. However , Deleuze accuses Plato of a sleight
of hand , because in the er Plato assin1ilates this transcendental
memory to precisely the process of ren1embering on the part of an
en1pirical psychological consciousness. Just as the errlpirical subject
can ren1ember events that happened to it in a former present , it is
capable
64
Fo 1' Deleuze , the object of 1'eminiscence , the l1 zemorandum , is that

which can only be 1' ecalled , just as the n1ate1' ial sign , the sentiendum ,
is that which can only be sensed 0 1' felt , and the cogitandum that
which must be thought but is beyond 1'ecognition. Eve 1' Y faculty has
its p 1' ope 1' object that mobilises and pushes it to its t 1'anscendent limit
- 'as though the object of encounte 1', the sign , we1'e the bea 1'e1' of a
p 1' oblem' (DR 140/182). In effect , the Idea-problem is al 1'eady the 1'e
in the sign , in an enveloped or inroluted state. One must be sensitive
to signs and open oneself to their violence (cf. PS 10 1/194). They are
the problem-conditions for thought and pu 1'e c1'eation.
Deleuze's definition of lea 1' ning as an apprenticeship or interpretation of signs , which he first put forward in his book Proust and
S is still present in Difference and Repetition. Bu t: the traditional
hermeneutic te 1' minology ( thinking subject' , interpretation' , etc.)
gives way to a te 1' minology drawn from rnathematics ( differential
relations' , singula 1'ities etc.). Ideas-problelTIs a 1'e , for instance ,
equated with Albe 1' t Lautn1an's dialectical Ideas , which bear the
urgency of problems and are prior to the discovery of solutions. 65
Deleuze obviously wants to distance hin1self frorn a philosophy of the
subject. Instead , he st 1'esses the objectivity of Ideas and their mindblowing' effect , which takes f1' om us the power to say '1' (WP 55/55).
he subject stays not intact , but is transformed in the extrerne
expe 1' ience of lea 1' ning.
Deleuze's c1'iticisn1 of the dognlatic Image of thought is an essential
step in his p 1' oject of t 1' anscendental philosophy , that is the project
of nding the t 1' anscendental genetic and p 1' oductive conditions
of thought. It is first of all necessa 1'y to libe1' ate thought f1'om the
const1'aints imposed upon it by a lTIo 1' ally motivated o1'der of 1'ep 1'esentation and 1'ecognition. In this chapte 1', we have dealt with some
cha 1'acte 1' istics of this orde 1' and also begun to sketch a new Deleuzian
irnage of thought , that is an lTIage that 1'elates thought to a transcendental and genetic ground - 0 1' 1' athe 1' 'ung 1'ounding' (effonde111ent , DR 9 1/123) - of pure diffe 1'ences' 0 1' diffe 1'ential 1'elations of
p 1' oblematic Ideas. The encounter with these problematic Ideas - be
they expe 1'ienced as mate 1' ial signs , as signs of a 1' t , as philosophical
p 1' oblems l' whateve 1' - gene 1' ate the act of thinking in thought as an
involunta 1' Y and often violent adventu 1'e. Thought , if it is to conque 1'
the new and not only p 1'oceed on the common path of already p 1'eestablished and prejudged categories and values , needs to rid itself of
the dogmatic In1age and discover its unconscious , sub-representative
65
source. Pure difference or the differential is the model for a genetic

and non-sensible element , which is fundamental to Deleuze's renewal
of the project of transcendental philosophy.
In the foIIowing chapter , we wiII first look at Deleuze's early book
Nietzsche and Philosophy , in which his critical rethinking of the
nature of the transcendental is already to be found , and then foIIow
him in his engagement with Maimon's criticism of the Kantian
transcenden ta l.
Notes
1. The essay How Do We Recognize Structuralismwas presumably
written in 1967 and frst published in 1972.
2. Deleuze explicitly exempts Lacan from this criticism, since Lacan
introduced a third term above and beyond the real and the imaginary:
namely the order of the symbolic.
3. Deleuze's critique of the notion of ideology is clearly expressed since
his collaboration with Guattari in the 1970s and 1980s. In the 1972
interview with Catherine Backs-Clment, Guattari says: we don't
have any time for concepts like ideology , which are rea11y no help at
a11: there are no such things as ideologies' (N 19/3 2). At several places
in Anti-Oedipus 'ideology' is criticised as a confused and misleading
notion. Deleuze and Guattari contend that the concept of ideology is
an execrable concept that hides the real problems , which are always of
an organizational nature' (AO 344/412). On their critique of psychoanalysis as an analysis of the subjective, as defned by ideology' , see AO
345/413. See also ATP 4/10: there is no ideology and never has been.'
4. Even the judgment of knowledge envelops an infnity of space , time ,
and experience that determines the existence of phenomena in space
and time ("every time that. . ."). But the judgment of knowledge in this
sense implies a prior moral and theological form , according to which
a relation was established between existence and the in nite following an order of time: the existing being as having a debt to God' (CC
127/159).
5. Deleuze , Nietzsche' , p. 69. Cf. also NP 104/119.
6. See also NP 89-90/102: 'There has never been a more conciliatory or
respectful total critique. [. . .] Kant merely pushed a very old conception of critique to the limit , a :onception which saw critique as a force
which should be brought to bear on a11 claims to knowledge and truth
but not on knowledge and truth themselves; a force which should be
brought to bear on a11 claims to morality but not on morality itself.
Thus total critique turns into the politics of compromise: even before
the battle the spheres of influence have already been shared out. hree
66
7.
8.
9.
10.
11.
ideals are distinguished: what can 1 know? what should 1 do? what can
1 hope for? Limits are drawn to each one , misuses and trespasses are
denounced , but the uncritical character of each ideal remains at the
heart of Kantianism like the worm in the fruit: true knowledge , true
morality and true religion.'
C f. Zourabichvili , Une philosophie de l'vnement' , pp. 22-3: The
failure of grounding is not alien to the fragility of this postulate [of the
intimacy with the outside]. It is not surprising that the necessity eludes
when we attempt to close thought upon itself' (my translation ,
D. V.).
Cf. DR 274/351: Is this not the most general characteristic of the
ground - namely , that the circle which it organises is also the vicious
circle of philosophical "proof" , in which representation must prove
what proves it , just as for Kant the possibility of experience serves as
the proof of its own proofSee also LS 19 /3 0.
C f. Sauvagnargues , Deleuze: L'Empirisme tra l1 scendantal, p. 84.
But the form of exteriority of thought - the force that is always external to itself, or the f nal force , the th power - is not at all another
image in opposition to the image inspired by the State apparatus. It is ,
rather , a force that destroys both the image and its copies , the model
and its reproductions , every possibility of subordinating thought to a
model of the True , the Just , or the Right (Cartesian truth , Kantian just ,
Hegelian right , etc.).' (ATP 377/467)
Deleuze thereby aligns himself with the philosopher Salomon Maimon ,
one of Kant's early critics yho wanted to replace the transcendental
conditions of possible experience with genetic conditions of real experience. Deleuze believes that this alternative approach to transcendental
philosophy f rst put forward by Maimon is also taken up by a range of
later philosophers (Schelling , Bergson , Nietzsche , Foucault) and even
the f lm-maker Pasolini. See the following quotations: (1) Thus it is
not the conditions of all possible experience that must be reached , but
the conditions of real experience. Schelling had already proposed this
aim and de f ned philosophy as a superior empiricism: this formulation
also applies to Bergsonism' (DI 36/49). (2) he Nietzschean and the
Kantian conceptions of critique are opposed on f ve main points: 1.
Genetic and plastic principles that give an account of the sense and
value of beliefs , in
67
12.
13.
14.
15.
16.
followers remain Kantians (bringing principle down to fact)' (CIT

276/42-3 , note 8/8).
As Nietzsche puts it , untimely thought means acting counter to our
time and thereby acting on our time and , let us hope , for the benefit of
a time to come' (Nietzsche , On the Uses and Disadvantages of History
for Life' , p. 60). It should be noted that the concept untimely' appears
here in the context of Nietzsche's discussion of the use of classical
studies (philology and history) , the above quotation being preceded by
the words: 'That much , however , 1 must concede to myself on account
of my profession as a classicist: for 1 do not know what meaning classical studies could have for our time if they were not untimely.' However ,
m Schopenhauer as Educator Nietzsche also uses this concept with
reference to philosophy , in which context it becomes significant for
Deleuze ( Schopenhauer as Educator' , pp. 133 and 145-6).
See also DR 167/217; for references from Deleuze's later work , see
ATP 377/467 and CC 82/106.
Martin , Variations , p. 157.
Arguing in a similar way , Zourabichvili says: The oscillation of critique between the theme of a "thought without image" [. . .] and that of
a "new image of thought" [. . .] indicates perhaps the moment in which
Deleuze confronts the question in itsel f. In fact , this oscillation reflects
the paradox of a transcendental philosophy , which in posing itself as
immanent seeks the conditions that are "not larger than the conditioned" and that constitute a sort of "plastic" transcendental field [. . .].
Now , what is the worth of a theory that pretends to get by without an
image , although it describes the conditions of an act of thinking? [. . .]
In reality , the paradox is that the new image - the "rhizome" [. . .] is
the image 0/ thought without image , an immanent thought that does
not know in advance what thinking mea (in 'Une philosophie de
i vnement' pp. 63 -4; my translation , D. V.).
Deleuze , Rhizome (Introduction)' , in ATP 3.25/9-37. Deleuze also
utilises the model of the rhizome to designate a creative becoming' of
language in literature , which defies the language of the majority , its
ordinary syntax and pre-existent semantic distribution: Creative stuttering is what makes language grow from the middle , like grass; it is
what makes language a rhizome instead of a tree , what puts language
in a perpetual disequilibr
68
The Dogmatic lmage of Thought

18. In his book La Msentente: politique et philosophie (Paris: Editions
Galile , 1995) , Jacques Rancire argues in a similar way. he revolutionary subject is not an identifiable subject captured by the order of
representation (the law of the police' in Ranci 's terms). Rather the
bearers of rerolutionary struggle are processes of subjectication that
is dissolved subjects which elude all codes of representation and assemble and disassemble always anew.
19. This interview with Deleuze was conducted immediately prior to the
student revolts in Paris which finally led to the general strike in May
1968.
20. Cf. the opening sentence of Part 1 of Descartes' Discourse on the
Method.
2 1. See Kant's letter to Herz (26 May 1789). The translation of the letter
appears in Appendix 11 of Maimon , Essay , p. 234.
22. For more on Plato's method of division and the role of myth , see DR
59-64/82-9 and L5 253-66/292 307.
23. For Plato , the paradigms of those who only simulate or mimic' the
wise are the Sophist and therepresentational poets' , cf. Plato , 50phist
235a-236d and Repuhlic , Book X, 601a-606d.
24. Patton Anti Platonism and Art' , p. 154.
25. As Aristotle explains , terrestrial' and aquatic' are not simply denotations of locality but of quality. hat is to say , a terrestrial animal does
not become an aquatic animal when it is found in water , and vice versa.
See Topics , Book VI , 6 , 144b 145a , in The Complete Wor5 vo l. 1,
p.244.
26. Deleuze chooses as an example of specific differences the predicates
with feet' and with wings' (DR 30/46). Although Aristotle indeed
cites these predicates as specific differences (Topics , VI , 6 , 143a-b , in
The Complete Wor5 vo l. 1, pp. 241-2) , they are (if provided in this
conceptual couple) not the best example , since the predicates with
feet' and with wings' are not contraries , i.e. not mutually exclusive
in the object (for there are animals with wings and feet). The reason
why with feet' is indeed a specific difference is because there exists the
opposite predicate without feet' within the same genus ( a l11 m'). (As
Aristotle explains , a genus can also be divided by a predicate and its
negation as in the case of a length with breadth' and without breadth'
(a line); Topics , VI , 6 , 143b, in The Complete Works , vo l. 1, p. 242).
27. Cf. Aristotle , Topics , VI , 6 , 145a , in The Complete 'orks vo l. 1,
p.244: The differe
69
differences , but only to the objects that are determined by the specific
differences. For instance , the genus animal' can be said of the human
species but it cannot be attributed to the specic difference rationa l'
(Topics , VI , 6 , 144a , in The Complete Works , vo l. 1, p. 243).
29. 'All errors of subreption are always to be ascribed to a defect in judgmen never to understanding or to reason' (CPR A 643/B 671).
30. Deleuze agrees with Bergson at the true and the false are functions
not so much of propositions that serve as solutions to problems , but
first and foremost of problems themselves. He distinguishes false
problems , that is inexistent or badly posed problems , from true problems , and compliments Bergson on having found a method (method of
intuition) for the intrinsic determination of the ralue of problems (see
B 15-2113-11).
3 1. Zourabichvili remarks that Deleuze advocates a new , differential conception of violence , which describes an aggressive critical force necessary for philosophical thought. This critical aggressivity (agressivit
critique) essentially differs from an aggressive thought that finds its
sources in negation and turns stupidity into malerolence and cruelty.
See Zourabichvili , Une philosophie de l'vnement' , pp. 34--5.
32. A detailed account of Deleuze's denition of conceptual personae (as
distinct from aesthetic gures and psychosocial types) can be found in
W Chapter Three.
33. As Deleuze remarks , the Cogito is indeed a nonsensical proposition
to the extent that this proposition purports to state both itself and
its sense' (DR 276/353). According to Deleuze , this is precisely what
defines a nonsense word (such as Lewis Caroll's word snark' or the
Stoic expression lituri') (DR 155/201) , whereas any signifying word
or proposition refers back to some other proposition or object whose
sense it expresses. he Cogito has no other reference than its selfreference , that is to say, the power of reiteration in indefinite regress (1
think that 1 think that 1 think . . .)' (DR 155/202).
34. This remark on the private thinker comes from a discussion of
Nietzsche - see ATP 376/467.
35. Deleuze attributes the production of counterthoughts' to 'private
thinkers' such as Kierkegaard , Nietzsche and Shestov (ATP 376/467).
But this is true of Dostoyevsky's idiot as well , whose way of thinking
contrasts with the opinions and
70
'he Dogmatic Image of Thought
43. See L5 105 /1 28 and 123 /1 49.

44. See also CPR A 59 60/B 84: 'These criteria [of truth] concern only the
form of truth , i.e. , of thinking in general , and are to that extent entirely
correct but not sucient. For although a cognition may be in complete
accord with logical form , i.e. , not contradict itself, yet it can still always
contradict the object. The merely logical criterion of truth , namely the
agreement of a cognition with the general and formal laws of understanding and reason , is therefore certainly the conditio sine qua non
and thus the negative condition of all truth; further , however , logic
cannot go , and the error that concerns not form but content cannot be
discovered by any touchstone of logic.'
45. See CPR A 62/B 87; see also A 277/B 333 and B 149.
46. he example is taken from the novel H use 0/ Leaves (New York:
Pantheon Books , 2000) , written by Mark Z. Danielewski.
47. Cf. Lebrun: "Grounding" , in this sense , amounts simply to certifying
in a good and proper way that the pretension to universality which
is inscribed in my proposition turns up to be irrefutable because it
responds precisely to the condition that is alone able to render it
valid' (in 'Le transcendantal et son image p. 209; my translation ,
D. V.).
48. C f. DR 159/206; NP 112 , 104 /1 18 , 110/125; B 16/5.
49. Zourabichvili , Une philosophie de l' vnement p.33: It's as a function of a certain problem that a question becomes possible and above
all a proposition acquires sense. Sense is nothing but the relation of a
proposition , not to the question that it answer i.e. the sterile double ,
but to the problem outside which it has not sense' (my translation ,
D. V.).
50. Aristo tI e, Topics , Book 1, 4 , 101b, in The Complete Works 0/ Aristotle ,
vo l. 1, p. 169.
5 1. Zourabichvili demonstrates the art . posing problems with reference
to Deleuze's early book on Hume , in Une philosophie de l' vnement' ,
pp. 33 -4. (For the reference to Deleuze , see E5 105-6/118-19). For
further examples of problems defined in Deleuze's manner , see Patton's
article 'The World Seen From lithin' (1997).
52. Russell , for instance , admits that it was only when he read Leibniz's
Discourse on Metaphysics and the letters to Arnauld that he saw how
Leibniz's system could be deduced from a few simple premises. Before
that: 1 felt - as many others have felt that the Monadology was a
kind of fantastic fairy tale , coherent pe
71
56. Couturat , On Leibniz's Metaphysics' , p. 23. For a reference to Leibniz ,

see his letter to Arnauld (4 /1 4 July 1686): there must always be some
basis for the connexion between the terms of a ' osition and it is
to be found in their concepts. That is my great principle with which
1 believe all philosophers must agree , and of which one of the corollaries is the common axiom that there is a reason for everything that
happens , and that one can always explain why a thing has worked out
this way rather than that [. . .]. One sees that from the above-mentioned
principle 1 draw surprising consequences , but it is only because one is
not accustomed to pursue far enough the clearest knowledge' (in The
Leibniz-Arnauld Correspondence , pp. 63 -4).
57. Couturat , On Leibniz's Metaphysics' , pp. 22-3.
58. Leibniz , On Freedom' , p. 107.
59. Couturat, On Leibniz's Metaphysics' , p. 25.
60. lbid. , p. 23.
61. lt is worth noting that Kant himself distinguishes between a transcendental use or rather misuse of the understanding that results from a
failure of the faculty of judgement to restrict the application of categ
ries to their territory' , and transcendent principles that eliminate all
lines of demarcation and attempt to seize a new territory beyond po
sible experience (CPR A 296/B 352-3). He opposes the latter to proper
immanent principles of the understanding that are of merely empirical
use. ln a similar way , Deleuze opposes the transcendent exercise of the
faculties to their empirical operation in recognition , but for Deleuze , in
contrast to Kant , the transcendent refers to something unrecognisable
(unthinkable , unimaginable , immemorial , imperceptible) within the
world , not beyond it.
62. Deleuze is presumably referring to the ancient Greek myth of Theseus
and the Minotau r. ln Crete , King 1inos had a labyrinth constructed to
imprison the Minotaur , a bull-headed creature who devoured men for
sustenance. Theseus volunteered to slay the monster , and Ariadne , the
daughter of Minos , who had fallen in love with Theseus , provided him
with a ba l1 of thread allowing him to retrace his path. By means of this
thread , Theseus , after having killed the Minotaur , eventually found his
way out of the labyrinth.
63. Rimbaud , 'Lettre Paul Demeny du 15 mai 1871' , in Lettres du
Voyan p.137: The poet makes himself clairvoyant [voyant] through
a long , immense and reason
72

64. Proust , A la recherche du temps perdu , p. 878.
65. See Chapter 3 , this volume. For more details , see also Duffy, Albert
Lautman' , in Jones and Roffe (eds) , Deleuze's Philosophical Lneage ,
pp. 356-79 , and Bowden , The Priority of Events , ch. 3, pp. 95-15 1.
73
The
Dema d forranscendental
Genetic
Conditio S
As we have already mentioned , philosophy , for Deleuze , is inseparable frorn critique He admires Kant for haring brought a revolution
to philosophy by means of his transcendental critique. he domain
of the transcendental is not the domain of the transcendent. Kant
undermines the traditional philosophical distinctions , in particular
with his notion of phenomenon which has a completely different meaning from the meaning that this tenn had acquired in the
philosophical tradition. Since Plato , philosophers had distinguished
the appearance of a thing from its n1etaphysical essence. They had
opposed the apparent world' of sensuous , perishable and illusive
appearances to the transcendent ntelligible world' of eternal
and true essences. Kant replaces the disjunctive couple appearance/essence with the conjunctive couple of that which appears
(Erscheinung) and its transcendental conditions of apparition. he
Kantian phenomenon is no longer a deceptive sin1ulacrmn or an
inferior copy related to an original essence , but an object of experience related to transcendental conditions which constitute it as a
possible object for us , that is as having a sense or signification. This
means that the dimension of signification takes the place of the metaphysical essence. 1 Truth is no longer hidden in a metaphysical realm ,
unattainable in principle , but can be attained in experience on the
premise that experiential cognition complies with the transcendental
conditions of experience. Transcendental logic provides a necessary
criterion for truth , namely the relation to an object (which is the
dimension of signification).
The probleln with Kantian transcendental philosophy , however ,
is that Kant defines the transcendental conditions as pure a priori
glven without being able to account for the genesis of these a priori
concepts thenlselves. Kant is first and foremost concerned with
guaranteeing the universality ar necessity of our objective experience , that is the necessary law-like interconnection of experiences
74
The Demand for Transcendental Genetic Conditions

in synthetic judgements , and he does so by founding the objectivity
of our experience on given , a priori concepts. Deleuze , as we have
seen , argues that the foundation that Kant provides is very fragile.
Kantian transcendental conditions suffer from a double defect. he
first defect is that they are too far from the actual; that is why the
slgmcation they constitute is too abstract. They are only conditions
of possible experience and not capable of generating real experience.
In the second place , they are too close to the actual; that is why they
are not truly transcendenta l.hey are traced from the empirical and
retrospectively presupposed as transcendental conditions.
Deleuze demands that a transcendental philosophy has to account
for the genesis of real experience instead of simply assuming condi
tions of the possibility of experience. A viewpoint of genesis has to
be substituted for a viewpoint of conditioning. his demand for
transcendental genetic conditions resonates in many of Deleuze's
books from the 1960s and explains his interest in post-Kantian
philosophers such as Maimon , Nietzsche and Bergson. he viewpoint of genesis abandons thehigh' idea of foundation (cf. NP 2/2)
and , instead , relates thought or the real to sub-representational ,
differential and immanent dynamisms of some kind whether it is
Maimon's iifferentials of consciousness' , Nietzsche's will to power
or Bergson's virtual melnory.
In this chapter we wiU focus on Nietzsche's and MaInon's respective contributions to a critique of Kant , a critique that is brought to
bear on the Kantian transcendental conditions themselves.
eleuze's Netzsche
Deleuze's book Nietzsche and Philosphy published in 1962 , is
symptornatic of a newly awakened interest in the study of Nietzsche ,
which reached its peak in France in the 1960s and 1970s. Before
the Second W orld Wa Nietzsche had been a rather rnarginal figure
in French philosophy , and the reading of Nietzsche had been promoted mainly by very conservative circles that brought Nietzschean
thoughts to bear on reactionary and elitist themes. A gradual shift
of perspective finally evolved during the war years , thanks to a
sn1all group of philosophers , such as Georges Bataille , Jean Wahl ,
Jean Hyppolite and Georges Canguilhem , who n1et in the salon of
Marcel Mor. Alnong those who freq uented these philosophical
reunions was the young Deleuze , at this time still a student in his
final year at the Lyce Carnot. 2 It was in the salon of Marcel Mor
7S
where he probably first came into contact with a new presentation

of Nietzsche. Some of these philosophers later became his teachers at
the University of the Sorbonne. Deleuze completed his diploma , the
DES (diplme d'tudes sup'ieures) on Hume under the direction of
Jean Hyppolite and Georges Canguilhem. Deleuze's investigations
on Hunle were later published in a book entitled Empiricism and
Subjectivity (1953) , which Deleuze dedicated 1'0 Jean Hyppolite in
sincere and respectful homage'. Though being a Hegelian hirnself,
Jean Hyppolite encouraged diversity in philosophical investigations
and thereby allowed a whole generation of students (Foucault and
Derrida among others) to deviate fron1 the French fficial' philosophy , which was very much influenced by Hegelian dialectic. 3 his
new generation developed a resistance to and increased criticisn1 of
Hegel's philosophy. he study of Nietzsche played an important part
in the rejection of the dialectical tradition. Dialectics had become
discredited as a n1ethod of thought , for it aimed to achieve sornething positive by means of contradiction and negation. Contrary to
the negative power of the dialectic , Nietzsche allowed conceiving
a n10vement of thought , which proceeds by an affinuative play of
differences and not by means of contradiction.
Deleuze's small volume on Nietzsche is indicative of ths new
reading of Nietzsche. Nevertheless , the Nietzsche it presents is unique
to Deleuze , being more a transformation or extension of Nietzsche's
thought than a 'faithf interpretation or comn1entary. Deleuze's
Nietzsche is important for this study mainly for two reasons:
1. Deleuze presents Nietzsche as an 'inverted' Kantian , who pushes
the critique of traditional n1etaphysics to its limits by bringing it
to bear on Kant's critical philosophy itself. Nietzsche does not
acceptthe facts of reason' (knowledge and morality)4 and instead
deluands a genetic account of the values of knowledge and morality , indeed the value of truth itself.
2. Nietzsche's search for the genetic element , which generates the
value of values , culminates in the theory of the wiU to power'.
Deleuze appropriates this theory and purges it alrrlost entirely of
anthropon10rphic terms. Central to Deleuze's interpretation is the
assumption of a (non-dialectica l) differential and genetic structure
of active and reactive forces , which are distinguished by their
quality , or di erence in quantity. We will argue that Nietzsche's
will to power is the pre-form of a productive transcendental principle , which is luobile , plastic and changing.
76

NIETZSCHE AND THE PROBLEM WITH TRUTH
According to Deleuze , one of Nietzsche's greatest achievements is

having called the affinity between thought and truth into question. 5
Thought is rather an ally of life or life forces to the extent that life
poses problems , which set thought into motion , and thought shatters
restrictions that are a hindrance to the exertion of life forces.
Life would be the active force of thought , but thought would be the
affirmative power of life. Both would go in the same direction , carrying
each other along , smashing restrictions , matching each other step for
step , in a burst of unparalleled creativity. Thinking would then mean
discovering venting new possibilities of life. (NP 10 1/115)
Thus thought and life both enter into a mutually enhancing
relation , in which life makes thought activand thought in its turn
invents new modes of being. By contrast, the search for truth , if truth
is defined as a metaphysical value , is disconnected frorn the impetus
of life forces. According to the metaphysical concept of truth , truth
is conceived as something transcendent , invariant and univers that
is an unconditioned given for all tin1es and places. As such , truth is
placed outside the world , out of reach with regard to perspectival
assesslnents , practical den1ands and transforn1ative factors active
within life. Furthermore , in the name of absolute truth , life is devalued and accused. The world we are living in is judged as erroneous
as mere appearance , while the truth lies beyond the sensible world in
a transcendent realm.
Nietzsche poses the question: 'w
r
b
at in us really wants "tn
1th
Searching for the origin 0f the wiU totruth
he wonders: who wants
truth in the first place? What will and what type of forces want truth
and for what reason? His genealogical approach differs considerably
from the traditional concern regarding the essence of truth (That is
trutl'). Nietzsche wants to know what the value of truth is. Why
should truth be more valuable than untruth? y couldn't the
world that concerns us - be a fiction?'7
Not without pride Nietzsche compliments himself on being the
first person to dare to pose this problem. For hir the will to truth
itself becomes problen1atic. Hitherto philosophers naturally claimed
that truth belonged to thought in principle. They refrained from
'relating truth to a concrete wiU of its OWI to a type of forces , to a
quality of the will to power' (NP 95/108). According to Nietzsche
the striving for truth rests on a metaphysical faith or conviction
77
that truth is worth rnore than appearance. This is an element of the

Christian faith , and in fact Plato had already expressed this conviction. 8 It seerrls obvious that the will to truth is not grounded on a
calculus of utility , since appearance , deception , disguise , simulation
or self-blinding have often proven much more favourable to life
than truth. Thus the unconditional will to truth must COlTIe from
another source. Nietzsche argues that it arises on n10ral grounds ,
since it expresses the will not to allow 0 eself to be deceived , or lTIOre
concretely: 1 will not deceive , not even myself!'9 Philosophers are
n1istrustful of this volatile , illusive and sensuous world , which they
conceive as being defcient and guilty. They accuse the world because
they cannot bear being deceived. 'The blind rage with which the
philosophers resist being deceived' is the expression of a will to truth ,
that is the desire for a true world beyond the world of rnere appearances. 10 Philosophers search for truth in 'the lap of Being , the intransitory , the hidden god , the "thing-in-itself" 11 Their n1etaphysical
aspirations lead thelTI to denounce the world they are living in and to
diminish life in the name of higher , superior' values. he will to truth
betrays a nihilistic will , which is hostile to life and destructive with
regard to life's rnost fundamental prerequisites. he ascetic ideal is a
sympton1 of this will to nothingness: it implies a hatred of the material world , a horror of the senses , a fear of happiness and beauty , a
condemnation of appearance and transience , and an annihilation of
wishing and longing. However , we should note that asceticislTI is not
necessarily negative and directed against life. As Nietzsche says in On
the Genealogy of Morals , there is a kind of cheerful asceticism of
an animal become fledged and divine , floating above life rather than
in repose' .12 The ascetic ideal can create the freedom and optimun1
conditions for thinking: the most favourable preconditions of higher
spirituality [Geistig eit]'.13 It is thus indispensable for the philoso
pher or the philosophical an lTIal (l a bte philosophe).14 Moreover , it
had a vital role as a precondition for the existence of philosophers:
according to Nietzsche it served the philosopher as a mask by giving
him the appearance of an already established type of contemplative
ma the priest or religious man in genera l.he phil
78

world' and believed in truth as the highest value. They only differed
in relation to the question whether the real world is attainable , unattainable for now or unattainable in principle. As such , Nietzsche
summarises the occidental history of philosophy in a stunning
short tale , which he calls 'How the "Real World" at Last Becarne a
Myth\15 There he narrates the delusion and final dissolution of the
ideal of the real world. It seems that with the demise of Christian
religion , Kantian rnorality and ultimately the belief in science , the
ideal of the real world loses every function. Nietzsche already foresees a new category of philosophers who doubt the value we ascribe
to truth , truthfulness or altruism. It is a type of philosopher who is
genuinely doubtful and suspicious:
In bourgeois life ever-present suspicion may be considered a sign of bad
character' and hence belong among things imprudent; [. . .] what should
prevent us from being imprudent and saying: a philosopher has nothing
less than a right to 'bad character,' as the being who has so far always
been fooled best on earth; he has a duty to suspicion today , to squint
maliciously out of every abyss of suspicion. 16
The philosopher of the future dares to pose the problem of ralues
and put the hierarchy of values into question. However , a proper
critique will not slnply lead to a destruction of established values
or a mere reversal of the existing hierarchy of values. It is true that
Nietzsche opts for re-evaluating appearance , the will to illusion ,
egoisn1 and desire , but these values do not have an intrinsic meaning
either. In order to detern1ine the nleaning of values , the philosopher
needs to investigate the conditions and circumstances under which
values grew up and developed. He has to ask who makes this
meaning?' Deleuze calls is Nietzschean method the nethod of
dramatisatio '. It consists in 'relating a concept to the will to power
in order to lnake it the synlptom of a will without which it could
not even be thought (nor the feeling experienced , nor the action
undertake (NP 78/89). To dramatise a value means to relate it to
the lnouth that utters it and the geohistorical' n1ilieu from which
it originates. The philosopher of the future will have to test values
constantly by evaluating their ongll1 in 'noble' or base' ways of
thinking.
If one wants to object that 'noble' and base' are thenlselves
ralues and why should noble' be worth more than base' e
answer lS: noble' and base' are not values but modes of existence of
those who eraluate and j udge. These ways of being are symptoms or
79
expressions of a particular will to power , which serves as a principle

for a style of life as well as for a rnode of thought.
This is the crucial point; high and low , noble and base , are not values
but represent the differential element from which the value of values
themselves derives. (NP 2/2)
The will to power is precisely this differential element.
ln the following , we will mainly refer to Deleuze's interpretation
of the will to power, which relies heavily on Nietzsche's physics of
force. he theory of forces is only sketchily outlined in the unpublished fragments collected in the posthumous opus The Will to
Power (1 906) , which suffered greatly under the distortions at were
caused by Nietzsche's sister. In the exposition that follows , we do
not atternpt a reading faithful to Nietzsche or consider his unpublished notes and books from a philological point of view. We will
rather present the Nietzsche of Deleuze's early book , Nietzsche and
Philosophy (1 962) , that is 'a rigorous and systematic thinker who
constructed a philosophy of nature around the complex concept of
will to power' .17
NIETZSCHE'S WILL 1'0 POWER
The will to power is by no means a wiU that wants power and

struggles for recognition. lf power is understood as getting oneself
recognised , then the will to power means nothing more than the
confirmation and conservation of current values that serve as criteria
for recognition. Such a wiU to power is conservative , not creative.
lt conforms to the rnodel of representation , for it makes power an
object of representation: power is represented in the 1110ney you have ,
the luxuries you enjoy, the honours you are endowed with or the
reputation you hold.
he
mania for representir for being represented , for getting oneself

represented , for hring representatives and representeds: this is the mania
that is common to all slaves [. . .] The notion of representation poisons
philosophy: it is the direct product of the slave and of the relations
between slaves , it constitutes the worst , most mediocre and most base
interpretation of powe r. (NP 81192)
The wiU to power , understood as striving for recognition and

having power represented , is the conception of the slave. lt is important to note that 'ITlaster' and slave' do not designate sociological or
80

historical types of hllrnan beings. Rather they are llsed as epitomes of
high or low , noble or base , ways of thinking and being; in Deleuze's
words we rnight say that the slave and the lnaster or aristocrat
fllnction as conceptllal personae' .18
The slave is a negative and reactive type , a rnan of ressentiment.
As Nietzsche says: the slave morality needs external stimllli in order
to act at all its action is fllndarnentally reaction\19 The eye of
ressentiment is directed toward the aristocrat or master. The slave
resents the master; he calls the master evil; and by negating him ,
he seeks to affrm himself: YOll are evil , therefore 1 arn good'. His
thinking and reasoning is based on the principle of contradiction and
negation. he syllogism of the slave' (NP 1211139) operates by a
dOllble negation: it begins by stating a negative ('the master is evil')
and by opposing the master to hin1self ( the master is the non-ego').
He assumes that the negation of a negation equals an affrmation.
In fact , the slave is a covert Hegelian'; he follows the logic of the
dialectic. 20
On the contrary , the master is an affrmative and active type. He
is aware of himself as an 'overflowing power' which likes to bestow
and share his wealth , not so much out of pity but rather out of an
excess of power'.21 The master labels himself as good. He actually
creates vallles. These vallles are not negatives of already existing
values. He creates values by determining what is good for him , that
is what enhances his power and his capacity to act. By contrast , he
labels bad' what diminishes his power and tries to separate it from
what it can do. he slave is 'bad' in the ralue-positing eye of the
lnaster. However , bad' in the mouth of the lnaster is by no means
the sarne as evil' in the 1I1 011th of the slave. As Nietzsche explains:
This bad' of noble origin and that evil' out of the cauldron of unsatisfied
hatred - the former an after-production , a side issue , a contrasting shade ,
the latter on the contrary the original thing , the beginning , the distinctive
deed in the conception of a slave morality - how different these words
bad'and evi l' are. 22
ThllS the slave creates the highly n10ralised notion of 'evil which
is born Ollt of resentment and hatred , while the n1aster only comes to
this bad' of noble origin after a double affnnation: frst he a rms
himself, then he affrn1s his difference from the slave.he negative
conclusion ('the slave is bad') is only the con1plementary shadow ,
which follows 'frOITl a triumphant affrmation of itself'. 23
'e should note that the n1aster says not only 'yes' to hin1self
81
but to the whole cornrnunity of rnasters: We noble ones , we good ,

beautiful happy ones.' he rnasters are capable of developing a
communal feeling' which is based on a free play or competition of
equal forces (agon).24 They respect each other and also respect their
enernies , for they regard someone as an enemy only if he is wort?
that is equal in power. Thus the logic of the rnaster is a logic of affirmation and difference. he negative , contradiction and opposition is
of no irrlportance in this logic.
Characterised in this way , 'master' and slave' do not, as we have
already mentioned , correspond to real characters , bllt rather to ideal
types. According to Deleuze's reading of Nietzsche, this distinction of
master and slave amollnts to a typology of forces. We have to reject a
simplistic psychological or anthropomorphic interpretation.
Here we must rid ourselves of all 'personalis t' references. The one that . . .
does not refer to an individual , to a person , but rather to an event , that
is , to the forces in their various relationships in a proposition or a phenomeno and to the genetic relationship which determines these forces
(power). (NP xi)
The master stands for an active type of forces , that is forces of
conquest and subjugation , while the slave stands for a reactive
type of forces , that is forces of adaptation and reglllation. 25 Active
and reactive forces do not simply differ with respect to qllantity;
rather they differ with respect to qllality. Quality , so Deleuze says ,
is irreducible to qllantity.26 Qllality can be defined as a dierence ol
quantity in relation to other forces. ha t is to sa y , a force is never
isolated; a force is always in a relation of tension to all other forces.
This relationship of difference with other forces makes each force
singlllar , such that no two forces are eqlla l. 'Difference in quantity is
the essence of force and of the relation of force to force' (NP 43/49).
Already in Beyond Good and Evil it becornes clear thatNietzsche
rejects the theory of materialistic atomiSlTI and instead plltS forward
a physics of dynamic quanta or power-quanta , which he largely
borrows from Boscovich. Roger ]oseph Boscovich (1711-87) was a
matherrlatician and physicist born in Ragusa , Daln1atia (present-day
Dubrovnik in Croatia). He is the author of Philosophiae naturalis
theoria (Theory ol Natural Philosophy , 1758) , in which he develops
a dynarnic interpretation of the world according to which phenorn
ena are explained in terms of force instead of mass.Nietzsche cites
hin1 as the one who won the greatest triumph over the senses by
renouncing 'the belief in the last part of the earth thatstood fast"
82

belief in "substance" , in "rnatter" , in the earth-residuum , and
By fo lI owing Boscovich , Nietzsche does not want
to claim that our material world is a mere delusion , appearance or
representation. Instead , he intends to find the genetic principle of it
aII , a more rudimentary world of forces and affects , which allows
organic forms and functions to develop and differentiate. He assumes
a 'pre-form of life' , that is a kind of instinctive life' , in which a wiII
has an effect upon another wiII , such that all mechanical events can
subsequently be interpreted as effects of the wil 1. 28 Denying continu
ity between forces and causality as a mechanistic principle , Nietzsche
explains the relation between forces as the effect of a wiII to power
which acts at a distance. The principle of action at a distance (actio
in distans) lneans that the will to power does not operate as a mediator of their interaction; rather the will to power is the difference or
distance which brings about the differentiation of forces separated
m space.
the
particle atom'.27
Suppose lally we succeeded in explaining our entire instinctive life as

the development and ramification of one basic form of the will- namely ,
of the will to power, as my proposition has it; suppose all organic functions could be traced back to this will to power and one could also find in
it the solution of the problem of procreation and nourishment it is one
problem - then one would have gained the right to determine all ecient
force univocally as will to owe1 The world viewed from inside , the
1Vorld defined and determined according to its ntelligible character' it
would be 'will to power' and nothing else. 29
Hence , according to this ontology there are two ways of regarding

the same world. The world as we know it consists of organic processes and definite entities (subjects and objects). The inner world' ,
or the world on a micro-level , however , is nothing but the wiII to
power , that is a differential complex of forces and discharges of
force endowed with the potential of creation. The wi lI to power is
never unitary but multiple. According to Nietzsche , 'there are treaty
drafts of wi lI (Willens..Pun tuationen) that are constantly increasing
and losing their power'. 30 The composition of forces changes and
new 'centres' and 'nodes' are formed. A new arrangement of forces
s achieved by the interference of chance that brings new forces into
relation and the effect of the will to power which is the differentiating
and determining principle of this relation (c f. NP 53/60). The will to
power is a necessary addition to the concept of force. According to
Deleuze , there needs to be an 'internal will' (NP 5 1/57) , an internal
83
elernent withollt which the relations of forces would rernain indetern1inate. One of the key qllotations that Deleuze considers to be
essential for the interpretation of the wiU to power is the following:
he
victorious concept orce' by means of which our physicists have

created God and the world , still needs to be completed: an inner will must
be ascribed to it, which 1 designate as will to power' , i.e. , as an insatiable
desire to manifest power; or as the employment and exercise of power , as
r1 ~;"
a creatlve dnve
, etc. 31
t-r-
However , the crllcial part 'an inner will mllst be ascribed to it' is
based on an error , that is a correction' that was made by Peter Gast
who together with Nietzsche's sister Elisabeth F rster-Nietzsche
prepared Nietzsche's handwritings for the print version. The original
line goes 'an inner world must be ascribed to it'. Deleuze cOllld not
have known of this error , since he llsed the French translation by
Genevi ve Bianqllis (1935 and 1937) , which relies on the Musarion
edition , in which this error was not yet changed back to the original
line. 32 However , we do not believe that Delellze's argun1ent depends
substantially upon this typographical erro r. The n1ain statelnent
rernains that the world viewed fron1 inside' is nothing bllt the will
to power. 33
Delellze's interest in Nietzsche's will to power arises in the
context of the philosophical project of a radical critique. Taking
IIp Nietzsche's method of dramati tion Delellze wants to relate
thOllght , that is our concepts and values , to the type of forces that
stipulate their lTleaning. To dramatise a concept or value means to
pose the question Tho speaks?' , that is to relate the concept or value
to a will to powe r. It is important to grasp that in Deleuze's reading
of Nietzsche , the will to power is not a personal will , but an internal>
genetic element which is capable of detennining the relations of
forces from a double point of view: the reciprocal genesis of their
difference in quantity and the absolute genesis of their respective
qualities' (NP 51157). Deleuze draws on n1athematical analysis and
the forn1ulation of the differential calculus to explain the issue. The
will to power
is added to force as the internal principle of the determination of its [the
force's] quality in a relation (x dx) and as the internal principle of the
quantitative determination of this relation itself (dy/dx). (NP 51158)
The rnathematical symbols dy and dx represent 'differentials
that is infinitely small vallles on the point of vanishng. The first
84

important point to notice is that the elernents dx and dy do not represent quantities with assignable values , since they are smaller than
any assignable value. Now , Deleuze describes two kinds of relations.
(1) Given an arbitrary quantity x the synthesis of a differential dx is
possible , which is in a variable relation to x. The synthesis of x and
dx represents a qualit} that is a difference of quantities that is not
measurable. (2) The relation between the differentials dx and dy is
slightly different: it relates two infinitely small variable quantities ,
which cannot be rneasured and are thus completely indeterminate
when considered in isolation. Yet , through their relation (dy/dx)
their values are reciprocally determined. Furthermore , it is important
to note that the relations between dx and dy , and between dx and
the variable x , are not relations which are external to their terms.
Instead , the relations and elernents are inseparable from each other.
In other words , the relations are internal , that is included in each
element. The crucial point , however , lies in the generative power of
differentials. Given the ratio dy/dx , there is a procedure (called integratio which allows us to generate the primitive function , that is
the function fron1 which the ratio dy/dx is differentiated. This process
of generation can be interpreted as involring a change of order: from
the sub-representational differential order to the order of actual
quantities and their relations , which are represented by the primitive
function. Deleuze sees in the rnathe l11atical procedure of integration
(the generation of primitive functions fr0 l11 differentials) a model to
explain the nature of the transcendental in philosophy. He will conceive a kind of transcendental function' , which by virtue of the reciprocal determination of its sub-representational differential elements ,
generates thought (concepts and values) and real experience alike. 34
This
w i11 be
t
h1
e
ft
uIn
d~a
a
ln1
n1e
1
1e
enta
t
al theme in Di:
P
where Deleuze presents his transcender
1tal em
1P1rlC1sm. However , we
can say that in Nietzsche and Philosophy , Deleuze's demand for a
transcendental genetic principle already finds an early expression in
his reading of Nietzsche's will to power. It is true that in Difference
and Repetition , the model of differential calculus t
85
an inner , genealogical and differential element. Where force acts

upon force , there is wiU to powe r.
This is what the iVill to power is; the genealogical element of force , both
differential and genetic. The will to power is the element from iVhich
derive both the quantitative difference of related forces and the quality
that devolves into each force in this relation. The will to power here
reveals its nature as the principle of the synthesis of forces. In this synthesis
- which relates to time - forces pass through the same differences again or
diversity is reproduced. (NP 50/56)35
We have not yet encountered the relation to time which is mentioned here. According to Deleuze , the temporal dimension is a
necessary complement to the will to powe r. Using Nietzschean
tern1inology , Deleuze equates this temporal dimension with the
'eternal return': the eternal return is a principle of selection , which
puts the differential relations of forces to the test of time. Certain
constellations of forces are fit to return , while others are expelled by
the selection of the eternal return. What comes back is that which
affirms difference and becoming; what is doomed to disappear is
that which contradicts affirmation , that is nihilistic forn1s of the
will to power , forces of reaction. The eternal return carries out
the synthesis of forces and ensures he reproduction of diversity at
the heart of synthesis' (NP 52/58) , while the will to power is the
differential and genetic principle of the synthesis of forces. They
are both linked together like Nietzsche's divine couple Ariadne/
Dionysus. 36 he will to power is Dionysian becoming: it is the first
affirmation , that is the creation of new values and possibilities of
life , which render life light and active. But this first affirmation is
in need of a second affirmation; that is to say , only as the object
of a second affirmation can becoming affirm its being: Dionysian
becorrl ng is being, eternity , but only insofar as the corresponding
affirmation is itself affirmed' (NP 187/215). Ariadne is this second
affirmation. As Nietzsche says: 'That everything recurs is the closest
approximation o{' a world o{' becoming to a world o{' being'37 - this
eternal recurrence is the wedding ring of Ariadne. lt should be noted
that the eternal return does not mean the reproduction of the same ,
but of diversity , of that which differs fror11 itself and from other
forces (NP 46/53). Identi
In other words , identity in the eternal return does not describe the nature
of that iVhich returns but , on the contrary , the ct of returning for that
86
The Demand fo 1' T 1'anscendental Genetic Conditions

which differs. This is why the eternal return must be thought of as a synthesis; a synthesis of time and its dimensions , a synthesis of diversity and
its reproduction , a synthesis of becoming and the being which is affrmed
in becoming, a synthesis of double affrmation. (NP 48/55)
Much more can certainly be said on Deleuze's interpretation of
the eternal return. But since the then1e of the eternal return will be
taken up in Chapter 4 of this book , we iVi lI now focus here on the
surprising formula of the will to powe1' as a p 1'inciple of the synthesis
of fo 1' ces. Deleuze's explanation of the will to power as a principle
of the synthesis of forces is a decisive move which puts Nietzsche
explicitly in line with Kantianisn1. In fact , Deleuze believes
that there is in Nietzsche , not only a Kantian heritage , but a half-avowed ,
half-hidden , riralry [. . .]. Nietzsche seems to have thought (and to have
found in the eternal return' and the \;v ill to power') a radical transformation of Kantianism , a re-invention of the critique which Kant betrayed at
the same time as he conceived a resumption of the critical project on a
new basis and with new concepts. (NP 52/59)
In the following section , we will shed light on this alleged
Kantianisn1 in Nietzsche. We will examine how Nietzsche overturned
the Kantian account of transcendental conditions , which define
what counts as an object for us and thus bestow sense or signification. Against the Kantian account , which remains one of external
conditioning , Nietzsche proposes a principle of internal genesis.
Nietzsche's will to power generates the particular senses and values
of phenomena by relating theln to a complex of forces , to vital
dynamisms immanent to the world.
FIRST TRACES OF A NEW TRANSCENDENTAL
In the rst edition of the C1'itique of Pu 1'e Reason , Kant describes

three syntheses , which constitute the possibility of experience: the
synthesis of apprehension in the intuition , the synthesis of reproduction in the Inagination and the synthesis of recognition in the
concept.hese syntheses all culrninate in the third synthesis of
recognition , which yields the cognition of the forn1 of an unspecified
object as a correlate of the fonnal unity of consciousness ('1 think').
The task of the a priori concepts of the understanding is to bring
about synthetic unity in the rnanifold of intuition. Kant's transcendental concepts have thus the sarrle function of providing unity as do
Aristotle's categories in the realn1 of logi c:
87
The same function that gives unity to the different representations in

a judgment also gives unity to the mere synthesis of different representations in an intuition , which , expressed generally , is called the pure
concept of the understanding. (CPR A 79/B 105)
However , since Kantian categories seek to capture the world of experience , Kant finds himself confronted with a particular difficulty: the
manifold of intuition cannot be brought directly under the synthetic
unity a priori. The synthetic unity a priori only applies to a unity
of pure intuition (a pure unity of space and time). he intermediary of pure intuition is supposed to n1ediate between the particular
(real experience) and the abstract , transcendental concept of an
unspecified object , constitutive of the possibility of experience.
It was the achievement of the post-Kantians to point out that the
Kantian transcendental concepts cannot be applied to the diversity
of real experience (neither directly nor through any 111ediation).
Nietzsche is a post-Kantian in this sense. He criticises principles for
being too large or too abstract in relation to what they condition. For
instance , he condemns Schopenhauer's metaphysical will' , which
is outside the order of space and time and which expresses itself in
every phen0111enOn according to different degrees of objectication.
This will is 'a mere empty word' because 'one has eliminated the
character of the N"ill by subtracting from it its content'. 38 Equal the
Kantian transcendental concepts are too abstract: the signification of
the forn1 of an unspecified object cannot be related to the senses and
values of particular experiences. Contrary to Kant's transcendental
conditions , Nietzsche's wiU to power' clings to real experience. he
wi I1 to power wants this relation of forces , this quality of forces.
This is why the wiU to power is not larger than what it conditions.
Nietzsche introduces the concept of plasticity in order to express the
intimate relation between the condition and the conditioned , which
is one of internal genesis. The important thing to grasp is that the
condition itself does not remain unaffected by what it conditions. In
other words , it is determined at the same time that it determines. The
will to power varies with the relations of forces that it produces and
can therefore be ca l1 ed an immanent and 'plastic' principle:
As plastic element it simultaneously determines and is determined ,
simultaneously qualies and is qualifed. 'hat the wiU to power wills is
a particular relation of forces , a particular quality of forces. And also a
particular quality of power: affrming or denying. his complex , which
varies in every case , forms a type to which given phenomena correspond.
88
The Demand for
Tllscendental
Genetic Conditions
All phenomena express relations of forces , qualities of forces and of

power , nuances of these qualities , in short , a type of force and will. (NP
85/97)
Depending on the particular quality of the synthesis of forces that

the will to power effectuates , the will to power is either affirmative
or negative. This means that the synthesis of forces can reslllt in a
configllration in which the active forces prevail over the reactive
forces , or else one in which the reactive forces forrn units and prevent
the active forces from acting Ollt. he qllality of the first configllration is affirmative , whereas the trillmph of the reactive forces is the
symptom of a negative or nihilistic will to powe r.
Another essential characteristic apart from the plasticity of the
will to power is its creativity. The will to power produces the
sense and value of phenomena. ln other words , the will to power
interprets' it bestows a fresh interpretation by irnposing a new
pllrpose hrollgh which any previous "meaning" or "pllrpose" are
necessarily obscllred or even obliterated':
Purposes and utilities are only signs that a will to power has become
master of something less powerful and imposed upon it the character
[Sillll] of a function; and the entire history of a thing ,' an organ , a custom
can in this way be a continuous sign-chain of ever new interpretations
and adaptations , whose causes do not even have to be related to one
another bu on the contrary , in some cases succeed and alternate with
one another in a purely chance fashion. 39
Things are signs or symptoms of relations of forces. Their history can
be read as a 'continllolls sign-chain' whereby the variolls chain links
are the reslllts of combats' between forces. These combats shollld
not be read as combats-against , bllt rather combats-between forces. 40
These are processes throllgh which forces augment their power by
a ppropriating other forces , which agree with them , thereby forming
a new con1position or intensive compound of forces. here is no final
end , no ultimate goal to sllch processes; it is an infinite becoming.
The evollltion of a thing is therefore not a progress towards a goal
bllt a sllccession of processes of attempted transfonnation , resistance
and appropriation inflicted on the thing. 'The form is fluid ut the
"n1eaning" [Sinn] is eren n10re so. 1
Hence the true essence of a thing is neither a transcendent intelligible being nor an abstract signi f1 cation constitllted by external
transcendental conditions , bllt rather the outcome of an infinite
process of sense-roduction. Truth is the limit object of a production
89
of sense. We should mention , however , that the characterisation

of truth as the lmit of a genetic series , which constitutes sense ,
is already an addition that Deleuze introduces in Dierence and
Repetition and which is a result of his involvement with differential
calculus as a technical model for internal genesis and creation (cf.
DR 154/200). However , the concept of lirnit provides a helpful tool
for the interpretation of Nietzsche , who denes truth as an infinite
process and an active detennination , that is as creation:
6ruth' is therefore not something there , that might be found or discovered
- but something that must be created and that gives a name to a process ,
or rather to a will to overcome that has in itself no end
We have nally con1e to the point where we can consider

Deleuze's thesis of a peculiar Kantianisln in Nietzsche. he Kantian
transcendental conditions are supposed to function as principles of
synthesis bringing unity to the manifold of intuition. The relation of
transcendental concepts to intuition is essential , because otherwise
the pure concepts would remain entirely empty, that is without
content and hence without signification. However , the signification
they constitute remains rather abstract , nalnely the signification of a
possible object of experience. Now , Nietzsche's will to power is supposed to function also as a principle of synthesis. However , the synthesis is not between concepts and intuition , but between active and
reactive forces. The will to power is a creative principle: it creates the
particular senses and values of phenornena. It is therefore a genetic
principle of real experience , not of possible experience. The differences between Kant's and Nietzsche's accounts are perfectly obvious.
While Kant adheres to a model of external conditioning, Nietzsche
allows for a model of internal genesis. 'hile Kant's transcendental
conditions are abstract , universal and invariant , Nietzsche's will to
power is plastic and changing. According to Deleuze , the will to
power is 'a good principle':
his
is because it is an essentially plastic principle that is no wider than

what it conditions , that changes itself with the conditioned and determines itself in each case along with what it determines. (NP 50/57)
he
principle of the Nill to power consttutes a 'superior en1piricisn1'

(NP 50/57). Tl1er
efo)re
1 t deserves tob
e called 'transcendental' pro
V
rided tha
t we accept a change in tlle n1e
er
d
1taa
l'\. Bu
1t WNh
1y shou
1ld Kan
1t have the final definition and exclusive

authority over his invention of the transceI
90
Deleuze , philosophical concepts are usurped and appropriated time

and again in the history of philosophy. In fact
the history of philosophy is completely without interest if it does not
undertake to awaken a dormant concept and to play it again on a new
stage , even if this comes at the price of turning it against itsel f. (WP 83/81)
In Difference and Repetition , Deleuze will call his philosophy a
transcendental empiricism' , and without doubt his transcendental
incorporates the characteristics of Nietzsche's plastic principle of the

will to powe r. I-I owever , in Nietzsche and Philosophy , Deleuze still
draws back from this appropriation of the term transcendental'. He
says that when he compared the will to power with a transcendental
principle , his aim was to point out the difference from psychological
determinations (NP 9 1/104). 'Nevertheless , in Nietzsche , principles
are never transcendental' (NP 9 1/ 104).his rneans that there are
irreducible differences between Nietzschean and Kantian principles.
Yet , it can be said that Nietzsche paved the way for Deleuze's
transcendental , which will be further explored in the course of this
book.
As a final note to the first part of this chapter , we would like to point
out two features that can be regarded as unfortunate with respect to
Nietzsche's concept of the will to powe r. The first is expressed by
the temptation to interpret the will to power in tern1S of a personal ,
psychological wil l. Although Deleuze and , indeed , Nietzsche himself
provide many reasons to abstain fron1 this interpretation , there is still
a need to clarify the objectivity' of the will to power as a transcendental and genetic principle. Deleuze wiU find an adequate solution
in the Maimonian term of differential Ideas , which are defined as
'objectivities' .
The second difficulty with the concept of the will to power concerns the question of its status with regard to our conscious representations (concepts , senses and values of things). It seems that the
will to power is necessarily sub-representative: it designates the 'inner
world' of the world we see , relnember , imagine and conceptualise.
However , what is stilllacking is an adequate expression to mark the
heterogeneity and distinctness of the will to power from the world of
actual objects and common ralues. In this regard , Bergson's concept
of the virtual wi I1 provide the terminological ITleanS to describe
the 'world as seen fr0111 inside In connection with the concept of
Idea , we will thus gain the concept of virtual Idea which Deleuze
91
will introduce as differential and genetic principles for the sense of

phenomena , for values and concepts of thought.
aimon's
Essay on Transcendental Philosophy
Deleuze's an1bition to substitute for Kant's account of external

conditioning a model of internal genesis whose conditions are not
larger than what they condition led him to appreciate the philosophy of Salomon Maimon (1753 1800). Mairnon was the first
philosopher 'to pass from a transcendental philosophy to a genetic
one'.43 Unfortunately 1aimon has been a ra ther rnarginalised
gure in philosophy and his extensive philosophical work , which
comprises several books , comn1entaries and journal articles , has
been unduly neglected. Although Main1on's astute mind and talents
were acknowledged by prominent philosophers of his day (e.g. Kant
and , in particular , Fichte) , he never quite received the attention that
he deserved One reason might be the prevalent anti-Sernitisln
in his time. Given his social background as an Eastern European
]ew born into an impoverished farnily and his unusual career as a
seltaught philosophe Maimon had to face n1any prej udices A
further reason for being unduly neglected Inight be the intricacy and
obscurity of his philosophical thought , which at least in part resulted
from his unusual style of writing. As Gideon Freudenthal points
out, Maimon was educated in the alnludic tradition and was used
to writing commentaries , a genre which essentially differs fron1 the
systematic form of philosophical discourse comn1on in the western
world Salnuel Atlas suggests that the grandeur and splendour of
Fichte , Schelling and Hegel and their metaphysical systems simply
overshadowed Main1on's philosophical investigations. 47 For these
and lnaybe other reasons , Maimon has not received appropriate
attention until this day.
Deleuze is one of the few philosophers who did recognise
Maimon's achievements , although he does not dwell extensively on
Maimon - in Dirence and Repetition he only devotes a few pages
to him. However , Maimon's name is n1entioned explicitly in rnany
of Deleuze's books , essays and selnnars , and traces of Ma lnon's
thought can be found throughout Deleuze's work Deleuze takes a
particular interest in l\1aimon's Essay on Transcendental Philosophy
(1790) , where 1amon laid out his criticism of Kant's transcendental philosophy and suggestions as to how to irnprove it. Rejecting
the Kantian viewpoint of external conditionir Main10n develops
92

a rrlO del of internal genesis based on the concept of differentials' ,
which he defined as Ideas of the understanding'. Maimon's theory of
differentials is inspired by the mathematical model of differential calculus. Deleuze shares Maimon's interest in the differential calculus ,
and he acknowledges that by means of the model of differential relations , a new concept of difference and of internal relation is thought.
Maimon developed the concept of 'intrinsic difference' (already
adumbrated in Leibniz) , that is difference as a positive , internal
quality , which does not merely result from external limitation or
determination This intrinsic difference is fundamental insofar
as any difference (spatio-temporal , quantitative or qualitative) can
be traced back to this kind of difference. As Nick Midgley says ,
1airnon is committed to a Ul110city of difference'. 50 o this extent ,
Maimon already paved the way for a philosophy of difference , where
difference is conceived as the sufcient reason or genetic (material)
principle of the rea l. 51
Surprisingly , there hardly exists any scholarship on the MaimonDeleuze nexus. In the last two years , only a few book chapters on this
topic have been published by Graham Jones , Daniel Srrl th and Beth
Lord respectively.52 In addition , the Ma l11on-Deleuze nexus is discussed briefly in the three 111onographs on Deleuze's transcendental
en1piricisn1 authored by Levi R. Bryant , by Christian Kerslake and by
Anne Sauvagnargues. 53
1AIMON'S CHALLENGE TO THE KANTIAN CATEGORIES
Mairrlon's rethinking of Kantian transcendental philosophy does not

aim to disn1antle it , but on the contrary , to improve it by introducing
Leibnizian elements. Maimon seeks to overcome the Kantian dualities between understanding and sensibility , concept and intuition ,
and form and matte r. Naturally , Kant himself could not approve of
this project, which undermines his basic tenet of the heterogeneity
of the two sources of human knowledge , receptive sensibility and
active understanding , a tenet that he basically intended to ward off
dogmatic idealism frol11 his transcendental idealisn1. In 'h e Critique
f Pure Reason Kant issued an attack against Lei niz for trea ting
sensibility and understanding as a single cognitive faculty that is
distinguished only in terms of the clarity of the representation. 54
Leibniz , he said , 'conceded to sensibility no kind of intuition of its
own , but rather sought everything in the understanding , even the
empirical representation of objects , and left nothing for the senses
93
but the contemptible occupation of confusing and upsetting the

representations of the former' (CPR A 276/B 332). Thus sensibility ,
for Leibniz , is just an inferior confused mode of cognition and not a
special source of representations (CPR A 2711B 326). Consequently ,
Leibniz allowed only conceptual differences between things. While
for Kant space and time are irreducible forms of sensible intuition ,
Leibniz defined them as 'outer' relations that have their ground in
something absolutely internal (CPR A 285/B 341). hat is , space and
time must be reducible to inner conceptual detenninations contained
in the concepts of purely intelligible things. These intelligible things ,
or things in thenlselves , are entirely thought without any schema of
sensibility (CP R A 286/B 342). Only a sort of non-sensible or intellectual intuition , that is an entirely different intuition and an entirely
different understanding than our own' (CPR A 287/B 344) can have
this immediate access to things in themselves. Although Kant does
not state it explicitly at this point , he refers to the notion of an intuitive intellect, which does not require 'schernas' or 'sensible images'
of space and time in order to represent objects. Instead , an intuitive
intellect entertains a direct relationship with objects , in the sense that
the very act of intuiting produces them at the same time (CPR B 139 ,
B 145).
After having read the Essay on Transcendental Philosophy , Kant
responded to Maimon's criticism with the accusation at Maimon
assumed precisely such an intuitive intellect as he originator
[Urheber] not only of sensible forms but also of their matter , i.e.
of objects' .55 For Kant , the idea of an intuitive intellect is absurd ,
since we are acquainted with no sort of intuition other than our
own sensible one and no other sort of concepts than the categories'
(CPR A 287/B 343). Kant insists that our human understanding s
discursive and not intuitive. It relies on something given to the senses ,
that is a spatio-temporal manifold of intuition to which it can relate
its categories. If we abstracted from sensibility , that is the manner in
which objects are given to u then our concepts of understanding
would have no relation at all to any sort of object. 'Without the data
of sensibility they would be merely subjective forms of the unity of
the understanding , but without any object' (CPR A 287/B 343). For
Ka
94
genera l. In Kan t:, this transcendental synthesis of apperception is the

supreme principle of all use of our understanding.
For Mairnon , on the contrary , the conjunction of form and matter ,
or concept and intuition , in an a priori synthesis remains incornprehensible , given that they stem from completely independent and
heterogeneous faculties. His sceptical challenge can be summarised
in his doubts about Kant's solution to the question quid juris?: by
what right do a priori concepts apply to something given in intuition?
Kant addresses this problem in the Transcendental Deduction ,
as well as in his chapter on the Schematism of the pure concepts
of the understanding. In a rst step (CP R ~ ~ 15-21) he tries to prove
the objective validity of the categories , that is their being necessary
conditions for the possibility of objects in genera 1. 56 Kant argues that
the categories are objectively valid because without them 1 would not
have cognition of any objects at all , not even of lUy own mental state.
In a second step (CPR ~~24-6) Kant goes on to prove the objective
reality of the categories. hat is to say , Kant explains how it is possible that a priori concepts necessarily relate to intuition. He argues
that a priori concepts are not applied to empirical intuition directly
(as this would beg the question quid juris?) , but to pure intuition ,
namely space and time. As pure intuition is given a priori , the question quid juris? appears to be solved. Kant concludes that the categories are applicable to actual or en1pirical objects through 'schernas'
r a priori spatio-ten1poral marks of the object. Strictly speaking , the
clain1 of objective reality of the categories includes two claims: (1)
the capacity of pure concepts to be applied to real (i.e. given spatiotempor yet a priori) objects , and (2) through this mediation to
actual objects of experience. 57
Maimon's rejoinder is very complex. He pursues several lines of
attack. The focus of his objections , however , lies on the objective
reality of the categories , whereas he is willing to admit their objective
validity , at least to SOIue extent. In reconstructing Kant's argument
for the objective validity of the categories , Maimon appeals to a
regressive transcendental argument starting from accepted facts of
knowledge , such as the one that Kant presents in the Prolegomena. 58
Unfortunately , he does not take into account Kant's attempt in
the Critique of Pure Reason to put forth a progressive transcendental argument that proceeds fron1 ground to grounded that is
without presupposing the fact of objective experience. The regressive
transcendental argull1ent can be recollstructed as follows: 59
95
1. Without the objective validity of the catego 1' ies , the 1'e would be no
objective expe 1'ience (that is a unive 1' sal and necessa 1' y connection
between distinct 1'epresentations) fo 1' us.
2. But objective expe1' ience does exist.
3. So , the catego1'ies of the unde 1' standing are objectively valid. 60
Since MairTl on doubts the second p 1'emise , he cannot concede the
objective validity of the catego 1'ies. Thus he only g 1'ants a hypothetical validity: the catego 1'ies a 1'e objectively valid if the fact of synthetic
a p 1' io 1'i j udgements is t 1' ue.
Maimon's qualms , howeve 1', begin with Kant's claim of the objective reality of the categories. 61 MaInon can be said to pu 1' sue an
exte 1' na l' and an 'inte 1' na l' app 1'oach to the p 1' oblem. he exte1' nal
app 1'oach consists of taking sides with Hume and doubting that
the 1'e a 1'e any facts of objective expe 1'ience' defined as a unive 1' sal
and necessary inte 1'connection between 1'ep 1'esentations. At least , we
cannot know of any such fact , since we have no justifiable c1'ite 1' ion
with which to dete 1' mine whether a given manifold can be thought
in a synthetic a p 1'io 1'i unity , stillless a criterion in which unity this
manifold could be thought. 62 That is to say , we lack any a p 1'io 1'i 0 1'
a poste 1'io 1'i crite 1'ion. For instance , in 1'esponse to the question to
which case does the synthetic a p 1'io 1' i concept of callsality apply , we
a 1'e unable to p 1' ovide an a p 1'io 1' i c1' ite1' ion. It might appea 1' that the
irreversibility of a temporal sequence can provide a criterion for its
objectivity , since in this case the sllccession of 1'ep 1'esentations enslles
in accordance with a necessa 1'y 1' llle. But 'irreve1' sibility' , as Maimon
a 1'gues , cannot be detected eithe 1' a p 1' io 1'i or a poste1' iori. Mere pe1'
ception leaves the question of an objective , i.e. irreve 1' sible , o 1'de 1'
between the successive 1'epresentations undete 1' mined , since all ou 1'
1' epresentations (subjective 0 1' objective as they may be) a 1' e equally
apprehended in a successive o 1'de 1' of befo 1'e and afte 1', and the 1'e is
no means to decide which of them is just subjective and which of
theln objective. hat is to say , we cannot go back in time and p 1'ove
that the sllccession of ollr rep 1'esentations was me 1'ely arbit1'a 1'Y and
could have been rep 1'esented in a reve 1'sed order , although we might
very well magine sllch a thing. Bllt 'then this reversed o 1' der of SllC
cession mllst happen in another time than the preceding one so that
in each tme only one kind of succession can be actual' .63 Thus the
irreversibility of a sequence , i.e. the clain1 that the reversed o 1' de 1'
of succession is impossible , can neve 1' be shown with 1'ega 1' d to a
96
The Demand fr Transcendental Genetic Conditions

particular perception. As Hume has argued , there can in principle be
no inference from the subjective order of sensations to an objective
order of states of the world. his means , however , that the evidence
of our sensory perceptions can testify at most to a constant conjunction between distinct representations , but never to a universal and
necessary interconnection. 64 Maimon concludes that we do not have
any criterion for deciding whether a particular succession of representations occurs in conformity with a necessary rule. Furthermore ,
even if we could somehow make out an objective , irreversible order
of succession , we would not know whether the determining rule is
the a priori concept of causality which prescribes a unilateral determination (a relation of subordination) or , for instance , the a priori
concept of con1munity which prescribes a reciprocal determination
(a relation of coordination). That is to say , it would ren1ain undeterrrl ned under which concept the rnanifold of representations can be
brought.
However , this line of attack , which is based on the question quid
facti? ('Are there any synthetic a priori judgements in experience?') ,
cannot undermine the deduction of a priori concepts in Ka 's
Critique , though it might well be a valid objection with regard to
Kant's Prolegomena. In the Prolegomena Kant presupposes the fact
of necessary truths in mathematics and pure science and then goes
on to infer the objective validity of the categories. In the Critique
of Pure Reason , Kant also suggests the factual existence of a priori
synthetic propositions with regard to mathematics and pure physics.
He admits , however , that these facts of a priori synthetic propositions can only serve an illustrative purpose; they cannot contribute
to the deduction , since the deduction has to take place on a transcendental leve l. 65 Hence , Kant demands not the factual existence ,
but the possibility of the factual existence of objective experience ,
that is he attempts to show that the categories can be applied to
lnt Ul tlon.
In order to counter Kant's deduction , Maimon has to prove that
pure a priori concepts are incapable of determining intuition. This
is the second line of attack pursued by v1aimon. lt can be called an
'internal' approach , since Main10n's argument affects Kant's proof
of the objective reality of the categories from within. Maimon asks:
'How can an a priori concept apply to an intution even to an a
priori intuition66 The crucial point of divergence between Ma lnon
and Kant concerns the notion of pure intuition. As Midgley remarks
97
in his Introduction to Maimon's Essay: Mairnon 'claims that either

there is no pure intuition , or to the extent that there is , it cannot fulfl
the function Kant derrlands of it.'67
For Kant , the question concerning the possibility of the synthesis
of a priori concepts and pure intuition does not arise , for both concepts and pure intuition are a priori. That is to say , pure intuition
(time and space) does not simply belong to intuition; as the formal
elelnent of in t: uition it cannot be derived from anything given in
perception and is therefore an a priori feature of cognition. 68 It can
thus fulfl the role of a mediating representation between intellectual
concepts and sensible intuition. Maimon objects that inasmuch as
pure intuition is still conceived as given (namely as formal intuition) ,
the synthesis of concepts (thought relations) and intuition (given
material) remains incolnprehensible. 69 His main concern , however ,
bears on the conception of pure intuition itself.
THE PROBLEM WITH PURE INTUITION
According to Kant , pure intuition conlprises not only space and tinle
as orms of intuition' but also fornlal intuition\70 Kant cla lns that
through a method of abstraction , it is possible to isolate and have
access to the formal properties of an object given in intuition. First
one needs to remove from the representation of an object everything that the understanding thinks about it through its concepts ,
and then all sensory content of enlpirical intuition. What remains
is formal intuition that is the extensive nlagnitude (extension and
gure) of the determinate , represented object (CPR A 20- 1IB 35).
Kant argues that space and time have a content of their own , pure
matter or extensive rnagnitude , which is logically independent of and
irreducible to concep t: s and empirical intuition.
he role of formal intuition is crucial for Kant's purpose in the
ranscendental Deduction. By means of ormal intuition' , Kant
wants to explain how it is possible that a priori concepts relate to a
posteriori intuitions. The signifcance of ormal intuition' for Kant's
argument is underlined by the fact that he introduces the distinction
between orrns of intuition' and onl1al intuition' only in ~26 of
the Transcendental Deduction , where he seeks to prove the objective
reality of the categories. There he wants to show that the categories
are not only conditions of the possibility of experience , but are also
valid a priori of all objects of experience. Kant argues:
98
The Denland for Transcendental Genetic Conditions

1. Space and time are not only forms of intuition but also formal
mt Ul t10ns.
2. Formal intuition contains a pure manifold that is combined in a
unity (the unity of a determinate , extensive magnitllde).
3. The unity of formal intuition presupposes a synthesis (a pllre synthesis of apprehension) , which neither belongs to the senses , nor
to the understanding.
4. he pllre synthesis of apprehension is performed by the faclllty of
irnagination in agreenlent with the categories , that is in agreement
with the synthesis of apperception. Conseqllently:
5. All synthesis , throllgh which even perception itself becomes
possible , stands llnder the categories (CPR B 161).
his is how Kant attempts to show that the categories determine
intllition , thOllgh not directly bllt throllgh the mediation of formal
intuition. 1n the case of mathematics , the formal intllition reqllired
is space , which is the determinable matter to be synthesised according to concepts. 1n transcendental metaphysics , the formal intllition ,
upon which the synthesising activity can act , is tirne. 1n his chapter
on Schematism , Kant elaborates time as a third thing' or mediating
representation' (CPR A 138/B 177) , which renders the synthesis of a
priori concepts and a posteriori intuition possible. The reslllt of the
synthesis of apperception , that is the synthesis under categories , is
objective experience or 'real objects'.
Let llS look , however , more closely at the synthesis of apprehension , that is the synthesis of the imagination , which brings about
cognition (albeit raw and confllsed , cf. CPR A 77/B 103) , that is
a combination of a given n1anifold in one intllition. As we have
already mentioned , if this manifold , which is brollght into a llnity ,
is pllre space and time , the product of this synthesis is formal
mt Ultln\
Maimon doubts that such a pllre synthesis of apprehension can
be conceived consistently. His main argulent is that dirence 1S a
necessary featllre of any synthesis (a synthesis being a unity of difference') , that is to say without difference there can be no synthesisJl
ln a first step , Maimon refutes Kant's claim that we can conceive an
errlpty llniform space (or time) , that is an absolute continuous and
homogeneous spatial (or tel11pora l) wholeJ2 According to Maimor
absolllte space , absolllte movement and the like are fictions or l11aginary beings (ens imaginarium) produced by the faclllty of imagina
tion , which imagines something as absolute , although it exists only
99
TRANSCENDENALIDEAS
in relation to something else. 73 Thus we can only imagine such a

thing as an errlpty llniform space , because we llnconsciollsly relate
it to differentiated sensible representations , for instance a rnanifold
of distinct objects. He provides the following example: 74 a river
can be irrlagined as a continuous and homogeneolls body of water
whose parts are indistinguishable. Yet Maimon insists that we can
only represent it to ourselves as such , beCallSe we refer the water to
distinct objects on the river's bank. If there were no distinct objects ,
the parts of the river would be indistinguishable. Maimon resorts to
Leibniz's principle of indiscernibles , arglling that if there is nothing
to distinguish parts or points quantitatively or qualitatively , these
parts or points are identica l. Hence the sllpposed extensive magnitude of a continllolls and hornogeneous eld (the river , for exarrlple)
wOllld collapse and shrink to nothing .7 5 he representation of space
as a continllolls and homogeneolls whole lacks the diversity that is
reqllired to see things apart from one another\76 Mairon's example
of the river shows that it is irrlpossible to distinguish and identify
'parts' of an indeterminate , continuous and homogeneous whole ,
be it space or time .77 It follows that a synthesis of a pllre spatio
temporal manifold in which the elements are indistinguishable IS
impossible. Kant simply assumes a pllre manifold of space and of
time , which can be synthesised to deterrninate and measurable extensive magnitudes (cf. CPR A 77/B 103). But how are we supposed to
conceive a pure synthesis of homogeneous spatial or temporal parts
that are indeed indistinguishable? The notion of formal intlltion
(allegedly being the result of a pure synthesis) looks suspiciously
inconsistent. The Kantian pure synthesis of apprehension is a fiction
just as the representation of an empty uniform space or tirne. Kant
models the pure synthesis of imagination on an ernpirical synthesis
of already spatially or temporally differentiated representations (for
instance , the synthesis of a particlllar triangle from three given lines).
He n1isses the point that a synthesis of a pure spatial or temporal
manifold in one (formal) intuition is indeed impossible.
Bllt what if we aSSllme for a mornent that a synthesis of pure
intllition is somehow possible. Kant adrrl ts that synthesis in general
proceeds obscurely:
Synthesis in general s [. . .] the mere effect of the imagination , of a blind
though indispensable function of the soul , without which we would have
no cognition at a11 , but of which we are seldom even conscous. (CPR A
78/B 103)
100

For Maimon , even if such an obscure synthesis occurs it cannot
constitute a solution to the problem: the unity of a pure spatial
or ternporal manifold (i.e. formal intuition) would be given to
the understanding but not produced by it .l8 According to Kant ,
the unity that belongs to forrnal intuition is the product of a pure
synthesis of apprehension which belongs neither to the senses nor
to the understanding. As Kant says , the unity of space and time
precedes all concepts' (CPR B 161). Formal intuition is the product
of a synthesising activity of the in1agination , which is performed
only in agreement with the categories , that is in agreement with the
synthesis of apperception (i.e. the synthesis of the understanding). lt
still rernains to bring this synthesis to concepts [which] is a function
that pertains to the understanding' (CPR A 78/B 103). herefore
it seems as if the pure synthesis of the spatio-temporal manifold is
given to the understanding qua synthesis , that is as a product of the
imagination , before the understanding has intervened at al l. Maimon
objects that inasmuch as formal intuition is still conceived as give
the synthesis of pure concepts and intuition remains incomprehensible. The legitimacy or truth of the application of pure concepts to
intuition has not been shown .l9 Thus Maimon cannot agree with
Kant's solution to the problem quid juris? Kant tries to 1 the gap
between thought concepts and given intuition by the intermediary of
formal intuition. But forn1al intuition cannot fulfil the function that
Kant ascribes to it.
Does Maimon's criticism of forn1al intuition (as a synthesised whole
frorn homogeneous spatio-temporal parts) compel him 1'0 reject the
possibility of geometry and mathematical natural science? Both sciences depend upon the possibility of deterrrning measurable spaces
and tirnes , and both claim a pure a priori status (that is to say , they
allegedly abstract from empirical representation). Maimon denies
these sciences a pure a priori status , but he does not deny thern
their possibility as sciences. Although they cannot claim apodictic
certainty , they can nonetheless claim assertoric certainty (i.e. a high
degree of subjective necessity , whic approaches the idea of objective
necessity).80 Ma lnon defines geometry and mathematical natural
science as 'real thought that is a creative thought which generates
real objects or new determinations of objects through rules of construction. The crucial point is that Mairnon envisages a new conception of synthesis , according to which synthesis is essentially genetic.
Contrary to Kant's view .
101
the understanding does not subject something given a posteriori to its a

priori rules; rather it lets it arise [lt entstehen] in accordance with these
rules (which 1 believe is the only way to answer the question quid juris? in
a wholly satisfactory way). 1) l
As we have seen , Kant assumes the pure synthesis of apprehension
as already given to the understanding but not produced by it. The
synthesis of apprehension acquires its objectivity , that is its objective significance or relation to an object , when it is brought under
concepts (synthesis of apperception). This relation is one of subsumption or conditioning: the understanding subjects the activity of
the imagination to its a priori rules. Maimon , by contrast , denlands
that the understanding itself produces the synthesis. 82 The elements
of synthesis must be wholly in1 n11ent within the understanding. ln
Maimon's view , the question that Kant fails to solve is: 10W can the
understanding subject sornething (the given object) to its power (to
its rules) that is not in its power?'83
This question would not come up if our understanding could produce
objects out of itself according to its prescribed rules or conditions without
needing to be given something from elsewhere. 1) 4
Maimon demands that both parts of the synthesis must equally be
thought. Concepts and intuition , form and lTIatter , nlust equally arise
through an internal genesis from the understanding. he underlying
idea is to dissolve the given into thought relations that precede any
cognition of objects themselres. 85 For instance , if we are able to
conceive of space as a thought relation given prior to any extended
objects , then we can eliminate space as determinable content , and
the two parts of the synthesis , i.e. the determinable (space) and the
determination (e.g. the deterrnination 'enclosed by three lines') , can
be thought equally. The model that Mainlon will use is that of di
ferential calculus. ln differential calculus , space is considered as a differential relation abstracted fro111 all quantity , that is as an intensive
magnitude (the quality of the quanturn). In other words , extensive
magnitude is reduced to its differentia l. 86 Fr OlTI there Maimon develops the philosophical concept of 'differentials' as real relations by
nleans of which he will explain how the understanding produces ,
although in an obscure manner space and tinle as extensive nla
102

HE MATHEMATICAL 10DEL OF THE DIFFERENTIAL OR
FLUXIONAL CALCULUS
Leibniz and Newton a 1'e comrnonly said to be the founde 1's of the
diffe 1'ential calculus 0 1' calculus of fluxions , though it is mo 1'e exact
to say that they gave to the infinitesimal p 1'ocedu1'es of thei 1' p 1'edecesso 1' s the algo 1' ithmic unity and p 1'eclslon necessa 1'y fo 1' fu 1' the 1'
development'.87 Leibniz and Newton both benefited f1'orn the wo 1' k
of othe 1' nlathematicians (such as Torricelli , Ba 1'1'ow , Wallis , G1'ego 1'Y
and Fe 1' mat) , but they we 1'e the fi 1' st to establish a p 1'agmatic set
of unive 1' sal 1' ules fo 1' solving the p 1'oblems unde 1' conside 1'ation.
Appa 1'ently , Leibniz and Newton developed the 1' ules of the calculus
independently f1'om one anothe 1', though the 1'e exists a la 1'ge cont 1'ove 1'sy accusing Leibniz of plagia 1'ism. 88 It is , howeve 1', plausible
that both of them indulged in sirnila 1' investigations , d 1'awing on
thei 1' comnlon p 1'edecesso 1' s' methods and applying these methods
to thei 1' 1'espective fields of study. While Newton was p 1'ima 1'ily
inte 1'ested in the physical phenomenon of movement and the rate of
change of bodies in motion , Leibniz dealt with classical geomet1'ical
p 1'oblen1s such as the problem of tangents and that of quad 1' atu 1'es.
Thus Newton's calculus can be labelled as dynam inasmuch as
Newton is conce 1' ned with the gene1'ation of va 1'iable quantities
f1'orn points , lines and planes. He called the gene 1'ated quantity a
nt' and the rate of change with which it va 1' ies a uxion'. Fo 1'
Newton , the fundamental p 1' oblem of the calculus amounts to: the
1' elation of quantities being given , to find the 1'elation of the fluxions
of these and conve 1'sely. 89 Leibniz , by cont 1'ast , held a compa 1'ably
static view of the calculus. He saw geomet1'ic quantities not so much
as gene 1'ated fluents' , but 1'athe 1' as agg 1'egates of in nitesimal elements. 90 Howeve 1', the algo 1'ithmic p 1'ocedu 1'es , which both Newton
and Leibniz codifi a 1'e ve1'y simila 1'. Mo 1'e significantl) both men
employed the idea of the infinitely small , a facto 1' which enorrnously
facilitated the ope 1'ations 1'equi 1'ed fo 1' the determination of g1'adients
of cu 1' ves 0 1' the cornputation of a 1'eas unde 1' cu 1' ves. Yet it must be
said that neithe1' Newton no 1' Leibniz made a serious effo 1't to cla 1'ify
the conceptions they employed and to find exact logical definitions.
They n1issed out on glvlng a clea 1' inte 1'p 1'etation of the meaning
of the notion of the infinitely smal l. In the mathematical debate
103
and that contributed to a confusion of thought and rrletaphysical

speculation. Differentials were designated as ultirrlate differences' ,
'quantities smaller than any given quantity' , 'qualitative' or relative
zeros' , 'ghosts of departed q uantities' ,evanescent quantities' , 'di
ferences on the point of vanishing' and rrlO mentary increments' or
decrernents of a flowing quantity 91
With regard to Newton , the infinitely small is usually equated
with his notion of uXlons\ his is , hoiVever not quite correct. As
Carl Boyer points ou the infinitely small is rather what Newton
designated as evanescent increments ,' or IT10n1ents' of a flowing
quantity.92 he notion of fluxion is based upon some other idea ,
the idea of a linting ratio of evanescent increments. A lin1iting
ratio is defined as the limit of the ratio of two quantities decreasing
until they approach zero. It is important to grasp here that fluxions
were always considered as ratios. In this way , Newton tried to
avoid dealing with the infinitely small , realising the lack of rigour
involved in a naive view of innitesimals: 'I have sought to den10nstrate that in the method of fluxions it is not necessary to introduce
into geometry infinitely small gures. Nevertheless he could not
completely expunge the in nitely small from his calculus since the
idea of the innitely small enters in the form of the elen1ents of his
limiting or 'ultimate ratios'. Thus , in the illustration of his lnethod
of fluxions , it becomes obvious that Newton could finally not elude
the infinitesimal point of view that characterised seventeenth-century
mathematical geometry.
Leibniz , on the contrary , seems to have been less concerned with
the notion of the infinitely sn1al l. He considers it a convenient fiction
which allows the facilitation of n1athematical operations. In Boyer's
words , 'he felt that the calculus , as a modus operandi , brought its
demonstrations with it.'94 This may be the reason why he did not
believe it necessary to provide a clear and consistent account of the
infinitely small , or what he calls 'differentials'. In the beginning we
see that Leibniz wavered in his attitude as to whether the differentials
were to be regarded as quantities with assignable or with inassignable values (that is as n1erely very sn1all quantities or quantities
ncon1parably smaller than any given quantity). However , starting
from the definitiol
104

can serve as a basis for understanding the mathernatical origin and
use of the notion of differentials. 96
In fact , it was Leibniz's research on the theory of number
sequences and his mathematical considerations on the nature of the
infinite that led him to a discovery that was of great impact in the
field of analytical geolnetry: the conception of infinite sequences of
line segments , distinguished into lifference sequences' and sum
seq uences'. Difference sequences' comprise an ordered sequence
of infinitely many terms with infinitely close values representing the continuous diminishing of a line segment. The differences
between the terms of the sequence were considered infinitely or
'incomparabl) small that is slnaller than any assignable interval
and yet unequal to zero. In his letter to Varignon with a note on the
Justication of the innitesimal calculus by that of ordinary algebra'
(1702) , Leibniz also refers to differentials as differences which are
on the point of vanishing'. Hence in a mathematical context differentials' can be de ned as these infinitely small differences between
successive values of a continually diminishing quantity.97 What is the
importance of the theory of infinite sequences of line segments for
the study of curves? lt was suddenly possible to interpret the curve
as an infinite assemblage of infinitely small straight lines. ln effct
this idea had already been developed by the ancient Greeks in an
attempt to solve the problem of sq uaring the circle'. According to
the farnous Archimedean-Eudoxan 'method of exhaustion' a polygon
was inscribed inside a circle. By increasing the sides of the polygon
to infinity , it was thought to approxirnate the curve and ultimately
exhaust' its arc length. hus a circle was considered an infinitangular polygon , whose sides are infinitely small and whose angles are
infinitely many. In Leibnizian calculus this ancient method was put
to use again. The line of the curve was thought of as identical with a
broken line stretched toward the curve to infinity. The infinitely small
straight segments of this broken line , if prolonged , were conceived as
tangents to the curve , which could be deternl ned by the ratio of the
differences in the ordinates and abscissas as these becolne infinitely
sn1al l. As Leibniz explains:
. to find a tangent is to draw a straight line which joins two points of
the curve Vhich have an infinitely small distance , that is the prolonged
side of the infinitangular polygon which for us is the same as the curve. 9'
In terrns of limit (although Leibniz did not use the termlimit'
tangent is the lin1it of a secant, which runs through
explicitly) the
105
two consecutive points of the curve and forms a 'differential triangle'

with the appertaining abscissa and ordinate. As the second point of
this secant approaches the first point the chosen point of tangency ,
the gradient of the secant tends toward the gradient of the tangent.
hey coincide in intuition as the two points of the secant line become
infinitely close to one another. This method of a differential triangle' ,
which depends upon the fundamental relation between ordinates and
abscissas , had in a sense already appeared in the works of previous
mathematicians such as Torricelli , Fermat and Barrow , but it was
Leibniz who developed this method for determining the differential
relations of infinitesimals. He conceived two infinitely close points of
the curve and expressed the infinitely small distance between thern
by means of infinitesirnals , or infinitely small differences. In order
to designate these infinitely srnall differences between two consecutive values of a variable , Leibniz adopted a characteristic notation ,
which has been maintained to the present day. Leibniz ernployed the
symbol dx to designate the differences in the values of the abscissa x ,
and the syrrlbol dy for the differences of the ordinate y, and dened
dy as the quantity which is to dx as the ratio of the ordinate to the
subtangent. 99 By lneans of this notation , Leibniz codified the rules for
deterrnining the gradient of tangents or , in other words , the rate of
change of a curve at the point where the tangent touches the curve.
he overall procedure involving these rules for determining differences of in nitesimals and their quotient is called differentiation.
Leibniz also codified a procedure for con1puting the area of a curvilinear figure by means of so-called sum sequences'. The area under
the curve was defined as an innite sequence of sums of approxirrlating rectangles whose sides are represented by the ordinates for infinites rnal intervals in the abscissas. This procedure is called integration
and the sum of these infinitely thin rectangles the integld. A great
discovery was made when Leibniz (and Newton) recognised the
inverse relationship between differentiation and integration. hey
discovered the remarkable property that if a geometric quantity
decreases continllally llntil it vanishes , then that same qllantity is the
sum of all the successive differences. In other words , the differential
of the integra
106

liberate the calculus from geo lTletric connotations and to provide a
rigorous arithmetic foundation according to the standards of matheITlatics as a formal discipline (see the work of Augustin Cauchy , in
particular his Cours d'analyse algbrique (1821) , Rsum des leons
sur le calcul innitesimal (1823) and Leons sur le calcul diffrentiel (1829) ,101 and Karl Teierstrass , in particular his introductory
lectures on analysis of 1859/60).102
MAIMON'S DIFFERENTIALS OF CONSCIOUSNESS
Maimon was familiar with both Leibniz's and Newton's account

of differential and fluxional calculus. Traces of both can be found
in Maimon's philosophical appropriation of the lnathematical terrn
of differentia l. For 1aimon the differential calculus provides a
metaphysical explanation of the generation of quantities , not only
mathematical quantities but also empirical quantities and their relations. He considers differentials as the genetic elements which cannot
be given in intuition , but which are nonetheless real and generate
perfectly determinate extensive quantities. Mairnon thereby draws
on the fact that the ratio dy/dx expresses a finite quantity designated
by a third term z , such that dy/dx equals z. Leibniz defined the term z
as the gradient of the tangent to a curve at a single point.hus while
z has a value that can be represented in intuition , the differential elen1ents of the ratio dy/dx cannot be intuited. With respect to intuition
differentials are equal to nothing , althollgh their ratio dy/dx does not
eqllal zero. Mapping the mathelnatical model of differential calcllllls
onto transcendental philosophy , MaInon says:
hese
differentials of objects are the so-called noumena; but the objects

themselves ansmg from them are the phenomena. With respect to
intuition = 0 , the differential of any such object in itself is dx = 0 ,
dy = 0 etc.; however , their relations are not = 0, but can rather be given
determinately in the intuitions arising from them. 103
Maimon ca l1 s the differentials noumena , at other times Ideas' or

'limit concepts' (Grenzbegriffe) , thus a l1 11ding to Kant's definition of
nounlenon bllt at the salne tirne completely modifying its meaning.
Kant openly admits that nounzenon is a problematic concept (CP R
A 254/B 309) , that is a concept which , in spite of the fact that it
contains no contradiction , can never be given a corresponding object
in intuition. In principle , it can withollt contradiction be thought as
designating a thing n-itself independent of sensibility. Such a thing
107
would be given to the pure understanding in a non-sensible or intellectual intuition. Yet Kant adds that we as hurnan beings cannot
comprehend the possibility of intellectual intuition , for the nature of
our understanding is discursive and not intuitive. Thus the concept
noumenon in its positive meaning , that is as designating a thing-initself, is inadmissible frorn an epistemological point of view , since
no object can be determined for the concept. Nevertheless , Kant
acknowledges that the concept of noumenon has an important use
if taken in its negative sense as a boundary concept' (Grenzbegriff)
(CP R A 255/B 310-11). As such it functions as a boundary for given
concepts , connected with other cognitions' (CPR A 254/B 310). It
does not designate a real object or given thing , but serves to provide
unity or an objective ground , in relation to which other cognitions
can be synthesised. The concept of noumenon in its negative sense has
all the characteristics of the Kantian 'transcendental Idea' of reason;
insofar as the Idea has no corresponding object in sensible intuition ,
it rernains an empty form of thought but serves the understanding to
be guided better and further in its cognition through concepts.
When Main10n refers to differentials , he uses the Kantian terrns
'noumenon' ,104 'limit concept' (Grenzbegriff) 105 or Idea of reason'106
interchangeably. But he does not use then1 in the Kantian sense.
This becomes apparent when he introduces the term 'Ideas of the
understanding' in order to refer to differentials defining them as he
infinitely s111all of every sensible intuition and of its fonns , which prorides the matter [Stoff] to explain the way that objects arise\107 Ideas
of the understanding are the material , genetic elements of an object
(or as Mairnon says , they provide the material completeness of a
concept , insofar as this completeness cannot be given in intuition'108) ,
while Ideas of reason ai111 at the formal totality of a concept. Maimon
dedicates a whole chapter to the end of opposing Ideas of reason
and Ideas of the understanding ,109 but unfortunately , in other parts
of the Essay , Ma lnon fails to exercise care in distinguishing clearly
between the two. Sometimes Maimon calls differentials Ideas of
reason but 1110St of the time he refers to differentials as 'Ideas of the
understanding However froln the way Maimon puts differentials to
use t becon1es cl
108

it. ll1 In this way Mairnon seeks to eliminate the Kantian sensation' ,
that is anything given to ollr consciousness throllgh callsal affection.
The givenness of sensation always indicates a referential relation to
something olltside' consciollsness. Maimon objects that we have no
right to posit the origin of sensation outside of llS. He demands a
Pllrely immanent account of cognition. According to this immanent
account , sensation only plays a role insofar as it is itself regarded as a
synthesis , whose elen1ents must be thOllght as a pre- or sllbconsciolls
manifold. In other words , every conscious perception (that is sensation or the perception of an object of experience) must ultimately be
dissolved into its elements: a rational manifold or differentials' .
Arguing that differentials are nothing bllt pure relations or intellectual genetic rllles , Maimon provides the following example taken
from mathematics. 112 Think of a right-angled triangle , one of whose
catheti rnoves infinitely in the direction of the angle opposite to
it , thereby remaining parallel to its original position. he triangle
becomes in nitely smaller and its sides eventually disappear in intuition. Nevertheless , althollgh the sides have disappeared cornpletely
and are reduced to their differentials , the relation still subsists.hat
is to say , the rule according to which a right-angled triangle can be
generated still holds between the differentials or in nitely small sides.
Leibniz and Newton were already referring to similar examples.
ThllS Leibniz claimed that the form of the triangle remained even
after all quantity had been abstracted from it. While being reduced
to zero , the evanescent quantity retains the character of that which is
disappearing. l13 Newton eqllally spoke of the ultimate forrns of evanescent triangles. 114 In order to think the transition from the nite to
the infinite , Leibniz and Newton both invoked the idea of continllity
or continuous motion. In the salne way , Maimon conceives of a continllolls transition from extensive quantities to the in nitely small ,
to the differentia l. However , we need to pay attention to Maimon's
characterisation of the differentia l.he differential is not a qllantity
in the con1n10n sense of the word , but a quality or intensive inner
n1agnitllde.
Consequently, the intensive magnitude (the quality of the quantum) is
in this case the liH
Mamon claims that qllality can be conceived as abstracted from
all qllantity , and yet stiU as being instantiated in a relation of
extensive quantities. 1l 6 He argues that the lawfulness of a relation
109
essentially belongs to the qualitative elements , after extension has

been rerrlovedJ17 For instance , in the case of the evanescent rightangled triangle , the sides , although reduced to differentials , lnaintain
the relation which is equal to ~2:1: 1. Maimon cornments:
50 , it is not a relation of number to number , since 1 have assumed both
to be infinitely small , omni dabili minora [smaller than anything given] ,

and consequently it cannot be expressed by any number in relation to any
unit , but only by the relation of one unit to another unit , i.e. this relation does not hold between the lines in so far as they are measurable , but
merely in so far as they are determined by their quality (by their position).
As a result , they are not extensive but intensive magnitudes. 1U~
Mairnon argues that pure relations do not hold between extensive quantities or numbers , but between differentials (i.e. qualities abstracted from all quantity). The ratio dy/dx gives a perfect
example for the discovery of this new type of relations. As Maimon
explains , dy/dx does not express an invariable nun1erical relation ,
like the relation of irrational 111agnitudes, but a lniversal functional
relation' . at is to say , '
dhese m
lagnltude
s [:dy and dx
] stand in a
ur
mined then the other is also determined\119 The lnain characteristic
of this universal funcrional relation seems to be , not so much the
variability of its differenrial elements ut first and foremost the fact
that none of the differentials can exist on its own outside the
relation. Outside of the ratio itself, the tern1s dx and dy have no
lneaning. It is only in and through their reciprocal relation that they
are determined. By contrast, if we consider the fraction 2/7, we notice
that each of the quanta which make up the elen1ents of the fraction
can be conceived separately. A quantum has a completely indifferent existence apart from its ratio. This n1eans that the relationship
between quanta is not essentia l. Hegel noted later that an algebraic
description of the fraction shows the salne deficiency. In an algebraic
formulation such as y/x = a , the ratio can be eliminated without loss
by transforming the original forn1ula into the expression y = ax. 120
Hence only the ratio dy/dx expresses an essential or pure relationship , in which the relation is prior to its terms. As Deleuze relnarks
in his lecture on Spinoza , the infintesimal calculus puts into play a
new type of relation the differential relation as pure relation.he
thought of relation as pure relation can only be n1adein reference
to the ir nite. This is one of the highly orignal moments of the
seventeenth century , 'l21
110

Maimon's reading of the differential calculus thus allows him to
conceive of relations that are prior to their objects and even generate
them. Conversel) every object rnust be resolvable into differential
relations. For an infinite understanding , the given completely disappears. What appears as given to our finite understanding , that is
sensation must be ordered in differential relations. l22 As Bergman
contends ,
understanding remains.'123 Hence , Maimon believes he has resolved
the question quid juris?: the pure concepts of the understanding are
not directly applied to intuition , but only to the limits of intuition ,
i.e. the real relations of differentials or Ideas of the understanding. 124
In this way the understanding relates its concepts only to a rational
n1anifold , which is entirely thought.
Just as in higher mathematics we produce the relations of different
magnitudes themselves from their differentials , so the understanding
(admittedly in an obscure way) produces the real relations of qualities
themselves from the real relations of their differentials. 50 , if we judge
that re melts wax , then this judgment does not relate to re and wax
as objects of intuition , but to their elements , which the understanding
thinks in relation of cause and effect to one anothe r. Namely , 1 hold that
the understanding not only has a capacity [Vermgen] to thin universal
relations between determined objects of intuition , but also to determine
objects by means of relations. 125
Contrary to Kant , Main10n conceives of the understanding not
only as a faculty of rules which needs to apply its a priori concepts
to something given to it from outside , but as a genuinely creative
faculty which creates the content in accordance with its a priori
rules. The mathematical n10del of differential calculus is supposed
to provide the proof of this creative capacity , which Maimon calls
real though t'. For Maimon , to create the world is to think the world ,
and therefore the thinker's ideal is to engage in an infinite progress
through which what is thought is always increased , while the given is
reduced to the infinitely slnall' .126
However , Maimon concedes that our understanding can probably
never achieve this ideal goal due to its lin1itation. Our consciousness is incon1plete , in other words 'we start in the lniddle with our
cognition of things and finish in the middle again 127 Our consciousness is a consciousness of representation , and it reproduces only
a part of the synthesis. The elements of the synthesis that is the
differentials fall outside the order of representation since they are
111
below the threshold of consciousness. They forrn a kind of primiti1Je

consciousness beyond the order of representation. Maimon defines
the genetic differential elernents as presentations' (Darstellungen).
hey do not represent anything , for in order to represent something
they need to be related to a represented whole , an object. hus
differentials can ultimately only be grasped as determined in and
through perceived objects (j ust as the differential is thought through
its integra l) .128 According to Mairnon , there is , in reality , a series
of degrees of consciousness. 129 0 consciousness of representation
lies in the middle between two limits or limiting ideas: primitive
consciousness and the consciousness of a cornplete synthesis. he
consciousness of a complete synthesis is the superior limit of our conscious representations , which can never be attained because of the
infinity that it involres. The consciousness of a complete synthesis
possesses the cornplete concept of an object, that is it knows its relation to all possible objects. 130 his idea of a higher consciousness
is rerniniscent of Leibniz's idea of a divine intellect that knows the
complete concept of an individual substance. However , Maimon
does not conceive a divine or infinite intellect external to independent individual substances (Leibniz's monads). Instead , he insists on
the possible transition from the finite to the infini that is from our
consciousness of representations to higher degrees of consciousness ,
or from our finite intellect to the infinite intellect. Only in this way
can he give a truly internal , genetic account of objects of experience anslng from unconscious differentials. As Bergman explains ,
Maimon deviates in this respect from Leibniz and comes close to
Spinoza's position. 'To the extent that Maimon believes in an infinite
understanding that created (thought) the world and in the identity
of our limited understanding with this infinite understanding he is
a disciple of Spinoza.'131 The immanent productivity of the infinite
intellect , which makes it the material cause of the world , certainly is
a Spinozistic element. Spinoza believed that all things are modes' of
the infinite substance and wholly intelligible in the divine thought.
Furthermore , Maimon's claim that the human llnderstanding is
the sarne with the infinite llnderstanding 'only in a limited way' ,
lS
112

Many readers will imagine they are catching a glimpse of 5pinozism here.
50 in order to prevent all misunderstandings of this kind , 1 now want to
explain myself once and for all: [. . .] our understanding is the schema for
the idea of an infinite understanding. [. . .] 501 differ from Kant merely in
this: instead of the three ideas that he assumes , 1 assume a single idea (of
an infinite understanding) , and 1 ascribe objective reality to this idea (not ,
it is true , viewed in itself for this is contrary to the nature of an idea but only in so far as it acquires objective reality for US).134
In this remark Mairrlon distances himself from Spinozism , which in

those days was condemned as atheist and politically radical thought.
Instead of saying that the h Ulnan understanding is a mode' of the
infinite understanding , Maimon claims that the human understanding
is the schenza for the idea of an infinite understanding. According to
Kant , the process of schen1atisation of an Idea proceeds by analogy.
Thus seen1ingly Maimon only clairns that the hun1an understanding is
like the infinite understanding , and he assumes the infinite understanding , i.e. God , as an idea which has objective reality for us (and not in
itself as a rnetaphysical , all-ernbracing and constitutive reality). his
conception of the h Ulnan understanding as a schema for the idea of an
infinite understanding seems a whole world away from an lnmanent
ontology in which the nite understanding is part of the infinite intellect and participates in an im111anent productivity. It must be said that
this ambiguity subsists in Maimon's philosophy. Perhaps one can say
that , unlike Spinoza , Maimon does not posit an in1manent ontology
fro111 the start , but rather projects it as the goal that we reach in our
philosophical investigations if they are carried through' rigorously.135
His proposed solution to the quid juris? problem certainly brings him
already on the road towards an in1rnanent ontology.136
Deleuze and Maimon

Deleuze's interest in Maimon lies precisely in the fact that the subcon
scious mechanism of reciprocally determined differentials provides
an explanation of the genesis of real experience , whereas the Kantian
conditions of possible experience are 'too general or too large for the
real' (DR 68/94). Ka 's transcendental philosophy 111ay claim to be a
philosophy of genesis , but to be precise , theris no genesis of the pheimenon but in fact there is a genesis of the intelligibility of phenon1ena\[37 Real experience only plays a role insofar as it is intelligible ,
that is t conforms with the conditions of possible experience. This
Ineans that real experience is reduced to that which can be represented
113
within the transcendental conceptual framework (DR 56/79) and that

what can be represented is decided in advance , namely solely identical
objects with extensive sensible qualities. That is excluded from the
Kantian account of objective experience is the flux of lTleaningless
intensive sensations or micro-perceptions below the threshold of
consciousness. According to Deleuz Maimon pa ves the wa y for a
transcendental (differential and genetic) psychology of perception'
(FLB 891118). This means that contrary to Kant's transcendental
conditions , which are abstracted from experiential cognitions and
arbitrarily projected as conditions of possibility of experience , the
Maimonian transcendental conditions designate genetic and differential mechanisms that generate perception or real experience.
We should not be confused by Deleuze's use of the tern1 transcendental sychology of perception': he is not pursuing a psychology of
the person or subject. Transcendental , differential and genetic conditions are objective structures or Ideas within a sub. representative ,
unconscious reahn: an inters l1 bjective differential l1 nconscious or
transcendental field.
Furthermore , it has to be noted that Deleuze's reading of Maimon
stands squarely within Mai l110n'S own project. Maimon is l1 ndoubtedly committed to a forn1 of rational dogn1atis l11: he defines differentials of consciousness as Ideas of the l1 nderstanding in order to give
a purely irnmanent account of the understanding's synthetic activity.
His solution to the question quid juris? consists in filling the gap
between given intuition and pure concepts by dissolving the given
into reciprocally determined differentials which are entirely thought.
Everything is thought , nothing is given. Deleuze adapts Maimon's
invention of differential and genetic Ideas to the Leibnizian notion
of infinitely minute perceptions. This means that differential Ideas
are not thought but sensed , although in an unconscious lnanner.
Therefore Dele l1 ze's reading of Maimon owes lnuch to his reading of
Leibniz and to his very own interest in a differential unconscious. In
the following , we willlook at how Deleuze develops the notion of a
differential unconscio l1 s taking as a point of departure the problem
of the unconscious nature of differentials , or in other words the
problen1 with the lnthought in tho l1 ght'.
THE PROBLEM WITH THE 'UNTHOUGHT IN TI-IOUGH
Maimon's solution to the question quid juris by means of which he

wanted to bridge the gap between what is given and what is purely
114

thought , is based on the premises that our finite understanding is
continuous with an infinite understanding , and that some kind of
original differential production takes place subcons:i o l1 s1y from the
point of view of our limited understanding. It seems that the original
differential production must be assigned to the unconscious activity
of an infinite understanding.
Now a certain difculty arises which in a way repeats the initial
problem on the level of the infinite understanding. he conflict
between the given and pure thought appears again. There is a
'lninimum of given (minimum de donn)'138 even in the infinite
understanding , namely precisely the differential rule of production
itself. In other words , the processes of differentiation cannot be
reduced to pure thought. If pure thought proceeds according to the
principle of reciprocal determination , if it is the nexus between differentials , there always remains a residue of something given. One is
compelled to introduce a given' , an ens reale , which is irreducible to
the pure thought of the infinite understanding. Maimon recognises
this difficulty , but seeks to resolve it n the same way as with a finite
understanding':
he
given intuited by an infinite understanding is either an objectum

reafe , signifying something present in the infinite understanding , but not
thought by it (this does not contradict its infinity , because this consists
only in the ability to think everything that is thinkable and the given is by
its nature not thinkable); or the given is a mere idea of the relation of the
concepts to something outside it , which in itself is merely a modification
of the understanding. In the latter case the actuality would not consist in
something outside the understanding , but merely in this relation. 139
Maimon's response seems far from satisfying. What he suggests

is that both can be the case. First, there can be sornething given to
the in nite understanding (though it cannot have been given from
outside). Second , this given can be conceived in analogy to an empirical representation consisting in the relation of a concept to some
external object. However , as Maimon has explained elsewhere , the
relation to something outside is illusory , since a representation is
nothing but an internal n1odification of the understanding. 'This
"outside us" signifies son1ething in whose representations we are
not conscious of any spontaneity , i.e. son1ething that (with respect
to our consciousness) is purely passive and not active in US. '1 40 The
pro len1 with transposing this 'solution' to the infinite understandng is that with the acceptancof sO lTlething given to the infinite
115
understanding , we have reintroduced an unintelligible and 'occult

quality' .141 Mairnon's insistence that this unthought' in thought does
not contradict the infinity of the innite understanding appears not
very persuasive and sounds rather like a capitulation to the problem.
Mairnon's solution' is open to serious criticism , which is presented
by Martial Guroult from a Fichtean perspective. For a Fichtean ,
the claim that there is sornething given to thought , which cannot be
brought to consciousness , is dogmatic and contradictory' .142 here
are two lnain reasons why our consciousness of representations can
in principle never shed light on the original differential production.
First , bringing differentials into our consciousness of representation would rnean destroying their definition as presentations' , i.e.
something that neither represents anything nor can be represented.
Second , if we define consciousness as a pure for itself that is not
only as a consciousness-of-something but also as a lucid , transparent
self-consciousness , there can be nothing unthought in it. This residue
would instantly dissolve by virtue of the capacity of consciousness to
penetrate itself. 143 Consequently , if one follows this line of argument ,
the manifold of differentials could only be outside consciousness ,
located in a non-consciousness. This position would in1ply a separation of the conditions of knowledge (forrns of thought and the sensible
given) fr Oln the conditions of existence (differentials or rules of production). In other words , the reason for knowing would be distinct
from the reason for being. As a conseq uence of this distinction , the
immanence of the subject of cognition would be undern1ined , since
the constitutive elen1ents of knowledge , i.e. the differentials , would
be an extrinsic reality located in an objectied infinite understanding.
he subject would actually be posterior to the reality of which it seeks
to gain knowledge. 144 he simple claim of imrnanence could not be a
change for the bette r. Apparently , we would have to conclude that the
111nite understanding was qualitatirely distinct fron1 the subject of
cognition , which is bound by the limitation of our cognitive faculty. If
we follow this argument to its ult lnate consequence , Maimon would
have only reinforced the Kantian dogrnatism of the thing-in-itself: the
occult character , wh
116

We can assume that Deleuze was very well acquainted with this
criticisrn of Maimon , in particular with the one issued by Guroult.
Indeed , it appears that Deleuze even discovered Maimon first and
foremost throllgh Guroult's books on Maimon and on Fichte, since
alrnost every time when Deleuze mentions Maimon , he also refers to
Gllroult. 147 In response to this criticism of Maimon , Deleuze argues
that the unconscious nature of differential Ideas has been completely
misunderstood , for it is precisely the unconscious nature of differential Ideas that provides a solution to the problem of the 'unthought
in thought'. It is a mistake to believe that 'the unconscious of a pure
thOllght rnust be realised in an infinite understanding' (DR 193/249).
It is eq uall y mistaken to condemn the differentials [. . .] to the status
of mere fctions unless they acquire the status of a fully actual reality
in that infinite understanding' (DR 193/249). The alternative between
rnere fictions and actual reality in an infinite understanding is simply
false. 148 According to Deleuze , the original differential production
does not refer to an in1te understanding , but to differential psychic
mechanisms in the unconscious of a finite self. Thus Deleuze elirninates the in nite understanding and restores the differential Ideas
to a differential unconscious within finite thought. He says that 'the
infinite is taken here only as the presence of an unconscious in finite
understanding , of something that cannot be thought in finite thought ,
of a nonself in the finite self' (FLB 891118). For Deleuze , there is no
contradiction in the notion of something unthought in thought. True ,
this 'occult quality' differs in kind fr0111 clear and conscious thought.
This is why it is located on a different plane that differs in kind fr0111
the order of representation. In other words , being of the order of
presentations , it does not belong to our consciousness of representation but to a differential unconscious. How can this be reconciled
with the claim of immanence of the differentials of consciousness?149
Deleuze sees a solution in a transcendental (differential and genetic)
psychology of perception' (FLB 89/118). The unconscious is both
the transcendental condition and the genetic origin of consciousness. This distinguishes precisely Maimon's transcendent from
the Kantian method of external conditioning: 'M
DELEUZE'S NOTION OF A DIFFERENTIAL UNCONSCIOUS

Since Deleuze develops the conception of a differential unconscious
with regard to Leibniz we will in what follows Inainly refer to his
117
lectures and book on Leibniz. We wi11 thereby attempt to make sense

of Deleuze's few explicit remarks on Maimon , on the assumption
that his interpretation of Leibniz is also applicable to Maimon's
case. We believe is to be justified , because Deleuze seems simply
to equate the conception of the unconscious of Leibniz with that of
Maimon. 150
For Deleuze , Maimon is the rst post-Kantian who returns to
Leibniz' (FLB 89/118) and draws a11 the consequences from the
'Leibnizian reinterpretation of the calculus' (DR 170/221). Deleuze
does not hesitate to identify Maimon's differentials of consciousness with Leibniz's minute perceptions' .151 He probably did not
know 1aimon's letter to Kant dated 20 September 1791 , in which
Maimon sharply criticises Leibniz's theory of minute perceptions. 152
According to Mairnon , the conception of obscure representations
that fa11 below the threshold of consciousness can only be of use in
anthropology , that is a doctrine that deals with hurnan beings insofar
as they are endowed with a body. Obscure representations are
modifications of the body , and not of the soul or pure consciousness.
Maimon's rationalistic commitments actua11y forbid equating his
differentials of consciousness with Leibniz's obscure representations.
For Mairnon , differentials of consciousness are the object of a transcendental critique of cognition that is concerned with genetic intel
lectual relations or laws. Thus it is irnportant to note that Deleuze
sensualises' Mairnon's differentials of consciousness by interpreting
them as unconscious micro-perceptions\153 It is true that Main10n
occasiona11y refers to differentials as differentials of sensations'. In
fact , Maimon states:
Sensibility thus provides the differentials to a determined consciousness;
out of them , the imagination produces a finite (determined) object of
intuition; out of the relations of these different differentials , which are its
objects , the understanding produces the relation of the sensible objects
arising from them. 154
But it should be noted that Ma lnon refers here simply to what he calls
a 'subjective order nan1ely how it presents itself to our consciousness of representation. 155 By contrast , the 'objective order' , which is
established by a true critique of our cognition , dispenses with sensibility or intuition altogethe r. lt only deals with Ideas and concepts. 1S6
With regard to the objective order , there are first and foremost the
'Ideas of the understanding' that provide the non-intuitive , intellectual matter , that is the manifold of differential relations. On the basis
118

of these Ideas , the faclllty of understanding determines real objects
and the relations between then1. he faculty of reason provides Ideas
of formal totality , in order to coordinate the acts of the llnderstanding so that a maXimUITl of llnity prevails aITlO ng its concepts. ThllS ,
from the point of view of a trlle critique , empirical perceptions must
llltimately be resolved into Ideas of the llnderstanding. 157
However , Delellze not only modifies the historical Maimon by
characterising differentials as Leibnizian minllte perceptions , he also
reinterprets Leibniz insofar as he defines minllte perceptions not only
as smaller than cOnSCiOllS perceptions bllt as different in kind. This
characteristic , i.e. their difference in kind , is drawn from mathematical considerations of differentials as intensive qualities in contrast
to extensive qllantities , which make up the terms of the primitive
fllnction. Delellze's redefined conception of differentials , drawn
both from Maimon and Leibniz , paves the way for a transcendental
philosophy of perception which ellldes a mere empirical psychology
or 'anthropology'.
After these preceding remarks we will now olltline Deleuze's
conception of a differential llnconsciolls that he attriblltes eqllally to
Leibniz and to Maimon.
The theory of a differental llnconsciolls is based on two claims: a
metaphysical claim , according to which the infinite world is contained within a nite consciollsness or n10nad; and a psychological
claim , according to which every consciolls perception implies an
infinity of tiny , obscllre perceptions (FLB 87/1 16). he comn10n
strllctllre of both is that something finite consists of an infinity llnder
a certain relation' .158 The fact that the rnind is finite and the world
infinite explains the darkness and obscllrity of the rnind , its depths.
The mind bathes in an llnconsciolls that consists of infinitely rninllte
perceptions lacking an object , 'that is , hallllcinatory rrlcroperceptions' (FLB 86/115). They are the representatives' (FLB 86/115) or
rather 'presentations' of the world. he world exists only in these
presentations; in other words , the world is Plllverised into a mass of
dancing particles of dllSt' (FLB 86 /1 15) , and this dllSt is spiritllalised
into infinitely n1nllte perceptions (FLB 87/116). hese tiny and
obscllre perceptions are not yet representation , that is to say these
perceptions are neither related to a whole (an object) , nor endowed
with consciollsness. Before we have clear and conscious experience
of objects , there is an 'unconscious lived experience' sllch as we
experience in the case of fainting , dizziness , or sleep.159 In these cases
119
our consciousness loses the power of self-consciousness and plunges

into a flow of rrl nute unconscious perceptions. The extrerne lirnit of
this line would be death , where death is not defined as the state of a
person who ceases to live , but on the contrary a state of catalepsy , in
which one is entirely reduced to minute perceptions.
In order to describe these distinct levels of experience - unconscious
lived experience and conscious experience of representations
Leibniz dissociates Descartes' criteria of the clear-and-distinc t' and
the 'obscure-and-confused'. For Descartes , knowledge that is derived
from the senses can only be obscure and confused , while knowledge
that is grasped by the understanding or by the naturallight of reason
is clear and distinct. Leibniz by contrast, couples the obscure with
the distinct and the clear with the confused. That is to say , the minute
and unconscious perceptions , though they make up the obscure part
of our mind , are perfecdy distinct. The sound of each tiny wave or
drop of water that together compose the roaring of the sea is distinct.
Sirnilarly , the infinitely small appetites that precede the conscious
feeling of hunger are distinguishable in elementary fonns of hunger
for salts , for proteins for fat , etc. On the other hand , our global
conscious perceptions are clear but confused. We perceive the global
noise of the sea very clearly , but it rernains confused insofar as we
cannot grasp its composition. Likewise , the feeling of hunger that
starts to rumble in our stomach is perceived clearly , but again it is
confused , since we do not consciously distinguish its composites. The
unconscious , as Leibniz supposedly conceives it , consists of minute
perceptions and minute inclinations or appetites , which are nothing
but the obscure and distinct differentials of conscious perception.
There is a genesis of our clear and conscious perceptions , which
starts from the differentials of consciousness. 160
lt is irnportant to grasp here that the conception of a differential
unconscious is very different from a Freudian unconscious , which
is characterised by drives or desires that are in conflict with consciousness. 161 For Freud , the unconscious expresses a force that
attracts all representations or inclinations opposed to or rejected by
consciousness. The unconscious and consciousness are two antagonistic forces. By contrast , the differ
120

sense the noise of the sea , it is drawn into clarity , although - living
near the shoreline - we have become accustomed to the noise and
usually do not perceive it any rnore (cf. Leibniz's example of the
grinding noise of the water mill which 1 do not perceive consciously
because 1 am already accustomed to it). This last example in particular shows that what is brought to light is in some sense selected'
from the habitu the regular or the ordinary. Only that which is
remarkable and notable for sonle reason becomes a clear and conscious perception. Leibniz uses the term singular' as an equivalent
for remarkable' and notable'. According to Deleuze , Leibniz is
inspired by the mathematical concept of 'singularity' which leads
Leibniz to create a philosophical theory: It is a very curious theory
that Leibniz was no doubt the first to introduce into philosophy , that
we could call the theory of singularities. '1 62
he theory of singularities is supposed to provide a solution to the
question as to how clarity emerges frorn obscurity , in other words
how clear and conscious perceptions are produced from tiny , obscure
perceptions. he process involved is not one of simple addition of
parts to a whole (FLB 87/116). It is not the case that very small
quantities are added up until they compose a deternlinate measurable quantity. On the contrary , our conscious perceptions must be
a product of genetic , differential relations established among minute
perceptions. The relationship between conscious perceptions and
minute perceptions is one of derivation or integration: 'The integral
is what derives fr Oln and is also what operates an integration , a kind
of totalisation , but it's a very special totalisation , not a totalisation
through addition. '1 63
Deleuze uses the theory of singularities as well as additional considerations on derivation and integration (from the eighteenth and
nineteenth centuries) for elaborating the conception of a differential
unconscious. It is clear that Deleuze's notion of a differential unconscious exceeds Maimon's theory of differentials of consciousness by
fa r. Deleuze takes advantage of later developments of the calculus ,
for instance the notion of a function and the derivative , but ironically
uses this additionallnaterial to contribute to the differential point of
view of the infinitesimal calculus. In the end , Deleuze's rereading of
the differential c
121
involved , which , howeve 1', does not interrupt the C01 nuity
between the unconscious and consciousness , so that the innnanence
of the diffe 1'entials is rnaintained. It is this pa 1' adoxical p 1'oblenl
which the rnathematical model is supposed to solve. We will fi 1' st
have to ask what a singula 1'ity signifies on the mathen1aticalleve l.
caesu 1' a'
DELEUZE'S REREADING OF THE DIFFERENTIAL CALCULUS
ln the discussion of cu 1' vilinea 1' figu 1'es , singula 1'ities designate distinctive points whe 1'e the natu 1'e of the cu 1've changes , that is whe 1'e it
shows a behaviou 1' that falls outside the o 1'dina 1' Y 0 1' 1'egula r. This
is the case , fo 1' instanc when the cu 1've 1'eaches a maximum or
llllmn1um , 0 1' else a point of in f1 ection , which ma 1' ks a change in
concavity. Singula 1'ities a 1'e indicated by the diffe 1'ential 1'elation. For
instance , if the diffe 1'ential 1'elation dy/dx , which in te 1' ms of intuition gives the g 1'adient of the tangent at a single point , is equal to
ze 1'o , then the g1' adient of the tangent is at that point ho 1' izontal and
the cu 1' ve of the o 1'iginal function (the so-called p 1' in1itive' function)
1' eaches a local maximum 0 1' local minimum. Singula 1' ities divide the
cu 1' ve of the function into pa 1'ts 0 1' neighbou 1' hoods , whose points a 1'e
of o1' dina 1'Y 0 1' 1'egula 1' cha 1' acte r. 'Each singula 1' ity is the sou 1'ce of a
se 1' ies [of 1'egula 1' points] extending in a dete 1' mined di 1'ection 1'ight
up to the vicinity of anothe 1' singula 1'ity' (LS 52-3/67; cf. also DR
278/356-7).
Th 1'ough a p1'ocess of 1'epeated diffe 1'entiation at a singula 1' point
(that is in taking the second de 1' ivative , the thi 1'd de 1'ivative , etc.) , it is
possible to cha 1' acte 1'ise not only one point of the cu1' ve , but a whole
1' ange of points in the neighbou 1' hood of the singula 1' ity.164 That is to
say , the qualitative natu 1'e of all the 1'egula 1' points in the neighbou 1'hood of a singula 1'ity can be cha 1' acte 1'ised 0 1', in geomet1'ical terrns ,
the 1' ate of change of the b1'anches of the cu 1' ve , whethe 1' they a 1'e
1' ising 0 1' falling q uick can be deterrnined lllo 1' e and mo 1' e accu1' ately. Given that all the 1'egula 1' points a 1' e continuous ac 1' oss all the
diffe 1'ent b1' anches between distinct singula 1' ities , the whole cu 1' ve can
finally be dete 1' mined in going f1' om the neighbou 1' hood of one singularity to the neighbou 1' hood of a subsequent singula 1' ity and so on.
ln the nineteenth centu 1'Y, the mathen1atician Karl Weie1' st1'ass
p 1' ovided a method to 1'ep 1'esent the cu 1've of a p 1'iIllitive function
wthin a specific domain th 1' ough a powe 1' se 1' ies. Given a p 1' in1itive
(and diffe 1'entiable) function , we can produce an infinite powe 1'
series of the form a o a1x a3x3 . . a n x l1 y SUn1111ing the
122

successive derivatives taken at a particular point of the function , such
that the values of a are calculated using the values of the respective
derivatives. Such a power series expansion , or Taylor series , can be
said to represent the primitive function in the domain. 165
Deleuze recognises an important fact on this poin t:: although we
are usually confronted with a function in calculus class that we then
have to differentiate by taking derivatives , this usual priority can be
reversed. In Deleuze's rereading of the calculus , the primitive function does not precede the differential relation , but is only the ultimate
result or by-product of the progressive determination of that relation.'166 In fact , the differential relation dy/dx precedes the primitive
function and can even be said to generate it. Weierstrass's method
proves to be an effective means to determine the characteristics or the
behaviour of a function within the neighbourhood of a given point.
Thus if we take the power series obtained at each singular point of
the primitive function , Deleuze assllmes that we can constrllct the
entire cllrve of the primitive fllnction on the basis of mllltiple power
senes. his is a qllite different procedllre from the determination of
the prinl tive fllnction by exercising the inverse operation of integration (developed by Callchy). Delellze presents Weierstrass's method
of approximation llsing sllccessive derivatives as an alternative
approach that is still tied to geolnetric conceptions and intllitions
(insofar as it determines the qllalitative behaviollr' of the cllrve) and
therefore faithflll to the infinites lnal calcllllls in its early geonletrical
phase. 167 his exposition by Delellze is all the n10re sllrprising , since
it was actllally Weierstrass who provided the calcllllls with a 'rigorOllS' fOllndation throllgh a programme of redefining geometric ideas
in Pllrely arithmetic terms. Weierstrass is best known for separating
the calcllllls fr Oln geometrical conceptions or intllitions , which were
considered to be vaglle and problematic. He thereby paved the way
for a modern interpretation of the calcllllls on the basis of set theory.
Delellze , however , criticises the arithmetisation of the differential calclllllS becallse of its failllre to think the infinite and the explllsion of
dynamic concepts ( llxions' and lllents' passage to limits ranish
ing differences' , etc.) .1 68 It eliminates the in
123
Apart frorn the generative power of differential relations , Deleuze

adds another irnportant observation: there is a difference 01 ind
between the derivative or differential relation and the primitive function (DR 172/224). While the primitive function deals with relations
between finite quantities and is therefore bound to representation ,
the differential relation prevails between dy and dx , which are not
quantities , but qualities that fall outside representation.
In other words , the differential relation is not a formula that relates x to
y over some range of values for x [. . .]. Rather , the differential relation
relates x to y not in breadth , over a range of values , but in depth; it operates in each point on the function , condensing the quality , the character
of the entire function into every poin t. 169
We are now in a position to draw some conclusions from the
rnathematical example of infinitesimal calculus , which Deleuze
covers in a unique way from Leibniz to Weierstrass. It has been
shown , firstly , that singularities and the differential relations at
these distinct points play a decisive role in the generation of the
curve of the primitive function , and secondl) that the differential
relation differs in kind fr Oll l the primitive function. heir difference
in kind from extensive , n1easurable quantities renders them suitable
candidates for a transcendental philosophy of perception. Released
from their n1athematical context, differentials can be interpreted as
minute perceptions , which are not j ust smaller but different in kind
from conscious perceptions. The reciprocal relations between the differentials produce the conscious experience of the world , which can
be interpreted as the curve of the prin1itive function. his is then how
Deleuze envisages the process of generation of conscious perceptions
from tiny obscure perceptions:
We have to understand literally - that is , mathematically - that a conscious perception is produced when at least two heterogeneous parts enter
into a differential relation that determines a singularity [. . .] For example ,
the sound of the sea: at least two waves must be minutely perceived as
nascent and heterogeneous enough to become part of a relation that
can allow the perception of a third , one that excels' over the others and
comes to consciousness. (FLB 881117)
In ths sense , the roar of the ocean s the result fro lTl productive
processes whose differential rule rernains llnconscolls. The m]ath
em-natical model serves as a transcendental (genetc and dffer
~er
n1
l
explanation of the
p sy
c hol
logy of
f perception whch replaces the
E
124

Kantian account of a simple extrinsic conditioning where concepts
are applied to something given in intuition.
Thus differential calculus is the psychic mechanism of perception , the
automatism that at once and inseparably plunges into obscurity and
determines clarity: a selection of minute , obscure perceptions and a
perception that moves into clarity. (FLB 901119)
Through Leibniz and 1aimon Deleuze enriches the notion of the
transcendental: the transcendental is in1manent , genetic , differential
and unconscious. It provides a transcendental model of genesis in
the sense that unconscious differential mechanisms generate real
conscious phenomena.
THE NOTION OF 'VIRTUALITY'
It is important to note that Deleuze's conception of a differential
unconscious is not identical with that of Leibniz or Maimon. 170 There

is a fundamental difference: Leibniz's differential unconscious , i.e.
the rnultitude of infinitely minute perceptions or the rnass of dancing
particles of dust' of the world (FLB 86 /1 15) , is contained within the
monad. The 1'elation of the Leibnizian 1110nad to the world is one
of inclusion and closure. As Deleuze 1'erna 1'ks in his discussion on
Leibniz: The dange 1' is that representation conque 1's diffe 1'ence , that
difference , even the infnitely small , is united in an infinite analytic
identity. This would mean nothing l l1 0 1'e than 'allowing the identical
to rule ove1' infinity itself' (DR 264/339). Similarly , Mairnon's differential Ideas 1'isk being engulfed in an ove1' all identit) namely the
1n nite understanding , whose p 1' inciple task is still to produce unity
through its conceptual rules. Although Deleuze provides a new focus
on Maimon by insisting on the unconscious and sub- 1'epresentative
status of differential Ideas , Maimon nevertheless 1'emains a 1'ationalist: for Maimon , there is nothing outside the understanding , the 1'e is
no extnns lC g1ven.
Obviously , Deleuze's own conception of a diffe 1'ential unconscious
cannot be equated with Leibniz's or Maimon's. For Deleuze , the differential unconscious is not included or enclosed within the subject;
rather , the differential unconscious is a fracture 0 1' rift that opens the
subject to an absolute 'outside' , a true exterior (cf. DR 169/220).
Equally , the differential and problematic Ideas a 1'e not contained in
the subject , but they are immanent in the empirical world. However ,
this does not mean that Ideas are 'actual' like empirical things.
125
Deleuze eludes the alternative between actual Ideas and Ideas as

fictions by stipulating them as 1J irtualldeas within a virtu unconscious realm , which is one facet of the empirical world. The notion
of virtuality originates froln Deleuze's stuclies of Bergson and Proust
and is further elaborated in Difference and Repetition. Being a vital
component of his theory of ldeas , the notion of virtuality needs to be
examined here.
To begin with , the virtual ought not to be confused with the possible. The possible designates an abstract conceptual possibility of
an object that is fabricated only retroactively in the image of what
resembles it' (DR 212/273). It is nothing but a copy from actual
facts , a rnere abstraction. As Bergson says in his serninal essayhe
Possible and the ReThe possible is only the real with the addition
of an act of n1ind which throws its image back into the past , once it
has been enacted.'171 We are prone to the illusion that the possible
precedes the existence of real things and only waits to be realised.
We believe that le possible would have been there f1'om all time ,
a phantom awaiting its hou 1" .172 Be 1'gson explains that we probably
confuse two meanings when we speak about the possible in such a
manne 1'. When we say that son1ething is possible , we want to say
that it is not impossible. This implies that the concept of a possible
object is not selcont1'adicto1'Y and that there is no hindrance to its
1' ealisation. However , from the negative sense of the te1'm mposs
ible" you pass surreptitiously , unconsciously to the positive sense 173
Absence of cont1'adiction and hindrance now means 'pre-existence'
unde 1' the fonn of an Idea or individual concept. lt seems that the
ldea 0 1' concept already includes a l1 the cha 1' acteristics of the object,
such that existence adds nothing to it but a simple realisation. In
Deleuze's words: The whole of existence is related to a preformed
element, from which everything is supposed to emerge by simple
"realization'" (B 20/9).
It is true that philosophers have conceived the p 1'ocess of realisation not as simply as is suggested here. he p 1'ocess of realisation of
a possible object proceeds in accordance with two rules: limitation
and opposition. From the totality of possible objects , the 1' ule of
lnitation restricts the range of possibilities to certain possibles that
pass into the 1'ea l.he 1' ule of opposition ensures the possibility of t
126

and exclusion. hat is to say , the process of realisation of objects is
ultimately a determination by means of negation in accordance with
the principle omnis determinatio est negatio. However , for Deleuze ,
this principle of determination as negation cannot explain the
process of passing into existence: very tinle we pose the q uestion in
terms of possible and real , we are forced to conceive of existence as
a brute eruption , a pure act or leap which always occurs behind our
backs and is subject to a law of all or nothing' (DR 2111273). The
possible lacks an account of the genesis of real objects. It rnay allow
for the possibility of an object but this possibility remains abstract.
he merely logical possibility of an object does not sufce to explain
its reality , that is to provide a sufficient reason for its existence.
For this reason , Deleuze takes recourse to the Bergsonian notion
of the virtua l. Bergson is not only the author who rejects the concept
of possibility , it is also he who develops the notion of the virtual to
its highest degree and bases a whole philosophy of memory and life
on it' (B 43/37). hanks to Bergson , the logic of the possible can
be replaced by the reality of the virtua l. The virtual is not an empty
image abstracted frorn actual experience and thrown backward by
the mind as a conceptual possibility. he virtual is a full reality: it
is real without being actual and id 1 without being abstract (DR
208/269).174 One of its most fundamental features is its creative
potential whose realisation is wholly unpredictable. The virtual
does not pre-form or pre-j udge the actual objects coming into existence in the way that conceptual possibility does. he virtual is a
creative force , a fullness constantly differentiating itself and creating
something unforeseeable and new.
Bergson introduces the notion of virtuality in the context of the
distinction between two types of multiplicity (cf. B 38/30-1): a
spatial or numerical multiplicity of discrete quantities assembled
within honlogeneous space and a temporal or qualitative multiplicity (duration) , which is not denumerable. It consists of differences
in qualities or intensities. For exalnple , the process of dissolution
of a sugar lump in water can be regarded as a temporal rnultiplicity
or duration. The two multiplicities differ from one another insofar
as spatial ITlultiplicities include differences in degree , while
127
that is in terrns of qualities or intensities. Qualities or intensities

make up the virtual facet of the world , which subsists or insists in
the extended and material world. One could also say that virtual
and actual are the dimensions of one movement or becoming: the
virtual cannot be separated frOITl a movement of actualisation. The
virtual beCOITleS actualised through processes of differentiation. 'For
actualization comes about through differentiation , through divergent
lines , and creates so many differences in kind by virtue of its own
n10vement' (B 43/36). In other words , processes of differentiation in
the virtual are the real ground fron1 which divergent lines of evolution emerge. The movelnent from the virtual to the actual designates
genuine creation and replaces thefalse movement of realisation [of
the possible] understood as abstract limitation' (DR 212/274).
We can already see where Deleuze is heading: Leibniz's in nitely
n1inute perceptions or Maimon's differentials of consciousness are
transforn1ed into pure virtualities (DR 279/3 S7). he reality of the
virtual consists of the differential elements and relations along with
the singular points which correspond to them' (DR 209/269 70).ln
this way , Deleuze eludes the illusion of a subordination of difference
to identity. The di erentials are no longer contained in an overall
identity (the infinite analytic identity of the lnonad or the infinite
understanding) , but are set free within a virtual realm , which is one
facet of the pluralistic , empirical world.
According to Deleuze , any hesitation between the virtual and
the possible [. . .] is disastrous , since it abolishes the reality of the
virtu (DR 212/274). Deleuze accuses Leibniz of failing to grasp the
distinction between the virtual and the possible. Although no one
has been better able to imlnerse thought in the element of difference
and provide it with a differential unconscious , surround it with little
glimrnerings [of ldeas] and singularities' (DR 213/27S) , Leibniz still
conceived the realisation of Ideas as realised possibles
ln summary , Deleuze argues that ldeas are virtua l. They subsist
within a virtual , sub-representative and unconscious reahn. The differential unconscious is a fracture or rift in the subject, which opens
the subject precisely to this virtual realm , which possesses the reality
of a task to be perfonned or a problen1 t
In the next chapter we wiU see how Deleuze expands the notion of
vrtualldeas in order to in c1 ude the Kantian characteristic of ldeas as
problems. As we shall see , Deleuze does not stick to the notion of 'di
ferentials of sensation' or to a '(genetic and differential) psychology
128
The De111and for Transcendenta! Genetic Conditions
of perception' that we have discussed in this chapter. Differentials of

sensation are only one type of Ideas , but there are various other types
of Ideas that concern all the faculties of cognition (DR 193/249).
Among these Ideas , Deleuze pays particular attention to those that
generate thought: 'Ideas must be called "differentials" of thought ,
or the "Unconscous" of pure thought' (DR 194/251). He continues
that thought thinks only on the basis of an unconscious , and thinks
that unconscious in the transcendent exercise' (DR 199/258). The
transcendent exercise of thought means the suspension of common
sense and good sense. Ideas have the role of a paradoxical or problematic element that is capa le of pushing the cognitive faculties to
a disjoint transcendent exercise. It is a q uestion , therefore , not of a
common sense but , on the contrary , of a "para-sense" (in the sense
that paradox is also the contrary of good sense' (DR 194/250). In
fact , Deleuze finds a model of ths disjoint paradoxical exercise of the
faculties in the Kantian account of the experience of the beautiful and
the sublin1e in Kant's Critique of the Facu!ty of ]udg111ent.
Notes
1. LS 253/292; see also Deleuze's ecture course on Kant (14 1arch

1978).
2. Dosse , Gilles Deleuze et Flix Guattari, pp. 116 and 159.
3. Cf. Leigh , Deleuze , Nietzsche and the Eternal Return' , p. 208.
4. Kant's fears of such a radical critique subverting the ground of reason
itself are well expressed in his Critique of Practical Reason , p. 126:
Nothing worse could happen to all these labors , however , than that
someone should make the unexpected discovery that there is and can
be no a priori knowledge at al l. But there is no danger of this. It would
be like proving by reason that there is no such thing as reason.'
5. Cf. Nietzsche , On the Genealogy of Morals , p. 589: The will to truth
requires a critique -let us thus define our own task - the value of truth
must for once be experimentally called into questio
6. Nietzsche , Beyond Good and Evil , p. 199.
7. Ibid. , p. 237.
8. Nietzsche , Gay Science , p. 20 1.
9. Ibid. , p. 20 1.
10. Nietzsche , Beyo d Go d and Evil , p. 236.
11. Ibid. , p. 200.
12. Nietzsc 011 the Genealogy of Morals , p. 544.
13. Ibid. , p. 533.
14. Ibid. , p. 543 [French in the original].
15. Nietzsche, Twilight of the ldols , pp. 40-1. Nietzsche does not exempt
129
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
Kant from this accllsation , since Kant at least with the Critique ol
Practical Reas n - reintrodllces a transcendent , intelligible world.
Nietzsche , Beyond Good and Evil , p. 236.
Patton , Deleuzian Concepts , p. 1 1. As Palll Patton explains , Delellze
does not content himself with this systematic Nietzsche. In his later
writing
s (particlllarly in
tar
y Nietzsche' who hardly belongs in philosophy at al l. He is rather
the inventor of a new kind of discollrse , a cOllnterphilosophy that
is defined by its essential relation to the olltside , to intensity and to
lallghter' (Patton , Deleuzian Concepts , p. 12).
As we have already mentioned in Chapter 1 in the section on the
concept of error , the term conceptllal persona' , which first appears
in What Is Philosophy? , refers to the sllbject of enllnciation of a phi
losophe r. According to Delellze and Gllattari , a philosopher creates a
conceptllal persona or sometimes even several personae (for instance ,
Nietzsche's Zarathllstra , Dionyslls , the priest , the higher man , Socrates ,
etc.) who carry Ollt the movements of thOllght and play an important
part in the creation of concepts. It is no longer the philosopher who
speaks , bllt the conceptllal persona who says 'I' (cf. WP 64/63).
Nietzsche , On the Genealogy ol Morals , p. 473.
Boglle , Deleuze and Guattari , p. 16.
Nietzsche Beyond Good and Evil , p. 395.
Nietzsche , 0 The Genealogy ol Morals , p. 476.
Ibid. , p. 47 2.
Nietzsche , Human , All Too Human , p. 175.
C f. Delellze , Nietzsche' , p. 73.
See also Nietzsche , The Will to Power , section 564 (1885 and
section 565 (Fall1886) , pp. 304-5.
Nietzsche , Beyond Good and Evil , p. 210.
Ibid. , p. 238.
Ibid. , p. 238.
Nietzsche , The Will to Power , section 715 (November 1887 March
1888) , p. 38 1.
Ibid. , section 619 (1885) , pp. 332-3.
Note at the Kallfmann/Hollingdale translation still mistakenly
renders Nietzsche's an inner world (eine innere Welt) , as 'an inner
will'. For more on the textllal criticism of The Will to Power , see
Wolfgang Mller Lallter "Der Wille Zllr Macht" als BllCh der
"Krisis" philosophischer Nietzsche-Interpretation' , in Behler et a l.
(eds) , Nietzsche Studie vo l. 25 , especially p. 257.
Nietzsche , Beyond Good and Euil , p. 238. See also The Will to Power ,
secton 1067 (1885) , p. 550.
Cf. Somers-Hall, Hegel and Deleuze on the 1etapl
130
35. See also NP xi , 7/7-8 , 5 1/57-8 , 52/58 9 , and Deleuze , Nietzsche' ,

p.73.
36. In the terminology of DR , we can say that the Nietzschean couple
Dionysus/Ariadne equals the power of difference and repetition. See
DR 4 1/59.
37. Nietzsche , The Will to Power , section 617 (1883-5) , p. 330.
38. Ibid. , section 692 (March June 1888) , p. 369.
39. Nietzsche , On the Genealogy of Morals , p. 513. See also Nietzsche ,
The Will to Power , section 643 (1885-6) , p. 342.
40. It is worth noting that Deleuze's interpretation of the relationship
between forces effectuated by the will to power is much less conflictual
than Nietzsche's description of the will to power as the tyrannically
inconsiderate and relentless enforcement of claims of power' would
suggest (Nietzsche , Beyond Good and Evil , p. 220). However , Deleuze
insists tha t the notions of struggle, wa rt lry or even comparison
are foreign to Nietzsche and to his conception of the will to power'
(NP 82/93 , also DR 5 1/72). Deleuze substitutes a differential play of
forces for the Nietzschean tyrannically inconsiderate and relentless
power claims.
41. Nietzsche , On the Genealogy of Morals , p. 514.
42. Nietzsche , The Will to Powel section 552 (Spring-Fall 1887) , p. 298.
43. Deleuze , Lecture Course on Chapter Three of Bergson's Creative
Evolution , p. 72.
44. After having read the manuscript of Maimon's Essay on
Transcendental Philosophy , Kant acknowledges the excellence' of
Maimon's Essay in a letter to Marcus Herz , and admits l0t only that
none of my opponents has understood me and the principle question as well as Mr Maimon , but also that only a few people possess
such an acute mind for such profound investigations [as he does]'
(Kant , Letter to Herz , 26 May 1789' , in Briefwechsel, p. 395. An
English translation can be found in the Appendix to Maimon , Essay ,
p. 231). Fic hte confessed in a letter to Reinhold his limitless respect for
Maimon's talents and warned that future generations would ridicule
them for not having appreciated Maimon's accomplishments (Fichte ,
Letter to Reinhold , (March or April) 1795' , in Gesamtausgabe , vo l.
111, p. 282).
45. Not even Kant , who is commonly regarded as a leading gure of the
Enlightenment , can be exempt from the charge of anti-Semitism. In a
letter to Reinhold he wrote (in stark contrast to the favourable remark
he had made only five years earlier): As regards Mamon with his
"improvement
131
46. Cf. Freudenthal , Between Two Cultures' , p. 2.

47. Atlas , From Criticaf to Speculative Idealism , pp. 12-13.
48. See , for example , (1) Lecture Course on Chapter hree of Bergson's
Creative Evolution , pp. 72 , 77 and 78; (2) NP 52/58 footnote; (3) 'Te
Idea of Genesis in Kant's Aesthetic' , in DI 61; (4) DR 170/221 , 1734/224-6 , 192-3/249, 310/66 footnote , 324/226 footnote , 326/254
footnote; (5) Lecture Course on Kant , 14 March 1978; (6) FLB 89.
49. Guroult , La Philosophie transcendantale , p. 64. See also Appendix ,
remark no. 1 with regard to Leibniz , pp. 157-9.
50. Midgley , introduction to 1aimon Essay , p. xlviii.
51. In his excellent Introduction' to Maimon's Essay , Midgley mentions
the importance of Maimon's philosophy for a thoroughgoing philosophy of difference in which relations and differences are prior to their
objects' (Maimon , Essay , p. xliv) , but he does not refer to Deleuze
explicitly.
52. Jones , Solomon Maimon' , in Jones and Roffe (eds) , Deleuze's
Philosophical Lineage , pp. 104-29. Smith, Genesis and Difference:
Deleuze , Maimon , and the post-Kantian Reading of Leibniz' , in
McDonnell and van Tuinen (eds) , Deleuze and the Fold , pp. 132-54.
Lord Kant and Spinozism , chapters 5 and 6.
53. Bryant, Difference and Givenness , pp. 8-9 , 46 , 194 and 242. Kerslake ,
Immanence and the Vertigo of Philosophy , pp. 10 , 145-7 and 165 6
note. Sauvagnargues , Deleuze: L'Empirisme transcendantal , pp.
225-37. As far as we know , the first mention of the MaimonDeleuze
nexus can in fact be found in Michael Roubach's article Salomon
Maimon's Philosophy and Its Place in the Enlightenment: Wandering
in the Land ofDifference' in Freudenthal , Rational Dogmatist,
Empirical Skeptic , pp. 86 and 88. However , his reference remains too
brief and obscure to be helpfu l.
54. See CPR Appendix to the Transcendental Analytic: 'Remark to the
Amphiboly of Concepts of Reflection' , pp. 371-83.
55. Kant , Letter to Herz , 26 May 1789' , in Briefwechsel, p. 396. For the
English translation see Maimon , Essay , p. 23 1. Interestingly , in rhis
same letter Kant attempts to draw a distinction between Maimon's
assumption of an intuitive intellect and that of Leibniz (and Wolff).
He says '1 very much doubt that this was Leibniz's or Wolff's meaning ,
or whether it can really be inferred from their denitions of sensibility as opposed to the under
132
56.
57.
58.
59.
60.
6 1.
Spinozist monism , which he sees present in Maimon's way of thinking (Maimo Essay , p. 231)? For more details on Kant's rejection of
Spinozism see Lord's excellent study Kant and Spinozism.
We follow Henry Allison in the analysis of the ranscendental
Deduction as an argument consisting of two parts. See Allison , Kan t' s
Transcendental Idealism , p. 134: 'The essence of my interpretation
can be expressed in the formula that the frst part of the Deduction
is concerned with the objective validity (objective Gltigkeit) of the
categories and the second part with their objective reality (objective
Realitt).'
Allison does not seem to distinguish between real (i.e. given spatiotemporal , yet a priori) objects and actual (i.e. empirical) objects. See
his defnition of objective reality': The notion of objective reality has
an ontological sense. To claim that a concept has objective reality is
to claim that it refers or is applicable to an actual object [my empha
sis , D. V.]' (Allison , Kan t's Transcendental Idealism , p. 135). For
Maimon's purpose , the distinction between real' and actual' is vital ,
since he will use this distinction in the critique of Kan t. (Maimon
opposes the actual to the real , just as he opposes empirical' or arbitrary thought' to real thought'.)
What the accepted fact' in Kant's regressive transcendental argument
exactly is has been interpreted differently in the secondary literature
on Kant. It could be simply some knowledge that we possess , scientic propositions of knowledge , or a sort of Cartesian truth about
consc lO usness.
For the distinction between regressive and progressive transcendental
arguments , see Franks , All or Nothing , p. 208.
See Maimon , Essay , p. 186. (References to passages from the Essay
are to the page numbers of the German original edition , marked at the
top of each page of the translation.) Cf. also 1aimon Philosophisches
Wrterbuch , p. 48 and Maimon , Streifereien , p. 73.
See , for instance , Maimon , Streifereien , p. 73: 'Although critical
philosophy is to the highest degree systematic , that is , self-coherent
[unter sich zusammenhngend ], it does not refer to anything rea l. Its
transcendental concepts and principles , categories , ideas , etc. have
no reality. With regard to the origin of these forms of thought of the
understanding , and with regard to the cmpleteness and systematicity ,
it [tra
133
65.
66.
67.
68.
69.
70.
7 1.
72.
73.
74.
argues that Maimon misconstrues Kan t' s claims , and that it seems
clear that for Kant , the schema , as a transcende l1 tal determination
of time , cannot be as it were "read off" [ablesbar] from empirical
objects' , in Engstler , Ul1 tersuchu gen zum Idealismus , p. 87 (my
translation , D. V.).
CPR A 94/B 127: he unfolding of the experience in which they [the a
priori concepts] are encountered , [. . .] is not their deduction (but their
illustratio since they would thereby be only contingent.'
Maimon's letter to Kant (7 April1789) , in Maimon , Essay , p. 229.
Midgley introduction to Maimon's Essay , p. xxvii.
C f. CPR A 20/B 34: 1 call all representations pure (in the transcen
dental sense) in which nothing is to be encountered that belongs to
sensation. '
Maimon , Essay , p. 60: assuming that time and space are a priori
intuitions , they are still only intuitio l1 S and not a priori CO l1 cetS; they
make only the terms of the relation intuitive for us , and by this means
the relation itself. But not the truth and legitimacy of its use.' 5ee also
Maimon , Essay , p. 64.
The different meanings of pure intuition in Kant's CPR are distinguished in Allison , Kan t' s Tra l1 sce l1 de l1 tal Idealsm , pp. 94-8.
Maimon argues that consciousness requires synthesis , that is to
say 'something must be given that is thought by the understan
ing as a manifold (through unity of difference)' (Maimon , Essay ,
pp. 130-1).
Cf. l mon Essay , p. 180: The possibility of thinking space Nithout
objects is purely transcendent.'
Ibid. , p. 19.
Ibid. , pp. 346-7 and 135-6. 5ee also Maimon , Versuch eil1 er eue
Logi p.480.
75. his
76.
77.
78.
79.
80.
argument is clearly presented in Beiser, The Fate ol Reaso l1,

p.302.
Ibid. , p. 30 1.
Maimon's argument with reference to space can be transferred to time
as wel l. As Maimon explains: 50 if we nevertheless still represent in
space things like water that are identical in intuition , this takes place
only in relation to something different , i.e. the representation is transcendent. It is the same with time: if 1 have slept for , say , a few hours ,
then 1 can only perceive the time by means of the different positions of
the hands of a clock for example' (Maimon , Essay , p. 134).
Ibid. , pp. 60 and 64.
Ibid. , p. 185: 1 claim , that the synthetic propositions of mathematics are certainly t
134
8 1.
82.
83.
84.
85.
86.
87.
88.
89.
90.
9 1.
92.
93.
94.
95.
96.
97.
98.
neither a priori (in my sense of the word) nor pure.' See also Maimon ,
Philosophisches Wrterbuch , p. 199.
Ibid. , pp. 201.
Ibid. , p. 63.
Ibid.
We will not examine here Maimon's pril1 ciple of determi l1 ability , by
means of which he attempted to ground an objective (i.e. necessary)
synthesis of the determinable and the determination (see , for instance ,
Maimon , Essay , pp. 20 , 258 and 260). Maimon believed that with the
principle of determinability he had found a truly immanent , genetic
principle which served as an a priori criterion and even as a sufficient
reason ('real ground of possibili') for the synthesis of mathematical
concepts. But in fact , Maimon could not do without something extrir
sically given , namely space as the highest determinable Moreover
the principle of determinability cannot be applied outside the realm of
mathematics. herefore we agree with Guroult that Maimon eventually felt the need to search for another superior principle' , which iVill
be difference' or the differentials of consciousness'. Cf. Guroult , La
Philosophie tra l1 sce l1 da l1 tale , p. 53.
Maimon , Essay , pp. 394 5.
Boyer The History of the Calculus , p. 299.
Ibid. , p. 188. There Boyer provides further references with respect to
this matte r.
Ibid. , p. 194.
Ibid. , p. 193.
Ibid. , pp. 12-13 , 212 , 216 and 219.
Ibid. , pp. 20t-2.
Isaac Newton , quoted in Boyer, The History of the Calculus , p. 202.
Ibid. , p. 209.
Ibid. , p. 213.
We base the following account on Bos' arti c1 e Differentials , HigherOrder Differentials and the Derivative in the Leibnizian Calculus'.
C f. Duffy,The Mathematics of Deleuze's Differential Logic and
Metaphysics' , in Duffy (ed.) , Virtual Mathematics , p. 119.
Cf. Leibniz , A New Method for Maxima and Minima , as 'ell
as angents 'hich Is Not Obstructed by Fractional or Irrational
Quantities' , quoted in Bos , Differentials , Higher-Order Differentials' ,
p.19.
99. Boyer , The History of the Calet.s p. 210. In fact , the definition
of dx and dy , which Leibniz provided in the first published account
of the calculus ('A New Method for Maxima and Minima , as 'ell
as Tangents 1hich Is Not Obstructed by Fractional or Irrational
Quantities' , in Acta eruditorum , 1684) , treats differentials as finite ,
135
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
11 1.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
assignable quantities. However , from his definition of the tangent

given above , it becomes clear that his avoidance of differentials was
only superficia l. Recent studies show that Leibniz regarded the differential as fundamental throughout his work. Modern mathematics , by
contrast , subordinates the notion of differential to the notion of limit
(in a sense similar to Newton) by defining the derivative to be the limit
of the ratio dy/dx.
Ibid. , p. 10.
Cauchy uvres Ser. 2 , vols 3 and 4.
Since Weierstrass frequen t1 y did not publish his results when he
actually achieved them , it is rather difcult to put a precise date on his
discoveries. Most of Weierstrass's work became known to the mathematical world through his lectures at the University of Berlin. A collection of his lecture notes can be found in his Mathematische Werke.
Nlaimon , Essay , p. 32.
Ibid. , p. 32.
Ibid. , p. 192. He further calls differentials he limits of objects of
experience' - see pp. 186 and 187.
Ibid. , pp. 32 and 355.
Ibid. , p. 82.
Ibid. , p. 75.
Ibid. , pp. 75-83.
Ibid. , pp. 192 and 355-6.
Ibid. , pp. 168-9 , 190 and 193.
Ibid. , pp. 394-6.
Cf. Bergrnan , The Philosophy of 50/omo l1 Maimo l1, p. 266.
Boyer , The History ofthe Ca/cu/us , pp. 218 and 219.
1.aimon Essay , p. 395; see also p. 122.
Ibid. , p. 355.
Bergrnan , The Phi/osophy of 50/0111011 Mai111011 , p. 260.
1aimon Essay, p. 396.
Ibid. , pp. 354-5.
Somers-Hall , Hegel and Deleuze on the MetaphysicaI Interpretation
of the Calculus' , p. 563.
Deleuze , Lecture Course on Spinoza , 17 February 198 1.
Bergman , The Philosophy of 5010111011 Mamo p.258.
C f.lairnon Essay , p. 355: he pure concepts of the understanding
r categories are never directly related to intuitions , but only to their
elernents , and these are ideas of reason [sc] concerning the way these
intuitions arise.'
Ibid. , pp. 355-6.
Maimon , Phlosophsches Wrterbuch , p. 193 (my translation , D. V.)
and p. 187. See also Maimon , Essay , p. 419.
136

127. Maimon , Essay , p. 350.
128. Ibid. , p. 206.
129. Cf. Guroult , La Philosophie transcendantale , p. 65. See also
Gurou 1t, Fichte , pp. 115-17.
130. Maimo Streifereien p. 64.
13 1. Bergman , The Philosophy of Solomon Maimon , p. 218; see also pp.
221-2. Bergman explains Maimon's attraction to Spinoza with the
fact that both thinkers relied on the same background in Jewish philosophy , in particular the medieval philosopher Maimonides and his
(Aristotelian) doctrine of the identity of the ens intelligens , the ens
intelligibile and the intellectus (p. 34; see also chapters 9 and 10).
132. Maimon , Essay , p. 65. See also Spinoza , Letter 32: As regards the
human mind , 1 maintain that it , too , is a part of Nature; for 1 hold that
in Nature there also exists an infinite power of thinking which , in so
far as it is infinite , contains within itself the whole of Nature ideally ,
and whose thoughts proceed in the same manner as does Nature ,
which is in fact the object of its thought. Further , 1 maintain that
the human mind is that same power of thinking , not in so far as that
power is infinite and apprehends the whole of Nature , but in so far as
it is finite , apprehending the human body only. he human mind , 1
maintain , is in this way part of an infinite intellect' (in The Lettel pp.
849-50).
133. Kant , Letter to Herz (26 1ay 1789) , in Maimon , Essay , Appendix 11 ,
p.23 1.
134. Maimon , Essay , pp. 365-6.
135. Note Maimon's cautious formulation in the following quote: We
assume an infinite understanding (at least as idea) , for which the forms
are at the same time objects of thought , or that produces out of itself
all possible kinds of connections and relations of things (the ideas).
Our understanding is just the same , only in a limited way. This idea
is sublime and will , 1 believe (if it is carried through) , overcome the
greatest difficulties of this kind' (Maimon , Essay , pp. 64-5).
136. In her book Kant and Spinozism (2011) , Beth Lord defends a strong
Spinozistic reading of Maimon. 'Maimon's purpose is to show that
transcendental idealism cannot work unless it posits the reality of a
Spinozistic supersensible substrate underlying appearances. Maimon
thereby attempts to do just what Kant suggests is impossible for us
in the Critique f Judgment: to think a supersensible substrate that is
both an intelligent and an immanent cause' (p. 105). See also p. 126:
Certainly , Maimon aligns hin
137
137.
138.
139.
140.
14 1.
142.
as actualizing itself as Ollr own minds and the objects of Ollr sensible
intllition. That is , we necessarily think of the in nite intellect as actllal
becallse Ollr llnderstanding is a mode of the infinite intellect, and Ollr
thinking is a limited instance of its thinking. The idea of the infinite
intellect is in us beCallSe we are in it.'
Delellze , Lectllre COllrse on Chapter Three of Bergson's Creative
Evolution , p. 72.
Gllrolllt , La Philosophie transcendantale , p. 59 and p. 84.
Maimon , Essay , p. 25 1.
Ibid. , p. 203.
Gllrolllt , La Philosophie transcendantal p. 84. See also Gllrolllt ,
Fichte , pp. 122 and 13 1.
Gllrolllt , Fichte , p. 122. Sllch a critiqlle of Maimon can indeed be
fOllnd in Samllel Atlas , From Critical to Speculative Idealism , p. 117:
For to have something "given" and not to be consciolls of it is
selcontradictory. '
Gllrolllt , Fichte , p. 122.
143.
144. Ibid. p.126.
145. Ibid. , pp. 122 and 13 1.
146. Vllillemin , L'Hritage Kantien et la rvolution Copernicienne ,
pp. 48-9 (my translation , D. V.).
147. See DR 310/66 note 13 DR 324/226 note 6/1 , DR 3256/254
note 15/1. Fllrthermore , Delellze's admiration for Gllrolllt can
also be seen with respect to his essay on Gllrolll t' s original genetic
strllctllral method which enables Gllrolllt to adopt a new approach to
Spinoza's Ethics. This method implies bringing Ollt the strllctllre of a
philosophical system , that is its constitlltive differential elements , their
interrelations and organisation in series' , along which the strllctllre
or order of reasons' evolves synthetically (cf. Delellze, Gllrolll t' s
General 1ethod for Spinoza' , in DI 146-55/202-16). Olivier Revalllt
d' Allonnes , one of Delellze's close friends when they were both Stlldents at the University of the Sorbonne , reports that Gllroult has
always been a great model for them , in particular because of his
method of text-interpretation. 1 have always fOllnd that Gilles was a
great student of Gllrolllt' (qlloted in Dosse , Gilles Deleuze et Flix
Guattari , p. 122; my translation , D. V.).
148. Samuel Atlas , for instance clearly states that e two possibilities
differentials as fictions or as actllal realities in an in nite understanding
- are mlltllally exclllsive and therefore necessitate a choice. Atlas
arglles that Maimon must be interpreted according to a strictly idealistic view of the differentals. Differentials shollld be llnderstoo
138
'he Demand for Transcendental Genetic Conditions
149. Maybe it should be emphasised here that the term :erentials of

consciousness' does not imply that differentials are contained within
consciousness. As Martial Guroult explains: There are therefore
differentials ol conscioness which are neither objects of intuition ,
nor of consciousness , but the generic (and genetic) elements of the
intuitions of this consciousness' (Guroult , La Philosophie transcendantale , p. 60; my translation , D. V.).
150. In the discussion following his presentation The Method of
Dramatization' Deleuze speaks indiscriminately of a Leibnizian or
Maimonian conception of the unconscious (DI 115/161).
151. 'For Mamon , as for Leibniz , reciprocal determination of differentials
does not refer to a divine understanding , but to tiny perceptions as
representatives of the world in the finite self' (FLB 89/119).
152. Maimon's letter to Kant , 20 September 1791 , in lmmanuel Kanfs
Wer vo l. X , pp. 92-4.
153. 'Inconspicuous perceptions are thus not parts of conscious perception ,
bu t req uisi tes or genetic elements , "differentials of conscio usness" .
Even more than Fichte , Salomon Mamon the first post-Kantian
who returns to Leibniz - draws all the conseqllences from this kind of
psychic automatism of perception' (FLB 89 /1 18).
154. Maimon , Essay , pp. 31-2.
155. Ibid. , p. 81.
156. Ibid. , p. 82.
157. Ibid. , p. 205. According to Cassirer , the seeming ambiguity about the
provenance of differentials - are they differentials of sensation' , Ideas
of the understanding' or Ideas of reason'? is due merely to a lack
of clarity in Maimon's expression. While it is true that Maimon refers
the diffrentials sometimes to sensibility , sometimes to the under
standing and sometimes to reason , he does so in different manners
and from different points of view. Cassirer arglles that the overall
idealistic character of his theory remains unaffected: the sensible
manifold [can be resolved] into a rational manifold'. See Cassirer , Das
Er enntnisproblem in der Philosophie , pp. 100-1.
158. Deleuze , Lecture COllrse on Spinoza , 17 February 1981.
159. Deleuze , Lecture Course on Leibniz , 29 April 1980.
160. Ibid.
161. Cf. DR 106/140 and Lecture Course on Leibniz , 29 April 1980.
162. Deleuze , Lecture Course on Leibniz , 29 April 1980. For Delellze , the
mathematical concept of singularity is very fruitflll for philosophy.
Mathemati
139
163.
164.
165.
166.
167.
ordinary into the realm of philosophy , Deleuze actually creates new

criteria for the relevance or necessity of a proposition. This means that
we are no longer concerned with the question whether a proposition
is true or false , but rather whether it is relevant , remarkable or interesting (cf. WP 1111106 , 82/80). As Deleuze says , the notions of relevance , necessity , the point of something , are a thousand times more
significant than the notion of truth. Not as substitutes for truth , but as
the measure of the truth of what I' m saying' (N 130/177). Generally
speaking , the art of thinking would be the capacity to distinguish singular points from ordinary points , that is remarkable and interesting
ideas from banal and stupid remarks.
Deleuze , Lecture Course on Leibniz , 29 April1980.
Leibniz had also considered a repeated application of differentiation ,
which would yield higher-order differentials. He ded the secondorder differential ddx or d 2x as a variable innitely small with respect
to dx. 1n the same way , one could define further differentials of higher
orders that further application of the operator d would yield (Bos ,
Differentials , Higher-Order Differentials' , p. 19). However , Leibniz's
definition remains vague , since he could not explain in what way
the differentials of higher order differ from firstorder differentials.
He also did not distinguish between differentials of independent and
dependent variables (Boyer he History of the Calculu p.211). he
point is that Leibniz lacked the concept of a function , which describes
a unidirectional relation between an independent' variable x and a
dependent' variable y. 1n the absence of the concept of function , the
concept of derivative could not be developed eithe r. Thus in order to
outline a theory of singularities' , Deleuze takes advantage of later
developments of the calculus.
C f. Duffy,The Mathematics of Deleuze's Differential Logic and
1etaphysics' pp. 128-9.
Evens , 'Math Anxiety' , p. 11 1.
As Duffy says: Deleuze is therefore able to cite the contribution of
Weierstrass's theorem of approximation in the development of the
differential point of view of the innitesimal calculus as an alternative
point of view of the differential calculus to that developed by Cauchy ,
and thereby establish a historical continuity between Leibniz's di
ferenti
140
1 71.
172.
173.
174.
175.
Deleuze's paper 'The Method of Dramatization' , DI 115/161: To

sum things up , 1 don't have the same conception of the unconscious as
Leibniz or Mamon.'
Bergson , The Possible and the Real' , p. 100.
Ibid. , p. 101.
Ibid. , p. 102.
Deleuze takes this formula from Proust who characterises Ideas in
such a way: Rels sans tre actuels , idaux sans tre abstraits'. In
Proust , A la recherche du temps perdu , vo l. 111 , p. 873.
Cf. B 29/20: Dualism is therefore only a moment , which must lead to
the re-formation of a monism.'
141
Ideas as
1 lems
In the previous chapter we saw how Maimon took Kant's transcendental philosophy in a radica11y new direction with his notion of
differential Ideas of the understanding which serve to explain the
genesis of real experience. We then saw how Deleuze reconstucted
Maimon's Ideas of the understanding as virtual Ideas belonging to
an intersubjective differential unconscious. In this chapter we will
continue our examination of Deleuze's theory of Ideas by means of
a closer analysis of the Kantian theory; in particular we wi11 see the
importance to Deleuze of Kant's characterisation of Ideas as 'ob.
lems. We will also see how Maimon's freeing of Ideas from exclusively belonging to the faculty of reason is taken further in Deleuze's
notion of Ideas that 'occur throughout the faculties and concern
therrl a11' (DR 193/249).
It is important to note that Deleuze uses the notion of 'faculty' in a
different sense than it has traditionally been used in philosophy. First
of a11 , he not only refers to sensibility , irnagination , 111enlory, understanding and reason , but also to what he ca l1 s a faculty of speech , a
faculty of 'sociability' , a faculty of vitality' , and he lea ves open the
possibility or faculties yet to be discovered , whose existence is not
yet suspected' (DR 143/186-7). In the second place , Deleuze argues
that faculties are not given ready-nlade , but enlerge and develop with
the Ideas or problems they encounter.hirdly it 111uSt be enlphasised
that the doctrine of the faculties cannot be grounded on a slnple
empirical psychologism. Rather , it needs to be related to an ontology
of Ideas or , in Deleuze's words , a dialectic of Ideas'. Fourthly , the
faculties are not faculties of a sovereign and self-present subject (a
Cartesian Cogito or Transcendental Ego). They can only be faculties
of a split subject , which is open to the 'outside' or to a differential
unconscious. In the light of this modifed notion of a faculty which
is free of a11 psychologism , Deleuze calls for a return to the doctrine
of the faculties: 'the doctrine of the faculties is an entirely necessary
component of the systern of philosophy' (DR 143 /1 86).
Now , Deleuze ascribes to each faculty a corresponding Idea
142
Ideas as Problen1s
with its own 'transcendent object'. In saying that an Idea has a
transcendent object , this is not to say that the Idea is a concept of
something beyond any possible experience. The object of the Idea
is the problem or the problematic' (DR 169/219). This means that
each faculty discovers at its extren1e limit something which it cannot
grasp (from the point of view of its empirical exercise) but which
it is forced to grasp nevertheless. We have already encountered the
sentiendum of sensibility ,1 i.e. that which can only be sensed but
which is irnperceptible from the point of view of representation.
he sentiendum is the differential manifold of minute and obscure
perceptions , or the 'beir of the sensible' (DR 140/182). Deleuze
also introduces an imaginandum of the faculty of imagination (DR
143/1 86) , a memorandum of rnemory ,2 a cogitandum of thought. 3
The sentiendum , imaginandum , memorandum and cogitandum are
all Ideas or transcendental signs' for something which can never be
given in representation but which must be grasped nonetheless.
For Deleuze , reason is no longer the privileged faculty of Ideas
as it was for Kant , since there are Ideas corresponding to each of
the faculties. However this does not n1ean that there are types of
ldeas that exclusively concern only one corresponding faculty. As
Deleuze says , ldeas are 'not the exclusive object of any one [faculty]
in particular not even of thought' (DR 193/250) , but they occur
throughout the faculties and concern them al l. A sort of communica
tion arises between the faculties , which is initiated through the Ideas
they encounter.o put it in more concrete terms , Ideas set their
corresponding faculty into motion and carry it to its extrerrle lirrl t ,
but at the same time this violence is communicated from one faculty
to another (DR 194/251). For instance , sensibility is set into motion
by the encounter with its sentiendum and it forces memory in its
turn to remerrlber the memorandum. Memory then forces thought to
grasp the cogitandum , that which can only be thought. 'The violence
of that which forces thought develops from the sentiendum to the
cogitandum' (DR 141 /1 84). Thought can only grasp its cogitandum
at the extremity of the 'fuse of violence' along which each faculty is
pushed to its transcendent exercise. The important thing to notice is
that the faculties no longer cooperate hanno
143
of what he lTleanS by a fundarnental enCollnter with a sign or Idea.

The narrator in Proust's In Search 0/ Lost Time encounters lTlany
different types of signs: the frivolous signs of society life , the deceptive signs of love , the sensuous signs of the material world and the
essential signs of art. SOIne are more obtrusive than others , but they
all in fI ict themselves on the narrator with necessity. For example ,
being haunted by jealousy the narrator tries to track down the lies
of his beloved and to intrude into her secret world. Every love affair
IS a dispute of evidence' (PS 117/143) and the jealous lover fnds
himself in a delirium of signs' (PS 122/150). His task is to decipher
the spiritual element' , the essence' or 'truth' out of the cOlnplex of
sensation. Proust opens a kind of 'inner space' in the empirical , a
space of essences or ldeas that are nevertheless in the sensible. lt is
important to note here that the sign is not to be identifed with the
object that emits the sign. For example , at one point in the novel
the narrator experiences the rising up of the in-itself of Cornbray'
(the town of his childhood) or a splinter of the pure past when he
takes a bite of the madeleine , but the secret of the sign does not lie in
the taste of the cake dipped in tea (DR 122/160 note). Equally , the
love affair with Albertine does not reveal the secret of love: the narrator's love for Albertine is a repetition of his love for Gilbert , which
is a repetition of his love for his mother which is a repetition of
Swann's love for Odette. These different repetitions form two adult
series (the love between Swann and Odette and the love between the
adolescent narrator and Albertine) that are brought in COn11TIUnication by an infantile series , i.e. the narrator's childhood events when
he was flled with love for his mother , knew Swann as the family's
friend and later adored Gilbert, the beautiful daughter of Swann and
Odette. These series coexist within an intersubjective unconscious'
(DR 124/163). ln sum , if we seek the truth of a sign ( the in-itself of
Combray' or the never-lived reality of the Virgin' , DR 85/115) , we
need to abstain from an objectivist temtation. The truth of the sign
is not to be found in the object emitting the sign. We rnay now be
tempted to look for the truth of the sign in subjective associations of
ideas. In the frst book of In Search 0/ Lost Tinze , we see Swann as
a prisoner of his own sllbjective states. He is captured by
144
Ideas as Problems
it is real without being actllal , ideal without being abstract'. In its
pllrest form it is best manifested in the signs of art that aim to make
the invisible visible.
While the main focus in Deleuze's early book Proust and Signs
is on the different types of signs , it i\Till be seen that in subseqllent
books Delellze mostly sllbstitlltes the notion of Ideas-problems for
the concept of sign. We believe that the reason for this is that the
concept of sign uSllally presllpposes a consciollsness that interprets
and deciphers the signs given to it. By contrast , Ideas-problems ,
according to Delellze's fnal defnition of Ideas as strllctural multiplicities of reciprocally determined differential elelnents , are objective and llnconsciolls strllctures that evolve throllgh self-organising
processes. This defnition of Ideas allows Deleuze to replace the subjective procedllres of interpretation with unconscious or impersonal
processes elicited by the genetic force of Ideas.
For the development of his own 'calculus or dialectic of Ideas' , we
i\T iU show that Deleuze adopts irnportant aspects of Kantian Ideas
and further invokes sorne mathematical theories and concepts which
he borrows fronl the French philosopher and mathernatician Albert
Lautman (1908 -4 4 ).4
Kant on Ideas of Pure Reason

The opening sentence of the frst Introduction to Kant's Critique of
Pure Reason reads as follows:
Human reason has the peculiar fate in one species of its cognitions that it
is burdened with questions which it cannot dismiss , since they are given
to it as problems by the nature of reason itself, but which it also cannot
answer , since they transcend every capacity of human reason. (CP R A vii)
Kant's investigation starts off with the startling observation that
reason poses questions and problems to us which cannot be
solved by our capacity of thought. Reason nds within itself pure
concepts which apply neither to any object of experience nor to
anything outside experience. It is nonetheless tempted to use these
concepts in the determination of transcendent objects , of things n
thernselves , and thus experinlents with a transcendent application
of these concepts to objects outside experience. herellpon such
selproclaimed sciences as rational psychology (psychologia rationalis) , rational cosmology (cosmologia rationalis) and transcendental
theology (theologia transcedentalis) elnerge , prornising to provide
145
knowledge of transcendent objects. But, as Kant will show , they

inevitably st l1 rnble into transcendental ill l1 sions thatistranscenden
tal paralogisn1s and antinornies that give rise to antithetical worl
concepts (for instance , 'he world rn l1 st have a beginning in time'
and 'The world -series is in nite'). The controversies abo l1 t issues
transcending any possible experience derrlarcate the traditional battlefield of metaphysics. Kant calls the concepts of pure reason , which
trespass our errlpirical limitations and lure l1 S into metaphysical
speculations of thought , 'transcendental Ideas'. The transcendence of
our finite perspective is , however , only the consequence of a misuse
of transcendental Ideas and cannot be imputed to reason or the ldeas
themselves.
The ideas of pllre reason can never be dialectical in themselves; rather it is
merely their misllse which brings it abollt that a deceptive illllsion arises
Ollt of them; for they are given as problems for llS by the natllre of ollr
reason , and this highest cOllrt of appeals for all rights and claims of ollr
speclllation cannot possibly contain original deceptions and semblances.
Presllmabl) therefore they have their good and pllrposive vocation in
regard to the natllral predisposition of ollr reason. (CPR A 669/B 697)
Kant draws the concept of ldeas frorn Plato. However , according
to the Platonic acco l1 nt , Ideas can actually only be intuited thro l1 gh
a forn1 of reminiscence which involves the task of transcending our
finite point of view , that is of transgressing the limits that nature has
set to the empirical exercise of Ol1 r faculties. ln the Transcendental
Dialectic , Kant explains:
Plato made llse of the expression idea in sllch a way that we can readily
see that he llnderstood by it something that not only cOllld never be
borrowed from the senses , bllt that even goes far beyond the concepts
of the llnderstanding (with which Aristotle occllpied himself), since
nothing encollntered in experience wOllld ever be congrllent to it. Ideas
for him are archetypes of things themselves , and not , like the categories
merely the key to possible experiences. In his opinion they flowed from
the highest reason, throllgh which hllman reason partakes in them; Ollr
reason, however, now no longer nds itself in its original state, bllt mllst
call back with toil the old, now very obscllre ideas throllgh a recollection
(which is called philosophy). (CPR A 313/B 370)
It sho111d be noted that Kant msconstr l1 es Plato by attributing to

him the notion of a highest reason' or dirine l1 nderstanding , within
which the ideas exist. 5 As the editors of the recent English edition of
Kant's Crtique of Pure Reason point Ol1 t , 'it was not Plato's doctrine
146
Ideas as Problems
that ideas are the thoughts of God , but this doctrine did originate
in syncretistic Platonisrn from the period of the Middle Acaderny.'6
Paul Guyer and Allen W. Wood explain that Platonists like Philo
of Alexandria , Plotinus and 5t Augustine , as well as all those who
merged Platonism with Christian religion , developed a theory of
Ideas as existing within the divine n1ind. However , despite this
rnisinterpretation , Kant is right in describing Platonic Ideas as something transcendent , as archetypes of actions and things in the world ,
for which we have to reach out in a process of reminiscence. Plato
considered Ideas as e~cient causes , not only in morality but also in
regard to nature itself .7 That is to say , Plato thought that Ideas are
a constitutive ground and can actually cause their object. Kant , on
the contrary , will reject both claims: Ideas of pure reason are neither
transcendent nor do they have a constitutive function. Instead , he
defines them as a priori concepts given within the rnind that serve
a good and consequently immanent use' (CPR A 642/B 670). This
Inmanent use aims at the understanding and is always regulative ,
never constitutive. Kant only allows moral Ideas to have a constitutive function as original causes. Thus practical reason , through its
Ideas , brings about the reality of the objects of these Ideas. As Kant
says:practical reason even has the ca usality actually to bring forth
what its concept contains' (CPR A 328/B 385). Yet , Ideas of pure
reason have a regulative use only , nalnely to give unity a priori
through concepts to the understanding's manifold cognitions , which
may be called "the unity of reason and is of an altogether different
kind than any unity that can be achieved by the understanding' (CP R
A 302/B 359). Transcendental Ideas demand a unied use of the
understanding and an extension of this unity of the understanding , if
possible , to the unconditioned , that is the totality of conditions. The
unconditioned , Kant says , contains a ground of synthesis for what is
conditioned' (CPR A 322/B 379). In a certain respect , Kant endorses
Plato's theory of Ideas:
If we abstract from its exaggerated expression , then the philosopher's

spiritual flight , which considers the physical copies in the world order ,
and then ascends to their architectonic connection according to ends ,
i.e. , ideas , is an endeavour that deserves respect and imitation. (CPR
A 318/B 375)
In fact , Kant attempts something quite sin1ilar to Plato. ]ust as Plato

sought to capture the COS1110S in a unified and ordered whole , so
seeks to comprehend all the cognitions of the understanding
147
in a unified systern. The transcendental Ideas of pure reason point

all the concepts of the understanding in one and the sarne direction ,
that is they let the concepts of the understanding converge at one
point , which , although it is only an idea (focus imaginarius) [. . .]
nonetheless still serves to obtain for these concepts the greatest unity
alongside the greatest extension' (CPR A 644/B 672). However ,
while Plato considered Ideas to provide an objective , constitutive
ground , i.e. an ideal architectonic for the cosmos as a uni whole ,
Kantconcedesthatthe
systematic unity (as mere idea) is only a projected unity , which one must
regard not as given in itself, but only as a problem; this unity , however ,
helps to find a principle for the manifold ar particular uses of the understanding , thereby guiding it even in those cases that are not given and
making it coherently connected. (CPR A 647/B 675)
Thus , Kant says , the systematic unity is merely something subjectively and logically necessary , as method' (CPR A 648/B 676 not
objectively necessary.
However , this seems to be a provisional conclusion only. We agree
with Henry Allison , Christian Kerslake and others that the systematic unity of empir al knowledge is more than a logical principle'
(CPR A 648/B 676) , the function of which is to help the understanding by means of Ideas to secure unity and coherence. Systematic unity
is more than a heuristic method , a recommendation or desideratmTI
pursued by reason , but rather a genuinely transcendental principle
that requires SOlTIe kind of deduction (albeit different from the one
of categories).8 As Allison points ou this raises the puzzle about
how a principle could have both transcendental status and a 111erely
regulative function'.9 We will come back to this point in the section
on the problem of a transcendental deduction of Ideas of reason
below which deals with the problem of the transcendental deduction
of Ideas of reason.
Now let us bring Deleuze into the discussion. That Deleuze takes
over fron1 Kant is the definition of Ideas as problems' constituting
'a unitary and syst!em
la
tic eld W
iVhic h orientates and Sllbsur
111les the
researches or inves
sti
ga
tions
s in such a n1anner
t i1a
t the answe
rs lfl
l
I
tu
r form
1 precisely cases of solution' (DR 168/219). Without the
presupposition of a problen1 i.e. a unitary systematic field , or
b nz 110 cases of solution could be found. lndeed although
Deleuze usually rejectsmethods' in the wake of his critique of the
dogmatic Image of thought , this can be said to be a method that he
148
Ideas as Problenls
adopts hirnsel f. For instance , in trying to understand a philosophy ,

we have to find out what drives this particular philosophical thin
ing. We rnust raise ourselves to the level of the underlying problern\
The way in which a philosopher actualises this problern within a
unitary , syrnbolic field under certain conditions and phrased in a specific terrninology deterrnines the solution. Solutions always occur as a
function of the explication of a problern or problernatic field (as in a
vector field the local trajectories of solution curves are deterrnined by
a singularity in the irnrnediate vicinity). ln an interview with Claire
Parnet , Deleuze cornrrlents on the reason for engaging in the history
of philosophy.lO e sees the engagernent in the history of philosophy
as a long apprenticeship' in which one learns how to restore the
problerns that have affected a philosopher and are often concealed
frorn view , and to discover what is innovative in the concepts created
in response. According to Deleuze , if one has not found the problerrl
to which a concept corresponds , philosophy rernains abstract. He
goes on to say that philosophical problerns do not always stay the
sarne: they are not sirnply discovered as sornething 'given' , but need
to be constituted. This 111eans that in articulating a traditional philosophical problern , in relating it to one's own conditions and stating
it in one's own terrns one no longer poses it in the salue way as it
has been posed before. In this sense , the constitution of problerns
and the creation of concepts as their solutions are an inexhaustible
task. There is an evolution that can best be described as a 'becorning
of thought'. Problerns evolve and do not sirnply disappear in their
solutions: they are the indispensable condition without which no
solution would ever exist' (DR 168/219) , but the condition itself is
plastic and changing. The problern is determined at the same time
that it determines what it conditions , in other words the condition
itself does not remain indifferent or unaffected by the way a solution
is deterrrl ned. Philosophers are compelled to commence and recom
mence perpetuall) to cope with problems that undergo transformations as soon as they are actualised. We must therefore discern the
following characteristic features of problems: (1) Solutions do not
pre exist their problems , rather problerns are essentially without
solution , that is without a fin
149
found in experience. For instance , the concept of an absolute whole

of experience is a transcendental Idea , 'since , because we can never
project it in an image , it remains a problem without any solution'
(CPR A 328/B 384). (2) Problems are not only transcendent but
immanent at the sarne time (DR 163/212). his rneans that they are
not isolated from the field of solutions but nstead entertain an immanent , intrinsic relation with their solutions. hey insist' or 'persist'
in the solutions.
It is important to point also to the differences between Kant's and
Deleuze's conception of Ideas. For Deleuze , Ideas are problematic,
objective structures providing a sufficient reason for the genesis of
thought. Kant on the contrary , still refers to them as subjective , a
priori principles providing unity and systematicity as transcendental
conditions of knowledge. For Kant , the faculty of transcendental
Ideas or problems is reason. Ideas or problerrls occur within the
rrl nd; they are bound up with the transcendental '1'. With regard to
the question of objective reality , transcendental Ideas are even more
ren10te fron1 objective reality than categories , because no corresponding object of experience can be given to thern in concreto. The
systematic unity and completeness that transcendental Ideas demand
can never be achieved in empirical experience (CPR A 567/B 595).
For this reason , 'no objective deduction of these transcendental ideas
is really possible , such as we could provide for the categories' (CPR
A 336/B 393).
Deleuze , on the contrary , releases transcendental Ideas from their
attachment to the subject and defines then1 as objective structures. In
a sense , Deleuze says , Kant has already pointed out a d 111ension of
objectivity , and in fact Kant claims that as a transcendental principle
the systematic unity of nature must be presupposed as objectively
valid and necessary' (CPR A 65 1/B 679) or at least rrlust have objective but indeterminate validity' (CPR A 663/B 691). In Deleuze's
words:
Kant likes to say that problematic Ideas are both objective and undeter.
mined. The undetennined is not a simple imperfection in Ollr knowledge
or a lack in the object: t is a perfectly positi objective structllre which
acts as a focus or horizon within perception. (DR 169/219-20)
Deleuze analyses three components of Kantian transcendental Ideas:
(1) Ideas are undetermined since no adequate object corresponds to
them in experience. (2) Yet , Ideas are determinable by analogy with
objects of experience , ahd (3) they carry the ideal of a con1plete and
150
Jdeas as Problems
determination , insofar as they grant affinity to all the concepts of the understanding , such that a continuous transition from
every specific concept to every other through a graduated increase
of varieties is made possible. he manifold of empirical concepts can
thus be brought under a few concepts of the understanding.
mnite
Ideas , therefore , present three moments: undetermined with regard to

their object , determinable with regard to objects of experience , and
bearing the ideal of an in f1 nite determination with regard to concepts of
the understanding. (DR 169/220)
Deleuze , however , criticises Kant for leaving two of the three
rnoments extrinsic: Ideas are determinable only in relation to objects
of experience , and bear the ideal of determination only in relation to
concepts of the understanding (DR 1701220-1). Contrary to Kant ,
Deleuze demands that transcendental Ideas need to be defined as
self-organising, intrinsic problematic structures. Let us now turn to
Kant's attempt to provide some ind of deduction , that is to sornehow
answer the q uestion of the objectivity of transcendental Ideas.
THE P)BLEM OF A TRANSCENDENTAL DEDUCTION OF IDEAs
OF REASON
It is not sufficiently clear how this deduction is to be understood ,

whether or not Kant actually undertakes it and what its formal
structure is. In any case , it can be argued that Kant needs to give
SOITle kind of deduction for transcendental Ideas in order to j ustify
the precise validity and demonstrations that ideas can have'. l1
Kant appears to prornise a deduction in Section Three of Book
One of the Transcendental Dialectic , but this interpretation is not
certain since the decisive passage in the Critique of Pure Reason
remains an1biguous due to a blurred spelling and a subsequent
difference in reading. Is Kant saying we can undertake a subjective introduction [Anleitung] to them [the transcendental Ideas]
from the nature of our reason' or we can undertake a subjective
deduction/derivation [Ableitung] of them om the nature of our
reason' (CPR A 336/B 393)? Moreover , no systematic argument in
the sense of a deduction can be made out in the section indicated
by Kant. Therefore op lIl lons as to what the purport of Kant's
subjective 'introduction/deduction' is diverge considerably. Per aps
Kant has in mind to provide a kind of 'birth certificate' for them
by tracing the transcendental Ideas fr0 I11 the categories of relation
151
that correspond to the three formal species of syllogism (categorical , hypothetical and disj unctive synthesis). According to this line of
argument the strategy of Kant's deduction consists in extending the
categories of relation to the unconditioned , i.e. the totality of conditions (cf. CPR A 322/B 379). Thus the categorical synthesis , which
corresponds to the category of inherence and subsistence (substantia
et accidens) and which attributes a predicate to a subject, necessarily
leads to the concept of the absolute unity of the thinking subject.
That is to say , by indefinitely running through a series of terms
which all function as predicates (i .e. determinations which can only
occur in connection with a detenninable) , one wiU finally reach a
term which is no longer a predicate but a subject which is thought
in itself and cannot be attributed to any other tenn. The hypothetical synthesis , which corresponds to the category of causality and
dependence and claims the existence of a cause for each effect , will
lead to the whole sum of conditions by ascending in the series of
conditions to the absolutely unconditioned , that is a presupposition
which presupposes nothing furthe r. The disjunctive synthesis , which
corresponds to the category of community and represents an aggregate of reciprocally detern1ining elen1ents , can be extended to the
highest rational concept of an All of reality (omnitudo realitatis) ,
that is the complete manifold of all real or possibly real predicates
as members of the division of this highest original concept. In this
way pure reason provides the transcendental Ideas of the soul , the
world and God.
However , the objection can be lnade that this tracing of transcendental Ideas from categories of relation is not the purport of a
deduction , no more than the tracing of the table of transcendental
categories fr0111 Aristotle's logical functions of all judgements is a
deduction. Kant describes it as merely providing the clue (Leitfaden)
to the discovery of the categories , their deduction being a quite
separate matter (CPR A 70/B 95). In a similar way the categories just
provide a schema , which allows for a classication of transcendental
Ideas , but does nothing with respect to a grounding or warrant for
their alleged necessity with respect to empirical inquiry. Hence what
a deduction really requires is to demonstrate the objective validity
and objecti
152
Ids
as Problems
of objective validity) , and second a demonstration of the emplncal instantiation of the a priori concepts (the problem of objective
realty). As Kant admits , the transcendental Ideas cannot be given
a corresponding object in experence , so they are even n10re remote
from objectve realty than the categories. However , we can still
demand a deduction in the rst sense: a justification of the objective
validity of transcendental Ideas. Although Ideas are not constitutive'
for nature , they are constitutive for knowledge. This means that transcendental Ideas are conditions of knowledge (though not of objects
of experience as the categories) , and as such their validity must be
justied by means of a deduction. As Kerslake notes , this deduction indeed occurs in the Appendix to the Transcendental Dialectic
under the heading 'On the final aim of the natural dialectic of human
reason'.12 There , Kant says:
One cannot avail oneself of a concept a priori with any security unless one
has brought about a transcendental deduction of it. The ideas of reason ,
of course , do not permit any deduction of the same kind as the categories;
but if they are to have the least objective validity , even if it is only an
indeterminate one , and are not to represent merely empty thought-entities
(entia rationis ratiocinantis) , then a deduction of them must defnitely be
possible , granted that it must also diverge quite far from the deduction
one can carry out in the case of the categories. That deduction is the completion of the critical business of pure reason , and it is what we will now
undertake. (CPR A 669-70/B 697 8)
In the following passage Kant declares that transcendental Ideas are

schemas. Although they do not determine any object of experience ,
they represent an irnagined object or thought-entity which serves to
represent other objects to us , ordered in accordance with the conditions of a maximum unity , that is in relation to the imagined object in
the Idea. As Kerslake puts it: 'The Idea thus is a schema of the concept
of unconditioned concepts for the orientation of "other" concepts.'13
According to Kant , we are j ustified in assuming the transcendental
Ideas as objective (CPR A 673/B 701) , since they provide us with
the systernatic unity necessary for the acquisition of knowledge. 14
However , the imagined objects in the Ideas should not be thought in
themselves , but only relative to the world of sense. Kant distinguishes
between assuming son1ething absolutely from assuming something
relatively (CPR A 676/B 704). In the case of transcendental Ideas ,
we assume their ideal objects only relatively , that is as regulative
principles. Kant concludes:
153
And this is the transcendental dedllction of all the ideas of speclllative

reason , not as constitutive principles for the extension of Ollr cognition
to more objects than experience can give , bllt as regulative principles for
the systematic unity of the manifold of empirical cognition in generp
throllgh which this cognition , within its proper bOllndaries , is cllltivated
and corrected more than cOllld happen withollt sllch ideas, throllgh the
mere llse of the principles of llnderstanding. (CPR A 6711B 699)
hus
the claim to validity is finally warranted in that transcendental

Ideas are schemas for ideal objects in relation to which our experience finds its n1aximum of systematic unity. Without this ideal focus
or horizon that reason posits through its Ideas we would have
no reason , and without that , no coherent use of the understanding , and , lacking that no sufficient mark of empirical truth' (CP R
A 65 1/B 679).15 According to Kant , this would rnean that the classification of natural things in terms of a hierarchical taxonomy of
genera and species (such as the classificatory systern of Linnaeus; cf.
CJ 20: 216) , or the construction of systen1atic explanatory scientific
theories , would be in1possible for natural sciences. With regard to
theory construction , for example , we would be confronted only with
a contingent, distributive unity of en1pirical laws , without being
capable of realising relations of derivation , i.e. the derivability frolll
higher-level laws , and the final comprehension of the indefinitely
many empirical laws in a collective unity under a few principles.
Therefore transcendental Ideas are necessary conditions for the
possibility of systematic empirical knowledge. If we ruled out this
possibility in advance , our empirical scientific inquiries into the inner
nature of things and the laws of nature would be absurd. This is why
the understanding , which is legislative with regard to possible objects
of experience , is ultimately dependent on reason and its Ideas of purposiveness , unity or totality in nature. Reason has an indispensable
task in unifying , totalising and conditioning through its Ideas (cf.
CPR A 326/B 382 3 ).
DELEUZE RE]OINDER
be sure , Deleuze s very critical of Kant's account of transcendental Ideas as unifying 0 1' totalising conditions that a 1'e supposed
to ground a systerrl of natu 1' e. As Deleuze says Kant held fast to
the point of view of conditioning without attaining that of genesis'
(DR 170/221). The Kantian account of conditioning is flawed , not
only in the case of catgo 1' ies as we have already discussed , but also
154
Ideas as Problems
in the case of transcendental Ideas. Kant conceives of Ideas and their
irnagined objects in analogy to real things. Thus he states that the
imagined objects in the Ideas should be grounded only as analogues
of real things , but not as things in themselves' (CPR A 674/B 702).
What Kant means by this analogical relation becomes clearer in his
discussion of the Idea of a supreme being as the schema of the ideal
object God'. Kant allows himself to posit an intelligent being or
being of reason (ens rationis ratiocinatae) , though not absolutely , but
only relative to the world of sense' (CP R A 677/B 705). He says that
we must consider all the things in the world as if they had
a supreme and all-sufficient ground [. . .], namely an independent , original , and creative reason , as it were , in relation to which we direct every
empirical use of our reason in its greatest extension as if the objects
themselves had arisen from that original image of all reason. (CP R
A 672-3/B 700 1)
In fact , what Kant has beforehand condemned as an illusion of pure

reason - nalnely the 'ideal of reason' of a supreme individual being
standing at the summit of the possibility of all things , providing
the real conditions for their thoroughgoing deterrnination' (CPR
A 582-3/B 610-11) - proves to be not only a natural and unavoidable illusion but also a useful and compelling fiction for the systematic
unity of the errlpirical world. Kant wiU pursue this thought further
in the Critique of the Power of Judgment where he calls upon us to
assume an understanding more powerful than our own for purposes
of reflection (not for determining judgements).16
However , the Kantian account of the fiction of a supreme being
becomes problelnatic and dangerously close to a transcenden tal
subreption' when he starts to determine this mere Idea of a suprerrle
being in anthropomorphic terrrlS as the highest intelligence , that is in
analogy with the ernpirical concept of an intelligence:
Still more , in this idea we can allow certain anthropomorphisms , which
are expedient for the regulative principle we are thinking of, without fear
of blame. For it is only an idea , which is by no means related directly to
a being different from the world , but rather referred to the regulative
principle of the world's systematic unity , but only by means of a schema
of that unity , namely of a supreme intelligence that is its author through
wise intentions. (CPR A 697/B 725)
Kant cont
155
suprerrle being in analogy with real things , only extending what we

are acquainted with in the empirical world to infinity.17
For Deleuze , however , the transcendental must be purged of all
resemblance to the errlpirical if it really is to serve as a ground (cf. LS
123 /1 49). The transcendental , conceived as a rnere abstract doubling
of the conditioned , only uncovers the conditions of possibility (of
experience or systematic knowledge in general) , and not , as Deleuze
demands , the conditions of genesis. Kant's procedure of thinking the
objects in the transcendental Ideas as analog l1 es of empirical things is
complicit with the logic of resemblance. In this way , transcendental
Ideas cannot , as Kant wants , be explained through the concept of the
unconditioned , insofar as it contains a gro l1 nd of synthesis for what
is conditioned' (CPR A 322/B 379). The ground that Kant offers is
no more than a copy (dcalque) of the empirical that is elevated to
a transcendental leve l. The Kantian transcendental Idea is the sum
total of conditions that are tho l1 ght in the image of the empirical
and rernain abstract forms witho l1 t any genetic potentia l. By contrast , Deleuze demands something unconditioned which is capable
of ass l1 ring a real genesis (cf. LS 19/30). For Deleuze , Ideas will be
conceived as genetic and differential n1ultiplicities. This means that
their determination is p l1 rely intrinsic and need not rely on a determ
nation by analogy with en1pirical concepts. I-I ence Dele l1 ze states that
thetranscendental Idea is
an internal problematic objective unity of the undetermined , the determinable and determination. Perhaps this does not appear sufficiently
clearly in Kant: according to him , two of the three moments remain as
extrinsic characteristics (if Ideas are in themselves undetermined , they are
determinable only in relation to objects of experience , and bear the ideal
of determination only in relation to concepts of the understanding. (DR
170/220-1 , my emphasis , D. V.)
Before we turn to Dele l1 ze's theory of Ideas that is heavily inspired
by differential calc111us , we will look at the role of Ideas in Kant's
third Critique. In The Critique of the Power of Judgment Kant
considers different ways in which rational Ideas can be presented in
sensible nat l1 re , for instance the presentation of Ideas in the bea l1 tif111
of nat l1 re , in the sllblime , and finally the aesthetic Ideas of the genius.
Daniel Smith arg l1 es that in this way rational Ideas enter into an
'operatve relatonshi with reflecting jlldgernent , which in contrast
to determining j udgement is made without concept and allows for a
free and indetern1inate play of the faculties. 18 'e wiU have to spell O l1 t
156
ld5
as Problems
what this operative relationship between Ideas and re f1 ecting judgement consists in. This will result in a very different picture to the one
that was presented in the Critique of Pure Reaso where the Idea or
what is only in the idea [works] as a ground for the harmonious use
of reason' (CPR A 693/B 721). Now , in the Critique of the Power 01
]udgment Kant uncovers a type of judgement which suspends with
the harmonious exercise of the faculties upon a supposed same object
and instead induces their discordant accord' (DR 146 /1 90). We will
have to show how Ideas are related to this paradoxical exercise of the
faculties that Deleuze characterises as a para-sense' (DR 146/190).
The Genetic Power of Ideas

Kantian Ideas of reason , as we have seen , represent the totality of
conditions or the ideal of a complete detern1ination in relation to
concepts of the understanding. Their object is a something in general ,
i.e. a complete systernatic unity with which we are not acquainted
in the empirical world. In other words , Ideas of reason are absolute
concepts without intuition. Nonetheless , Ideas of reason have a
good and irnmanent use for our faculty of cognition. They constitute
horizons or id 1 foci beyond possible experience towards which
the concepts of the understanding converge. In this way , Ideas of
reason endow our cognitions with a maximurn of systematic unity.
However , as has been pointed out , their misuse brings about deceptive illusions. Ideas of reason ought not to be taken as determinations
of objects , i.e. things in themselves. heir object is always 'indeterminate' and probleluatic' for our faculty of cognition. Yet reason
experiences an interest in fnding some ind of presentation of lds
in sensible nature. This is so because between the theoretical realm
of knowledge , i.e. sensible nature , and the practical or rnoral realm
of freedom , i.e. the supersensible , there is an incalculable gulf fxed
[. . .], so that from the former to the latter (thus by means of the
theoretical use of reason) no transition is possible' (C] 5: 175-6).
But if it is possible to show that sensible nature is suitable to express
or symbolise sOlnething supersensible or receive the effects of moral
concepts (e.g. the concept of freedom) , then theoretical and practical philosophy can be connected. In the Critique of the Power 01
]udgment Kant fnally seeks to provide a passage from theoretical
cognition of sensible nature to practical cognition of the supersensible in the subjec t. The key notion is that of reflecting judgement
which comprises both aesthetic and teleological judgements the two
157
parts of the Critique 0 1' the Power l' ]udgment. We need to rnaintain
some caution , since aesthetic j udgernents and teleological j udgernents
are two quite different types of reflecting judgernent. 19 However ,
what is cornrnon to both is that , in contrast to deterrnining j udgernents , reflecting j udgernents are rnade without a concet. They do
not seek the deterrnination of an object , but instead bear witness to a
free and indeterrninate play of the subjective faculties. This free and
indeterrninate play , rnanifested in reflecting j udgernent in general ,
anirnates and strengthens the faculties and thus prornotes the receptivity of the rnind for the rnoral feeling (C] 5: 197).
According to Deleuze , Kant develops in the third Critique a
genetic viewpoint that is capable of accounting for the genesis of the
relations between the faculties , an account which was stilllacking in
the Critique 0 1' Pure Reason and the Critique 0 1' Practical Reason.
his rneans that for the first tirne Kant atternpts to give a sufficient
reason for the presurned a priori fact that the faculties can agree
in a harrnonious accord in spite of their distinct natures. In the
first two Critiques , Kant appeals to faculties that are ready-rnade ,
whose relation or proportion he seeks to deterrnine , already supposing such faculties are capable of sorne harrnony' (DI 6 1/86). But in
the Critique 0 1' the Power 0 1' ]udgment Kant pursues the question as
to how the free rnutual harrnony of all the culties arises. Hence,
for Deleuze , the third Critique plays a fundarnental role within
Kant's opus of critical philosophy: Beneath the deterrninate and
conditioned relations of the faculties , it discovers free agreernent ,
indeterrninate and unconditional' (DI 69/98). Focusing on the
aesthetic part of the Critique 0 1' the Power 0 1' ]udgnzen Deleuze
works out
three parallel geneses: [. . .] the sublime, or a genesis of the reason-imagination agreement; purpose connected with the beautiful , or a genesis
of the imagination-understanding agreement according to the beautiful
in nature; and genius , or a genesis of the imagination-understanding
agreement according to the beautiful in art. (DI68/97)
According to Deleuze's reading , there is an accord of faculties only
on the ground that each faculty is capable of operating according
to its own nature and entering into a free , indetern1inate and also
discordant play. This is why , in Deleuze's view Kant's Critique in
general ceases to be a simple conditioning to becorne a transcendental Education , a transcendental Culture a transcendental Genesis'
(DI6 1/86).
158
Ideas as Problems
THE NOTION OF REFLECTING jUDGEMENT
The novelty in the Critique of the Power of]udgment that will allow
for a genetic viewpoint of the free play of the faculties is Kant's
notion of reflecting judgement. John Zammito has argued that this
idea of reflecting judgement occurred to Kant quite late when he was
already in the process of writing the third Critique. lnitially , the third
Critique was conceived by Kant as a Critique of Taste' , in which he
wanted to present a kind of a priori principle for the faculty of feeling
pleasure and displeasure and elaborate this in a critique of the beautifu 1. 20 Then in early 1789 , Kant came up with the notion of reflecting
judgement , presumably when he was busy with writing the First
Introduction to the Critique of the Power of Judgment. he fundamental distinction of reflecting from determining j udgements led to
the transformation of the Critiqlle of Taste' into the Critique of the
Power of Judgment , i.e. a work which con1prises not only aesthetic
but also teleological j udgements. It is assumed that a considerable
body of material (suc as the 'Analytic of the Sublime ~~2330)
was inserted into the already composed Critique of Taste' only after
Kant has worked out the implications of the new idea of a critique of
the power of jlldgement. 21
Kant defines the distinction between detern1ining and reflecting
judgement in the following way:
he
power of judgment in general is the faculty for thinking of the particular as contained under the universa l. If the universal (the rule , the principle , the law) is given , then the power of judgment, which subsumes the
particular under it [. . .], is determining. If, however , only the particular is
given , for which the universal is to be found , then the power of judgment
is merely reflecting. (CJ 5: 179)
Reflecting judgernent thllS ascends from the particular to the llnivers that is from the manifold of particular empiricallaws in nature
to equally empirical but higher principles , in order to ground the
possibility of a systematic unity in nature. Zammito points out that
already in the Critique of Pure Reason Kant had considered such
cases in which the particular is given and the universal concept is
assumed only problematically. There Kant ascribed to reason in
its 'hypothetical use' the task of nding the universal concept , and
distinguished it frorn theapodictic use' of reason which applies to
cases in which the universal is given and only judgement is required
for subsuming the partcu]ar under it (c f. CPR A 646/B 674).
159
1n making the distinction between the apodictic' lfld the hypothetical'

llse of reason in jlldgments , Kant came closest to anticipating the key distinction of re f1 ective from determinant jlldgment which he enllnciated in
the Frst lntroducton [to the Critique of the Power of Judgme t].22
Thus Zarrlmito's understanding of the third Critique is that in it
Kant is transferring the hypothetical function of reason to the reflecting power of judgement. 23 lt should be noted that in the first Critique ,
judgement was not considered a faculty in its own right but a function belonging to the understanding (CPR A 69/B 94 , A 8 1/B 107).
In the third Critique the faculty of judgement is presented as an intermediate faculty between reason and the understanding. 24 It f111fils
the role of a lllediator , or in other words of an overarching fac111ty ,
since it brings together all the other cognitive faculties to the end of
producing a single cognitive product , i.e. the judgement.
At this point, Deleuze offers a quite different reading of the
power of jlldgement, which to our knowledge has no predecessors.
According to Deleuze , the power of judgernent is not a further
single faculty in the family of higher faclllties of cognition. Instead ,
he argues that since judgements always imply several faclllties , the
power of judgelllent simply consists in their accord , 'whether an
accord already determined by one of thelll playing a legislative role
or , more profoundly , in a free indetern1inate accord' (KCP 5 1/87).
Delellze concl l1 des that Kant did not discover a 'new' interrnediate
faculty of jlldgerrlent but only elaborated the accollnt of jlldgement
he had already given. he novelty of Delellze's reading is that the
power of judgement can no longer be interpreted as a spontaneo l1 s
and innate faculty. On the contrary , if the power of judgement consists in an accord of faclllties , the q l1estion arises how the accord is
effectively engendered? Deleuze distinguishes three general types of
accord of the cognitive faculties: the logical comlllon sense , the moral
common sense and the aesthetic common sense. His interpretation is
arguably still in line with what Kant himself says. For Kant certainly
makes use of the notion of 'comn1on sense' (sensus communis) ,
by which 'lllllSt be understood the idea of a cornmunal sense , i.e.
a faculty for jl1 dging' (C] 5: 293). Furthern1ore , he discriminates
between a logical cornmon sense (sensus conl1nunis logicus) and an
aesthetic con1mon sense (sensus cO n'zmunis aestheticus) ( 5: 295).
7hile the aesthetc corrlmon sense represents the effect of the free
play of our cognitive powers' (C] 5: 238) , the relationship of the
faculties in the logical common sense is 'lawful , l1 nder the
160
ldeas as Problems
of dete 1' minate concepts' (C] 5: 295). In othe 1' wo 1'ds , the logical
common sense designates an a p 1' io 1'i acco 1' d of imagination , unde 1'standing and 1'eason in which unde 1'standing is the legislative faculty
assigning to each faculty its specific task. Although the acquisition of
knowledge is actually in 1'eason's speculative inte1'est, pure 1'eason
leaves eve 1'ything to the unde 1' standing' (CPR A 326/B 383 -4), since
it cannot apply its principles immediately to objects of expe 1'ience.
With 1'espect to Kant's Critique of Practical Reaso Deleuze further
distinguishes a mo 1'al common sense' , i.e. an a p 1'io 1'i acco 1'd of
unde 1' standing and reason with 1'eason being the legislative faculty.
In sum , 'saying that judgement dete 1' mines an object is equivalent
to saying that the acco 1' d of the faculties is deterrnined , 0 1' that one
of the faculties exe 1'cises a detennining or legislative function' (KCP
50/85).
In the following we will look at Kant's account of the f1'ee and
indete1' minate account of the faculties with 1'ega 1' d to the expe 1'ience
of the beautiful , the subli 1l1e and the p 1' inciple of genius , and conside 1'
the inte1' vention of Ideas in the respective 1'eflecting judgements.
EXPOSITION OF AESTHETIC ]UDGEMENTS OF TASTE
In the 1'st two Critiques , Kant p 1'esupposes an a p 1'io 1'i dete 1' n1inate
acco 1' d between the cognitive faculties , that is a ha 1' monious 1'elationship with fixed p 1'opo 1' tions depending on the p 1'ed Olninant inte1'est
of 1'eason and the pa 1' ticula 1' legislative faculty in this given inte 1'est:
this agreement is always proportioned, constrained, and determinate' (DI 57/81). But by what 1'ight can we assun1e such an acco 1'd as
an a p 1' iori fact , since the faculties a 1'e essentially distinct in natu 1'e?
The faculties must fi 1' st of all , by themselves and spontaneousl) be
capable of an indeterminate acco 1' d without legislation. Now , in the
Critique of the Power of ]udgment , Kant envisions such a f1'ee and
indete 1' minate accord in the aesthetic common sense.
The aesthetic judgement, for instance 'this lily is beautifu l', judges
the 1'ep 1'esentation of a single object independent of concepts with
1' ega 1' d to the pleasu 1' e it a 1' ouses. This feeling of pleasu 1' e should not
i confused with the en1pi 1' ical satisfaction we expe1' ience when the
1' epresentation of an object is 'ag 1' eeable' to ou 1' senses and suitable
fo 1' our interests.
Any interest spoils the judgment of taste and deprives it of its impartiality
[. . .]. Taste is always still barbaric rvhen it needs the addition of charms
161
and emotions for satisfaction , let alone if it makes these into the standard
for its approva l. (C] 5: 223)
Kant is concerned with a higher form of pleasure , which is a completely disinterested pleasure (C] 5: 205) , not in the least biased by
the existence of the object and the sensible attraction and charm
elicited by it. For in the latter case , the aesthetic judgement of taste
would remain private; it iVould only bear on the personal liking of
the one who is immediately aHected. But the beautiful is that which
pleases universally and the pleasure that it induces is expected of
everyone else as necessary (C] 5: 218-19). JudgelTIents of taste lay
claim to universal consent and necessity. The difficulty for Kant will
be to demonstrate this universality and necessity of judgements of
taste , since from the logical point of view aesthetic judgelTIents are
singular propositions and are not grounded in any conceptualisation.
he universality and necessity they express must therefore be of a
special kind. In fact , the aesthetic universality can only be 'subjective'
because the predicate of beauty is not connected with the concept of
the object considered in its entire logical sphere , and yet it extends
it over the whole sphere of those who j udge' (C] 5: 215). Aesthetic
judgements require an intersubjective validity i.e. a universal assent
of the whole sphere of j udging subjects , which , however , must at the
same time be reconcilable with 'an autonomy of the subject judging
about the feeling of pleasure in the given representation' (C] 5: 218).
The con1pulsion toward an agreement can therefore only be based
on an indeterminate concept. This means that deterrninate concepts
of the understanding , which can be applied to objects of sensible
intuition through schemas , are excluded from aesthetic judgements.
Furthermore , the necessity of an aesthetic judgement can only be
called 'exemplarinsofar as it only demands the assent of everyone
at the example of a singular instance (C] 5: 237). Hence , with regard
to the judgement 'this lily is beautiful' , we assume that our aesthetic
experience of pleasure is communicable to and valid for everyone ,
and consequently that everyone feels compelled to agree with it. But
what exactly is the determining ground for the COlTIlTIUnicability of
my pleasure and the universal consent? he basis for the judgen1ent
of taste cannot be sought in the subjective private condition
162
Ids
as Problems
himself into the position of everyone else , merely by abstracting from

the limitations that contingently attach to our own judging; which is in
turn accomplished by leaving out as far as is possible everything in one's
representational state that is matter , i.e. , sensation , and attending solely
to the formal peculiarities of his representation or his representational
state. (CJ 5: 294)
Thus the material existence of the object plays no role in grounding
the judgement of taste. The higher form of pleasure , which accompanies the judgement of taste , is no immediate result frorn the object in
its sensual richness but rather from the reflection upon the form of the
object, i.e. its formal elements - the design or the composition (C] 5:
225). More precisely , the aesthetic form' is not simply conceived as a
property of the object, but is merged with the reflection of the object
in the irnagination. The detenllining ground for the pure judgement
of taste is thus the formal purposiveness of the object in relation to
the subjective faculties. As Kant says , the determining ground
can be nothing other that the state of mind that is encountered in the relation of the powers of representation to each other insofar as they relate
a given representation to cognition in genera l. The powers of cognition
that are set into play by this representation are hereby in a free play , since
no determinate concept restricts them to a particular rule of cognition.
(C] 5: 217)
Hence the reflection upon the aesthetic form of a given object, its
design or composition , occasions a purposive relationship of the subjective faculties , that is a certain state of mind , which is distinguished
through its universal cO l1lInunicability. What happens is that the
imagination is left free to synthesise according to its own lawfulness
without being constrained to schematise according to concepts of the
understanding (a 'lawfulness without law'). lmagination freely corresponds to the understanding as the faculty of concepts in general ,
such that there is a subjective correspondence of the imagination to
the understanding without an objective one' (C] 5: 241). According
to Deleuze , imagination in its free spontaneity does not schematise:
schematislll is always the act of an irnagination which is no longer
free , which finds its actions determined in conformity with a concept
of the understanding' (KCP 41171). At one place , Kant defines the
freedom of imagination as 'the fact that it schematizes without a
concept' (CJ 5: 287). Deleuze , however , insists that 'schen1atism
without a concept' is a bad choice of expression. The free and
spontaneous exercise of imagination consists in its re f1 ection upon
163
the object and its productive inventiveness. It appears that Kant

agrees that irrlagination considered in its freedorn cannot be taken as
reproductive , as subjected to the laws of association , but as productive and self-active (as the authoress of voluntary forms of possible
intuitions)' (C] 5: 240). Kant provides as examples English landscape
gardens or baroque furniture whose extravagant variety of forms
pushes the freedom of the inlagination almost to the point of the
grotesque' (C] 5: 242). In these cases , imagination is freed from
stiff (mathematica l) regularity that comes across as constraint and
induces boredom. Likewise , in natural instances such as the changing shapes of a fire in a hearth or of a rippling brook' imagination
sustains its free play (CJ 5: 244). Kant concludes that
taste seems to fasten not so much on what the imagination apprehends in
this field as on what gives it occasion to inve i.e. , on what are strictly
speaking the fantasies with which the mind entertains itself while it is
being continuously aroused by the manifold which strikes the eye. (C] 5:
243)
In the aesthetic comnlon sense imagination is thus discovered in its

original exercise: free reflection and spontaneous , inventive produc
tion. It relates to the understanding in a non-specified correspondenc that is it nds itself in agreernent only with indeterminate
concepts of the understanding (KCP 41/71).
In sum , in the First Book of the Analytic of the Beautiful , Kant
presupposes the indeterrninate aesthetic comnlon sense as the ground
for us to produce a determinate common sense. But he has not yet
attended to the question whether the aesthetic common sense or
taste is an original and natural faculty , or only the idea of one that is
yet to be acq uired and is articial [. . .] - this we would not and cannot
yet investigate here; for now we have only to resolve the faculty of taste
into its elements and to unite them ultimately in the idea of a common
sense. (C] 5: 240)
In Deleuze's reading , Kant occupies hinlself in the Analytic of the
Beautiful with the task of exposition and he postpones the crucial
question , whether the aesthetic common sense is natu 1'ally given 0 1'
whether there is a principle or Idea that can provide us with a rule for
producing the aesthetic common sense in us. According to Deleuze ,
Kant already iInplicitly suggests that it is not sufficient to assume
the aesthetic common sense as a natural fact , but that there needs to
be an explanation of how it is engendered in the soul according to
164
Ideas as Problems
a principle or Idea. Indeed , it appears that Kant presupposes a kind
of 'indeterminate norm' (CJ 5: 239) for the genesis of the aesthetic
comrnon sense. 25 However , a definite answer to the problem of the
genesis of the aesthetic cornmon sense can only be given in a deduction , that is a deduction of aesthetic judgement. Contrary to the
deduction of synthetic a priori j udgen1ents in the Critique of Pure
Reaso the deduction of aesthetic judgelnents is not concerned with
the right of conditioning or necessary subjection of given intuition
under a priori concepts (quid juris?).
Instead , the problem is now one of deducing the genesis of the agreement
among faculties: this problem could not make its appearance as long as
one of the faculties was considered legislative with respect to the others ,
binding them in a determinate relation. (DI 61186)
According to Deleuze , Kant must , however rst turn to the aesthetic
judgernents of the sublime in order to find the key for the required
deduction of judgernents of taste. Deleuze wiII offer a novel reading
which explains the confusing order of sections in the Critique of the
Power of Judgment - for instance the insertion of the Analytic of
the Sublime (~~23-30) between the Analytic of the Beautiful and the
Deduction of the judgn1ents of Taste , which in its turn is followed by
~49 which contains the analysis of genius. Deleuze explains the order
of sections from the systematic point of view of a problem , namely
the problem of a transcendental genesis of the relationship between
the faculties.
HE SUBLIME AND THE MODEL OF GENESIS
With regard to the experience of the sublirne Kant wi l1 explore the

relation of aesthetic judgements to Ideas of reason , and envision the
possibility of nding a presentation' of rational Ideas in sensible
intuition. In fact , he wiII show that aesthetic judgements , in particular the judgernent of the sublime , lnake us feel the reality of the Idea
of a supersensible substrate of humanity , and in this sense reveal the
animating principle which engenders the free original exercise of
each faculty and their reciprocal accord. That will emerge from an
exposition of the experience of the sublirne is that the accord of the
faculties s the product of a veritable transcendental genesis.
In the experience of the sub 1i me , the imagination is confronted
with something 'tuhich is great beyond all conzparison' (CJ 5: 248 ,
emphasis in origina l) for nstance with the Inn1ensty of a certain
165
representation , its formlessness or violent power. Although this

overwhelming experience is occasioned by the presence of certain
representations of natural objects , the sublime is str tly speaking
not the property of an object. It is only by means of a subreption'
that we ascribe to an object in nature what in fact has its source in
the self (C] 5: 257). However , Kant's account of the sublne in some
way resorts to this subreption , since it testifies to our capacity of
representing sublirnity in objects. 26
Kant distinguishes two types of the sublin1e - the mathen1atically
and the dynarnically sublime - and refers each of them to an aspect
of reason with its corresponding Ideas. Thus the mathernatically
sublime brings theoretical reason and the principle of 'totality' into
play , the dynamically sublime practical reason and the principle of
'autonorny\Let us begin with the mathematically sublirne. What
happens is that in the face of an imlnense representation (e.g. the
starry heaven above US 27 ) , reason poses a claim for absolute totality and demands of the imagination that it represents the imlnense
magnitude in one intuition. The underlying problem is that of
m nity. While reason is capable of completely comprehending the
infinite under one concept , i.e. an Idea of reason (C] 5: 255) , the
imagination fails to offer an intuitive whole of sOlnething infinite or
absolutely great (such as infinite space or past time). The aesthetic
estimation of a magnitude (which as aesthetic' proceeds independently of nun1erical concepts) involves two actions of the faculty of
imagination: apprehension and com'ehension (C] 5: 251). he
task of apprehension poses no specific problem; it can go on indefinitely. In other words imagination is able to take in spatio-ternporal
units ad infinitum. But a difficulty arises when the imagination
attempts to achieve con1prehension , that is to encompass the partial
representations in one individual intuition.
For when apprehension has gone so far that the partial representations
of the intuition of the senses that were apprehended first already begin to
fade in the imagination as the latter proceeds on to the apprehension of
further ones , then it loses on one side as much as it gains on the other ,
and there is in the comprehension a greatest point beyond which it cannot
go. (C] 5: 252)
Thus the faculty of irnagination 'soon reaches its maximurn' (C] 5:
252) and finds itself unable to fulfil the demand of reason to present
the Idea of a whole. Pushed to the limits of its power , imagination is confronted with its inadequacy for estirnating an imn1ense
166
Ideas as Problems
magnitude. We experience the impotence of the imagination as an
intense feeling of displeasure or pain. But imagination's striving to
advance to the infinite' and its subsequent failure awakens the feeling
of a supersensible faculty in us' (C] 5: 250) , namely reason , which
by means of its rational Ideas has the power to overstep the limits of
sensibility. In rnaking intuitable the superiority of the rational vocation of our cognitive faculty over the greatest faculty of sensibility' ,
the imagination discovers its own true vocation (C] 5: 257). Thus
even the irrlagination has a supersensible destination
for the imagination, although it certainly finds nothing beyond the sensible to which it can attach itself, nevertheless feels itself to be unbounded
precisely because of this elimination of the limits of sensibility; and that
separation is thus a presentation of the infinite, which for that very reason
can never be anything other than a merely negative presentation, which
nevertheless expands the sou l. (C] 5: 274)
Hence the con f1 ict between reason and imagination in which imagination seems to succumb by admitting its irrlpotence actually makes
the unattainability of rational Ideas in sensible nature palpable for
us and thus induces a feeling of the superiority of reason. Thereby
irrlagination achieves an (albeit rnerely negative) presentation of
rational Ideas and the con f1 ict of imagination and reason gives rise to
subjective purposiveness:
For just as imagination and u derstanding produce subjective purposiveness of the powers of the mind in the judging of the beautiful through their
unison , so do imagination and reason produce subjective purposiveness
through their conflict: namely, a feeling that we have pure self-sufficient
reason , or a faculty for estimating magnitude , whose pre-eminence
cannot be made intuitable through anything except the inadequacy of
that faculty which is itself unbounded in the presentation of magnitudes
(of sensible objects). (C] 5: 258)
he
violent relation between reason and imagination is resolved

into a discordant concord , a harn10ny in pain' (DI 62/87); in
Kant's words , imagination and reason are harmonious even in their
contrast' (C] 5: 258). This han110ny gives rise to a pleasure 'that is
possible only by 111eans of a displeasure' (C] 5: 260) and therefore
arnounts to what can be called a 'ne
167
boundless and turbulent ocean or rnassive waterfa l1 s (C] 5: 261).

hese natural forces rnake us aware of our physical powerlessness ,
endanger our possessions and threaten our health and life. Through
these violent irnpressions irnagination is pushed to the point where
he rnind can lnake palpable to itself the sublirnity of its own vocation over nature' (C] 5: 262). We rnight be abased as natural beings ,
but we experience a self-preservation of quite another kind' (C] 5:
261) , narnely as rnoral beings. We feel superior to nature within
us , and thus also that outside us' (C] 5: 269). Equally , just as in the
case with the rnathernatically sublirne , the irnagination first suffers
a deprivation of its freedorn. It is used as an instrurnent of reason
and its ideas' (C] 5: 269) and forced to treat nature as a scherna
for thern' (C] 5: 265). However , by being subjected to this violence
of reason , irnagination is finally endowed with a rnuch greater power
than it sacrifices: it is freed frorn con nernent within the borders of
sensibility and undergoes a necessary enlargement to the point where
it is capable of giving a negative presentation of the rnoral Ideas of
reason. Thus reason exercises a dorninion over in1agination n order
to enlarge it in a way suitable for its own proper dornain (the practica l) and to allow it to look out upon the infinite , which for sensibility
is an abyss' ( 5: 265).
The j udgernent of the sublirne , although it seelns to be instilled by
something absolutely great in nature , is in fact initiated by reason
and its Ideas. It expresses a 'predisposition to the feeling for (practical) ideas , i.e. , to that which is n10ral' (C] 5: 265). he j udgen1ent
on the sublirne thus has its 'foundation in h Ulnan nature' , or more
concretely in the supersensible substrate of the subject (the sou l).
It is irnportant to note , however , that , although the predisposition
to the feeling for Ideas of reason is already anchored in the soul , Kant
clairns that for judgernents of the sublin1e a certain culture or refinernent is required to rnake our rnind susceptible to Ideas of reason. 28
In Deleuze's reading , this culture' is not to be understood as an
education delivered by school , the church or other institutions. The
sublirne is not an affair of sorne ernpirical and conventional culture;
but the faculties which the sublin1e puts in play point to a genesis of
their agreenlent within in1rnediate discord. his is a transcendental
genesls , n
168
Ideas as Problen1s
reason is engendered in discord. Pleasure is engendered in pain' (DI

62/88). Thus the transcendental genetic condition of the harmonious
relationship of the faculties irnposes itself from the outside: it is the
exterior of thought, or the unthought in thought which is unthinkable but must be thought nevertheless. By contrast, Kant grounds the
transcendental genetic condition again within the subject in saying
thatthe lnifying point of all our faculties a priori' is in the supersensible , the sou l. For Kant , 'no other way remains to make reason selfconsistent' , that is to maintain the unity of reason (of theoretical and
practical reason) (CJ 5: 341). Kant conceives the sou l', i.e. the focal
point to which all faculties converge, as a genetic subjective principle
(unconditioned and self-grounding) from which each faculty extracts
its original force and unfolds a free , spontaneous exercise independent of any constraints of common sense , that is unbound by ernpirical
representat lO n.
For the case of the sublime Deleuze discerns a genetic model of the
relations of imagination and reason. Now he poses the question of
whether sirnilar geneses of the relations of the faculties are possible
for the subject elsewhere. If this were so , the model of genesis could
be extended to the other faculties and each faculty would have its
free , original exercise. 1n Kant's deduction of judgements of taste ,
Deleuze ses precisely this: a genetic account of the free , spontaneous
exercise of imagination , understanding and reason.
SYMBOLISM IN NATURE
Once Kant has worked out the theory of the sublime and along with
it revealed our capacity to represent sublimity in objects , he is able to
offer a transcendental deduction of j udgements of taste. he crucial
insight is that beautiful objects , or rather objects that are reflected
formally in imagination and deemed beautiful , can be taken to
symbolise 1deas of reason. This means that in the symbol a relation
is secured between the supersensible and the beautiful object in our
intuition. The possibility of symbolisation in nature provided Kant
with the opportunity to bring reason into his account of the beautiful
that he had previously grounded on a free play between imagination
and understanding alone. But he did not sufficiently explain how it
comes about that imagination is left free to synthesise and associate ,
while the understanding is only involved as s requisite for a cognition in genera l', 'without presupposing a determinate concept' (CJ 5:
217-18).
Kant suggests that judgements of taste have to do
169
CONDITIONS
OFHOUGHT:
with indeterminate concepts of the understanding and that it is only

through the intervention of reason that the understanding becornes
indeterrninate and irnagination free. In other words , reason will
provide (through its Ideas) a transcendental grounding for the free
harmony of the faculties in the sense of beauty.
We can easily see why j udgements of the sublirne are in no need of
a transcendental deduction: 'The sublime in nature is only improperly so called , and should properly be ascribed only to the manner of
thinking , or rather to its foundation in hUlnan nature' (C] 5: 280).
hat is to say , the sublime has its ground only in the relations of
the subjective faculties of our mind; it does not rely on any object
outside of us. o be sure , the feeling of the sublime is occasioned by
the presence of an immense or powerful representation of an object ,
but this object is itself completely unpurposive and the sublime is
only projected' upon it. In reality , judgements of the sublime are an
expression of the purposive relation of the cognitive faculties. The
analysis of the feeling of the sublime demonstrates how these relations are effectively engendered , and this account is simultaneously
a deduction of j udgements of the sublime. The case of j udgements
of beauty is more complex. The purposiveness of the relations of
the faculties in the experience of natural beauty is inseparable froln
the re f1 ection on the form of the object. Although it is true that this
relation to the object is merely one of re f1 ection and not of determination (as in judgements of cognition) , nevertheless this relation to the
object is enough for judgements of taste to require a special warrant ,
that is a deduction for their claim to (subjective) universality and
necessity. For this reason , Kant has to provide grounds for the free
play of the faculties within the re f1 ecting judgement on beauty and
the high feeling of pleasure connected with it.
ln the transitional paragraph (~30) that bridges the account of
the sublime and the deduction of pure aesthetic judgements , Kant
turns his attention to a striking feature of nature: the abundance
of beautiful objects which are scattered even in the depths of the
ocean , where the h Ulnan gaze hardly ever reaches (C] 5: 279). The
subjective purposiveness of the beautiful formations suggests the
existence of inner ends in nature. Kant , however , categorically rules
out the possibility of a reali
170
Ideas as Problems
when they condense or solidify (as for instance the formation of a
crystal) (C] ~58). Nevertheless , he acknowledges that the aptitude
of nature to produce beautiful objects deserves our attention. Nature
apparently has the positive property of instilling an awareness of the
internal purposiveness of our faculties through its beautiful objects.
In Kant's words nature has the property of containing an occasion
for us to perceive the inner purposiveness in the relationship of our
mental powers in the judging of certain of its products' (C] 5: 350).
Thus the internal accord of our faculties actually implies an external accord between nature and our intellectual faculty , albeit this
external accord can only be contingent: he agreernent thus has no
go (Dl 64/90). There cannot be a necessary subjection of nature ,
because in this case the aesthetic j udgernent would proceed by determinate a priori concepts and no longer be merely h utonomous
(l egislating only for itself). In any case , reason has a profound interest
in this contingent accord between nature and the subjective faculties.
It is of great importance
that nature should at least show some trace or give a sign that it contains
in itself some sort of ground for assuming a lawful correspondence of its
products with our satisfaction that is independent of all interest [. . .],
reason must take an interest in every manifestation in nature of a correspondence similar to this. (C] 5: 300)
Through the abundance of beautiful and free formations , nature
proves to be inherently qualified for the presentation of the concept
of ends or moral Ideas of reason. In Deleuze's reading , this intellectual interest of reason is a neta-aesthetic interest' synthetically
connected with our judgement of taste , which is nonetheless entirely
disinterested in the material existence of the object. The intellectual
interest is not an interest in beauty as such; it remains external to the
j udgement of taste. In other words , esthetic pleasure is disinterested,
but we feel a rational purpose when the productions of nature agree
with our disinterested pleasure' (DI 65/92).
The n1eta-aesthetic interest of reason is primarily attached to the
free materials' of nature , that is the f1 uid substances , or colours
and sounds. As Kant admits the charn1s in beautiful nature , which
are so frequently encountered as it were lnelted together with the
beautiful form , belong either to the modifications of the light (in the
colouring) or of the sound (in tones)' (C] 5: 302). It s remarkable
that Kant at this point no longer abstracts from the :harms' of an
object , its colour and sound. I-l itherto it appeared that he favoured a
171
kind of forrnalism' in art, since he insisted with respect to pictorial

or plastic arts that it is always the design that counts , or with respect
to music the formal structure of cOlnposition. Now he enlarges his
(0 mal aesthetic of taste by adding what Deleuze calls a nl
late
7 l
me
ta .
and sounds as the decisive features associable W11th rno)

ral Ideas. Kant
takes as examples the song of a bird symbolising joy and contentment with its existence , and the variety of colours syrrlbolising di
ferent virtues. What exactly is the role of the rnoral Idea here? Let us
consider one of Kant's exan1ples in greater detai l. When a white lily
is taken as a syn1bol of pure innocence , the concepts of whiteness'
and lily' are thereby artificially enlarged into indeterminacy (or as
Freud would put it , the concepts of the understanding become overdetermined'). In any case , through the intervention of reason and its
moral Idea the understanding exhausts its capacity of determination ,
and imagination is set free to associate new representations which
are not given in the concepts of 'whiteness' and lily'. At one point ,
Kant describes this procedure of imagination as 'the transportation
of the re f1 ection on one object of intuition to another , quite different
concept , to which perhaps no intuition can ever directly correspond'
(C] 5: 353). In other words imagination lnanages to render moral
Ideas sensible by fnding an indirect resentation of the morally
good. Kant also calls this indirect presentation of rational Ideas in
sensible intuition 'syn1bolisation' or ypotyposis'. However , the
d ef-nition he gives of the procedure of 'hypotyposis' again puts much
emphasis on the formal aspect and explains it in an analogy with
schematisation:
To a concept which only reason can think , and to which no sensible intuition can be adequate , an intuition is attributed with which the power of
judgment proceeds in a way merely analogous to that which it observes
in schematization , i.e. , it is merely the rule of this procedure , not of the
intuition itself, and thus merely the form of the reflection , not the content,
which corresponds to the concep t. (C] 5: 351)
According to this defnition , we abstract in symbolisation from the
perceptible content of a given object , only re f1 ect upon its forn1 and
use the forrnal rule or concept as a schema for an Idea of reason.
Kant sees here an analogy of symbol to schema. In schematisation a
rule , which specifies the concept , is applied directly to sensible intuition (in disregard of the particular content). In symbolisation , the
power of judgement has to perforn1 double task rst applying the
172
Ideas as Problems
concept to the object of a sensible intuition , and then , second , apply
ing the mere rule of re f1 ection on that intuition to an entirely different
object , of which the rst is only the symbol' (CJ 5: 352). For instance ,
the way in which a handmill functions , that is through the application of an external force , represents a n1erely externally constrained
mechanisrn and as such it is taken to symbolise the despotic rule of an
absolute lllonarch ruling only by his single V il l. On the other hand ,
the internal mechanisrn through which the soul governs the body is
taken as the forrnal rule for a purposive governance according to
internal laws and therefore is used as a symbol for a good and just
monarch. 29
In light of this definition of symbolisation and the given exarnples ,
it seen1S that a symbol is a kind of structural metaphor , in which only
the form or the structural organisation of an object is considered ,
then extracted as a rule and transported to another object that oth
erwise could perhaps not be given any intuition at al l. A freguently
used conternporary example of such a structural metaphor is the
symbolisation of politics or warfare by means of a sports game
(such as footbaU). Two things need to be said here as an objection
to Kant's definition of symbols in analogy to schemas. First, Kant
is not g uite right to parall1 the imagination's act of symbolisation ,
which proceeds in a loose accord with indeterminate concepts of the
understanding and is animated by indeterminate Ideas of reason ,
with the imagination's act of schelllatisation , which is conceived as
a rnere construction of deterrninate concepts in sensible intuition (as
in geometry , for instance). This is to say that the act of symbolisation is an auton OlllOUS and creative act of the imagination 30 which
is not c0111parable to the forced exercise of imagination in schematisation. 31 Second , Kant's previous examples of symbolisation , such
as the symbolic presentation of pure innocence in the object of the
white lily , certainly associate rational Ideas with the free materials
of sensible nature , that is colours , sounds , etc. Kant's n10ve to make
the re f1 ection on mere forn1 or structure and the principle of analogy
the basis for the imagination's free , spontaneous and productive
act of syn1bolisation is suspect. Symbolisation is rather comparable
to the creation of 'aesthetic Ideas' that Kant describes as the faculty
of genius. As we will see in t
173
faculties in the aesthetic cornrnon sense. 1n the exposition of aesthetic

judgements Kant had assumed a free , rnutual harmony between
imagination and understanding. Now , however , we can see that
reason must have already been involved from the start, that reason
indeed lies at the origin of the whole enterprise. This is because the
experience of the beautiful implies a moral disposition , or in Kant's
words
the beautiful is the symbol of the morally good , and [. . .] only in this
respect (that of a relation that is natural to everyone , and that is also
expected of everyone else as a duty) does it please with a claim to the
assent of everyone else , in which the mind is at the same time aware of
a certain ennoblement and elevation above the mere receptivity for a
pleasure from sensible impressions , and also esteems the value of others in
accordance with a similar maxim of their power of judgment. (C] 5: 353)
1n summary , the universal assent and feeling of pleasure connected
with the aesthetic judgement on a singular object have their transcendental ground in a subjective , yet universally valid , principle: the
free , indeterminate accord of the cognitive faculties that are animated
and enlivened by moral 1deas of reason and reason's interest in their
objective reality.
GENIUS AS THE FACULY FOR AESTITIC IDEAs
Kant's deduction of aesthetic judgen1ents of taste the principle of

which , as we have seen , is reason's lneta-aesthetic interest in nature's
production of beautiful objects and its suitability for symbolising
rnoral Ideas , left a problematic division between natural beauty and
the beautiful in art. Indeed , Kant denies that the beautiful in art provides any proof of a way of thinking that is devoted to the morally
good or even merely inclined to it' (C] 5: 298). He claims that there
is no inner afnity between the fine arts and a morally good way of
thinking. Art is either mere imitation of natural beauty , or is intentionally directed toward our satisfaction. hus Kant remarks that
the man in a museum , who 'gladly leaves the room in which are to
be found those beauties that sustain vanity and at best social joys
and turns to the beautiful in nature' bears witness to a beautiful soul
favourable to the moral feeling and deserves our esteem (C] 5: 300).
However , this separation of the beautiful in nature and in art is not
justified by the discussion in the Analytic of the Beautiful , where
Kant refers indiscriminately to aesthetic judgement on beauty as such
174
Ideas as Problems
and bases the feeling of pleasure it arouses on a free play of imagination and understanding. For this reason , Kant may have felt the need
to restore the unity of the two types of the beautiful in sections 58
and 59. He achieves this reunification by considering beautiful art as
the product of genius , the faculty for aesthetic Ideas , through which
art acquires its rule.
Kant distinguishes aesthetic Ideas from rational Ideas of reason.
While the Idea of reason is a concept (of the supersensible) to which
no adequate intuition can ever be given , the aesthetic Idea is an intuition for which no concept can be found adequate. The rational Idea is
an indemonstrable concept of reason , and the aesthetic Idea an inexponible representation of the irnagination (Cj 5: 342). Kant describes
the relation between both as oppositional , yet complementary: One
readily sees that it [the aesthetic Idea] is the counterpart (pendant) of
an id of reaso which is , conversely , a concept to which no intuition (representation of the imagination) can be adequate' (Cj 5: 314).
Contrary to Kant , Deleuze stresses that this is a false opposi
tion; there are not two sorts of Ideas' (DI 67/94-5). 'The aesthetic
Idea is really the Sa l11e thing as the rational Idea: it expresses what
is inexpressible in the latter' (KCP 48/82). He argues that for Kant
the objects of both aesthetic and rational Ideas are spiritual events'
(KCP 47/81) beyond the boundaries of experience. An Idea of reason
either denotes invisible beings , beings of heaven or heH , or it refers
to empirical events such as death or love and elerates them to events
of the spiri t. An aesthetic Idea , on the other hand , also expresses
son1ething inexpressible that cannot be found in nature , because it
invents a second nature whose phen0 l11ena are spiritual events (KCP
48/81-2).32
The imagination (as a productive cognitive faculty) is , namely , very powerful in creating , as it were , another nature , out of the material which the
real one gives it. We [. . .] transform the latter , no doubt always in accordance with analogous laws , but also in accordance with principles that
lie higher in reason (and which are every bit as natural to us as those in
accordance with which the understanding apprehends empirical nature);
in this we feel our freedom from the law of association (which applies to
the empirical use of that faculty) , in accordance with which material can
certainly be lent to us by nature , but the latter can be transformed by us
into something entirely different , namely inthat which steps beyond
nature. (CJ 5: 314)
Imagination is here described as a productive faculty that proceeds
freely in a spontaneous and 'transcenden t' exercise. It creates frOlTI
175
the material that it finds in nature a new and richer material (supplemented with further representations) and thereby aesthetically
enlarges the concept itself in an unbounded way' (C] 5: 315). Yet
the creation of a second nature' or an aesthetic Idea' requires more
than the rapidly passing play of the imagination' (C] 5: 317). What
is presupposed is he happy relation [. . .] of finding ideas for a given
concept on the one hand and on the other hitting upon the expression for these' (C] 5: 317) in order to make what is unnarneable in a
certain representation universally comrrlunicable. his talent is called
genius': it is a rather exceptional 'inborn predisposition of the mind
(ingenium) through which nature gives the rule to art' (C] 5: 307).
This does not mean that nature functions as the norrnative standard
that art has to imitate and that it can approximate to perfection if we
are skilled and diligent apprentices. Quite to the contrary , it means
that nature has equipped a few selected individuals with a certain
disposition of mental powers which cannot be attained by the follow
ing of rules but which is a natural gift' (C] 5: 307). The artist-creator
endowed with genius is able to create aesthetic Ideas that render
rational Ideas sensible and occasion much thinking , that is they give
more to think than can be grasped and made distinct in any determinate thought (C] 5: 314 and 5: 315). Aesthetic ldeas strengthen
our cognitive faculties and animate the mind by opening up for it
an immeasurable eld of representations related to a concept. The
creation of aesthetic Ideas itself is a procedure that for the main part
occurs unconsciously
and hence the author of a product that he owes to his genius does not
know himself how the ideas for it come to him , and also does not have
it in his power to think up such things at will or according to plan , and
to communicate to others precepts that would put them in a position to
produce similar products. (C] 5: 308)
According to Kant , the products of genius are charactersed by a
certain boldness in expression' and a deviation from the cornmon
rule' in the sense of a deformity'. However , it is not possible , as it
were , to read off from these products a rule or precept as to how
to produce art. Every atternpt to imitate the art of the genius will
becon1e a sort of 'aping and the spirit and boldness of the original
artwork will be lost in the in1itation , which w
176
Ideas as Problems
frorn coercion in his art in such a way that the latter thereby itself
acq uires a new rule' (C] 5: 318). Th us the work of a geni us casts an
appeal toward another genius across interrnediate spaces and times.
An intersubjective community of geniuses is thus constituted.
The question to ask here is , what function this Kantian romanticlsr concerning the genius has with regard to the overall aesthetic
project of finding an a priori and genetic principle for the free use
of the cognitive faculties. We can say that the genius also functions
as an a priori genetic principle for the unintentional subjective
purposiveness in the free correspondence of the imagination to the
lawfulness of the understanding' (C] 5: 317-18). However , the
genius stands on the side of the production of beautiful art. What is
1T10re , the genius is a very rare phenomenon being a avourite of
nature' (C] 5: 318). Thus , as Deleuze points out , 'we are faced with
this difficulty: how can such a genesis have a universal implication ,
if it is governed by the singularity of genius?' (DI 68/96). An answer
can be found in the distinction of two capacities that the artist or
genius ought to possess: one is spirit' , i.e. the animating principle
in the rnind that provides rich material for products of art. Spirit
is the capacity that constitutes the originality of the artist-genius;
it is given by nature as a gift and cannot be attained by the most
hard-working imitator and apprentice. However , spirit is only one
essential element of the character of the genius , because taste is also
required. The abundance of rnaterial has to be worked on and given
a form , and thus presupposes acadelnic training and the knowledge
of determinate rules (C] 5: 310).
Taste , like the power of judgment in general , is the discipline (or corrective) of genius , clipping its wings and making it well behaved or polished;
but at the same time it gives genius guidance as to where and how far it
should extend itself if it is to remain purposive; and by introducing clarity
and order into the abundance of thoughts it makes the ideas tenable ,
capable of an enduring and universal approval , of enjoying a posterity
among others and in an ever progressing culture. (C] 5: 319)
If the products of genius are to stand up to the power of judgernent,

then it is necessary that the artist-genius finds the form that not only
contents hirnself, but that is favourable to the judgelnent of taste and
therefore not detrimental to the freedon1 in the play of the cognitive
faculties on the part of the spectato r. If the artist-genius is able to do
both , create the materials and give forrn to his aI then the artwork
assumes a central place in culture. lndeed , art can then serve as a
177
rnodel , that is have an exemplary function as a standard or a rule for

j udging (CJ 5: 308). Thus it will instil in the spectator the feeling that
he is capable of freely and self-actively judging the beautiful object,
it wiU animate his faculties to a free accord , strengthen the lnind and
enhance its susceptibility for moral Ideas. In this way , the genetic
principle of genius , which pertains to an elect of nature , acquires the
value of a universal principle when considered in its effects on the
community of those who j udge.
We have seen throughout the preceding sections on Kantian ldeas in
the Critique of the Power of Judgment that Deleuze's interest focuses
on the different types of genesis of the proportion of mental powers
and the various ways in which rational Ideas can be presented in
sensible nature.
In the sublime the presentation is direct but negative , and done by projection; in natural symbolism or in the interest of the beautiful the presentation is positive but indirect , and is achieved by reflection; in genius or
in artistic symbolism the presentation is positive but secondary , and is
achieved through the creation of another nature. (KCP 48-9/83)
The crucial part of Deleuze's argument is that it is through the
encounter with rational Ideas rendered sensible by their incarnation
in nature that our faculties are animated to transgress the imposed
restriction of logical con1n1on sense and to act spontaneously and
freely within a discordant accord. Deleuze n1akes the aesthetic
common sense the object of a transcendental genesis (KCP 42/73). ln
all three cases of aesthetic judgement, reason brings its rational and
moral Ideas into play. ln other words , Ideas have a certain genetic
power that bears on the relations of the cognitive faculties and gives
rise to their paradoxical disjoint exercise in the aesthetic common
sense. In Deleuze words , ldeas , far from having as their milieu a
good sense or a comn1on sense , refer to a para-sense which determines only the communication between disjointed faculties' (DR
146/190). Daniel Smith argues that aesthetic j udgements provide
the rnodel for this operative relationship between ldeas and the
paradoxical exercise of the faculties:
It is precisely because of the intervention of the Ideas of reason (which
enlarge the concepts of the understanding) that the imagination becomes
free and the concepts of the understanding become indeterminate in
aesthetic judgments. 33
178
Ids
as Problems
Although Ideas of reason are never fully and positively incarnated

in nature , they nonetheless provoke the rrl nd's effort to rnake the
representation of the senses adequate to the Ideas (C] 5: 268). This
means that the ultirrlate unattainability' of cognising nature as the
presentation of something supersensible elicits the inventiveness and
creativity of imagination. Deleuze wiU retain this Kantian aspect of
the impossibility of cognition and transforrn it into the powerlessness' (DR 1471192) of thought to fully grasp the transcendental
elerrlent, i.e. the Idea or the problematic , which remains empirically
unthinkable. In fact , Kant discovered the Idea or the probleluatic
as the transcendental horizon , which essentially belongs to nature ,
things and events , and he also discovered the transcendent exercise
of the faculties and their discordant accord , which is elicited by the
contingent encounter with the transcendental element.
Deleuze wiU put to use all these Kantian aspects for his own conception of Ideas: the objective and problematic character of Ideas ,
their capacity of being presented in the sensible world (the sign') ,
their unattainability or transcendence fror l1 the point of view of
cognition , and their genetic power to release a transcendent exercise
of the faculties.
Deleuze's Dialectic of Ideas

Known for his 'anti-dialectic and 'anti-Iegelian' tendencies ,
Deleuze surprises the reader by presenting a dialectic' of Ideas in
Chapter Four of Difference and Repetition , entitled Ideas and the
Synthesis of Difference' ( Synthse idelle de la diffrence'). The use
of the term dialectic' is certainly n1eant provocatively but it also
implies the sincere resumption of a long tradition of dialectical
thought' albeit in a renewed manner. Deleuze is never simply against'
his ostensible enemies (Plato , Kant , Hegel , etc.): he rather rnobilises
son1e of their ideas and concepts and develops then1 further from the
point of view of a philosophy of difference or the differential'. As
Daniel Smith says:
In this manner , Delellze places himself sqllarely within the heritage
of his so-called enemies' the great philosophers of dialectics: Plato
and Aristotle , Kant , and Hegel md develops his concept of dialectics
throllgh them. In this sense , Delellze's reworking of dialectics extends
beyond both Hegel and Kan t. 34
179
According to Deleuze , 'dialectic is the art of problerns and

questions , the cornbinatory or calculus of problems as such' (DR
157/204). In his view , dialectical thought is seriously perverted when
the problematic' is rrl sunderstood:
Whenever the dialectic forgets' its intimate relation with Ideas in the
form of problems , whenever it is content to trace problems from propositions , it loses its true power and falls under the sway of the power of the
negative , necessarily substituting for the ideal objecticity [objectit] of
the problematic a simple confrontation between opposing , contrary or
contradictory propositions. This long perversion [dnaturation] begins
with the dialectic itself [that is , with Plato], and attains its extreme form
in Hegelianism. (DR 164/213)
Let us consider Aristotle's Dialectic , for exan1ple. Aristotle's Dialectic
deals with the manner of posing problems , whereas the Analytics is
concerned with the formal procedures by which syllogisllls reach
a conclusion. However , for Aristotle , the art of posing problellls
consists for the most part in making out the points of views that are
shared by the majority of people. hese generally shared opinions
form the places (the topoi) that serve as the pren1ises or the solution
of a problelll. As we have already seen , Deleuze criticises Aristotle
for falling prey to the natural illusion that is , for tracing problen1s
from propositions of con1mon sense (mere doxa). Deleuze also
blames Aristotle , as well as Kant and others for succmllbing to the
'philosophical illusion' on which we wi lI have lllore to say in the next
section on Deleuze's mathematical sources. But first let us now turn
briefly to Deleuze's criticism of Plato's , Kant's and Hegel's dialectic.
hat Plato recognised a profound relation between problems and
Ideas can readily be seen in the Socratic dialogue Meno. In the focus
of this dialogue is a geornetrical problelll that nally leads the disciple to 'remember' the original Idea. However , the Platonic theory
of problems is completely dependent on the classic master pupil
situation where the pupil understands and follows a problem only to
the extent that the rnaster already knows the solution and provides
the necessary adjunctions' (DR 180/233). According to this n10del ,
one ren1ains in an infantile state , in dependency on a n1aster who
sets the problelll , knows the solution and has the powerful authority
to determine the result as true or false (DR 158/205). he situation
seems different in Plato's aporetic dialogues , where the search for
truth that pursues the question What is X?' leads into contradictions.
The aporetic dialogues do indeed present problems without solution ,
180
Ideas as Problems
yet the aporia is in fact only a propaedeutic means for delineating a

domain of problems that will then be determined by other procedures
(cf. DR 188/243). It does not lead to a new conception or a new
rnanner of posing questions and problerns. Nevertheless , Deleuze
remarks in faTour of Plato that Plato's dialectic does not yet attribute
to the negative moment (the rnoment of contradiction) the decisive
role which it will play later in the Hegelian dialectic (DR 63/88).
Deleuze says that it is Kant who must be credited with rst stipulating the problern as the true object of the Idea and consequently with
discovering he real source of the dialectic' (DR 16 1/209). Ideas for
Kant are purely problematic. That is , they are no longer conceived as
transcendent essences as in Plato. However , as we have seen , Deleuze
insists that transcendental Ideas have to be released from the (transcendenta l) subject and their analogical relation to the elnpirical , in
order to acquire a status of 'ideal "objecticities [objectits]" possessing their own sufficiency and implying acts of constitution and
investment in their respective symbolic fields' (DR 159/206). One of
the i
mportant features of Ideas as obje
c 1' e structures is that they are
bot
h transcend
en
lt and im
una
nent. Ideas are fully irnmanent which is
to say that there is no gap between their ideal objectivity and their
inl1nanence in the ernpirical , historical world. There is an objectivity ,
Deleuze says , 'which implies that Ideas no nlore than Problems do
not exist only in our heads but occur here and there in the production
of an actual history' (DR 190/246). It follows that the question that
corresponds to Ideas cannot be that of essence at is X?' but has
to be diversified according to the incarnation of Ideas in empirical
circumstances. The leading questions are therefore 'who?' , when?' ,
'in which cases?' ow rom what viewpoint6hese q uestions
are those of the accident , the event , the 111ultiplicity of difference as opposed to that of the essence , or that of the One , or those of the
contrary and the contradictory' (DR 188/244). This is why Deleuze
finds the nature of the dialectic distorted when it is based on the
Platonic question What is X?' The answer to the question What
is X'is always God as the locus of the combinatory of abstract
predicates' (DR 188/243).
Deleuze's criticism 0
181
the real , when it compensates for the inadequacy of a concept that is

too broad by invoking the opposite concept, which is no less broad
and general?' (B 44/38). In fact , nothing moves in the dialectical
movement as conceived by Hege l. It is a false movement - in other
words , the abstract logical movernent of "mediation'" (DR 8/16).
Opposition and contradiction cannot be the moving force of the
dialectical process; rather they seemed to us to be surface effects
and conscious epiphenomena , while the unconscious lived on problems and differences' (DR 268/344). Deleuze especially criticises the
Hegelian dialectic for substituting the negative for the problematic
and differentia l.
However , the history of dialectical thought does not stop with
Hege l. Deleuze points to the largely unknown mathematician and
philosopher Albert Lautman (1908 -4 4) who - inspired by re f1 ections
on Plato and Heidegger - proposed the concept of 'dialectical Ideas'
which display an ideal reality distinct frorn the factual existence of
lnathernatical facts , objects and theories. Lautman's principal thesis
is that mathematics participates in a dialectic that dominates it. 35
In his essay Nouvelles recherches sur la structure dialectique des
lnathmatiques' (1939) , he sets out to analyse the nature of this
domination in terms of abstract Ideas and their concrete realisations in mathen1atical theories. As we wiU see , Lautlnan had a great
impact on Deleuze. lndeed , we must say that the characterisation of
ldeas as differing in kind from mathematical theories or solutions
and as being both transcendent and immanent con1es originally from
Lautman (cf. DR 178-9/231 2).
Nowhere better than in the admirable work of Albert Lautman has
it been shown how problems are first Platonic Ideas or ideal liaisons
between dialectical notions , relative to eventual situations of the exist
ent'; but also how they are realised within the real relations constitutive
of the desired solution within a mathematical, physical or other field. It is
in this sense, according to Lautman , that science always participates in a
dialectic which points beyond it in other words , in a meta-mathematical
and extra propositional power - even though the liaisons of this dialec
tic are incarnated only in effective scientic propositions and theories.
Problems are always dialectica l. (DR 163 -4 /212-13)
It is important to note that Lautman's version of Platonsm is , in

fact , very different from what is generally defined as Platonisn1 in
mathematics. For LautITlan , the realrn of ldeas is not separated from
its field of realisation through an irreducible rift such as that between
182
Ideas as Problems
the eidos and its representation , or the ideal model and its copy. 36
Lautman says:
We do not understand by Ideas the models whose mathematical entities
iVould merely be copies , but in the true Platonic sense of the term , the
structural schemas [schmas de structure] according to which the effective
theories are organized. 37
Thus Lautmanian Ideas are not exactly Platonic Ideas but structural
schen1as that establish possible relations between pairs of concepts. EXa111ples of such conceptual couples would be locallglobal' ,
intrinsi c/extrinsic' continuous/discontinuous nite/infinite' etc.
According to Lautrnan , we can have access to the realm of Ideas by
examining different mathematical theories that affirm the existence
of a common structural schema or dialectical Idea. This means that
through a descriptive analysis , relying upon concrete mathematical
exarnples , one can attain the realm of dialectical Ideas. 38 However
this is not to imply that dialectical Ideas are 111ere logical abstractions taken from the concrete. Instead they have to be understood in
a Heideggerian way as ontological questions or problems: Insofar
as "posed questions they [the dialectical Ideas] only constitute a
problematic relative to the possible situations of entities.'39 The type
of anteriority they involve is that of the question in relation to the
answer: 'It is of the nature of the response to be an answer to a question already posed , and this , even if the idea of the question comes to
mind only after having seen the response. o In his 'principal thesis' ,
a text published in 1938 , Lautman describes the type of anteriority
of dialectical Ideas as follows: 'The only a priori element that we
conceive is given in the experience of the exigency of the problems ,
anterior to the discovery of their solutions [. . .]. We understand this
a priori in a purely relative sense , and with respect to mathematics.'41
However , Lautman ernphasises that the logical schemas [schmas
logiques] of dialectical Ideas are lot anterior to their realization
within a theory' In other words , they are not anterior in the sense
of a chronological anteriority. Nor is the type of anteriority involved
that the logicists (lnen1bers and proponents of the Vienna Circle)
advocate , nan1ely the foundational anteriority of logic to lnathen1at
lCS According to this logicist view , the whole of mathernatics was
supposed to be reconstructed from a small nun1ber of primitive
logical concepts and propositions. As Simon DufT explains in his
essay on Lautman Lautrnan was strictly opposed to such logical
pOS lt1 Vlsm:
183
Against the logicist claim that the development of mathematics is

dominated a priori by logic, Lautman proposes a metaphysics of
logic' , and calls for the development of a philosophy of mathematical
genesls 44
Indeed , Lautman favours a very special type of anteriority of dialectical Ideas in relation to mathematics. He describes it in terms of a
relation of intrinsic genesis.
An intimate link [l ien intime] thus exists between the transcendence of
Ideas and the immanence of the logical structure of the solution to a
dialectical problem within mathematics. This link is the notion of genesis
which we give it [. . .] by describing the genesis of mathematics from the
Dialectic. 45
Lautman's concept of genesis y no means indicates a genesis that
takes place in one way only as in the case of Platonic Ideas. Rather
the matter is one of a reciprocal genesis. On the one hand , dialectical
Ideas have a productive powe r. One might say that the conceptual
couples produce a dialectical rnovement that gives rise to new theories. In Lautman's words: The genesis is then no longer conceived as
the material creation of the concrete fron1 the Idea , but as the advent
[venue] of notions relative to the concrete within an analysis of the
Idea. There are thus certain mechanisms 'in which the analysis of
Ideas is extended in effective creation , in which the virtual is transfonned into the real' .47 For instance , according to Simon Duffy , 'in
the case of the exarnple of the local-global conceptual couple , the
new n1athematical theory that was effectively created was Poincar's
qualitative theory of differential equations , or the theory of auton10rphic functions. '48 On the other hand , mathematical theories can give
rise to a dialectical Idea , insofar as they extract a logical schen1a or
pair of notions. As Lautman puts it:
But mathematical theories can conversely give rise to the idea of new problems that have not previously been formulated abstractly. Mathematical
philosophy , as we conceive it , therefore consists not so much in retrieving
a logical problem of classical metaphysics within a mathematical theory,
than in grasping the structure of this theory globally in order to identify
the logical problem that happens to be both defined and resolved by the
very existence of this theory
Lautman thus develops a new conception of metaphysics , a ITletaphysics relative to rnathematics , or as Simon Duffy calls it , a 'meta
mathematics in metaphysical terms\50 It is necessary to turn fr0111 the
184
Ids
as Problems
mathematical theories to the metaphysic i.e. the dialectic , which is ,

however , intimately connected with mathematics through a process
of intrinsic genesis.
he
importance of Lautman's theory of dialecticalldeas for Deleuze

has probably already become apparent. For Deleuze , Lautman's
work stands out as an original landmark in the history of modern
epistemology , for it recognises the double irreducibility of "problems'" (DR 323/211 , note 22/1) , that is to say , both the global formation of a transcendental problen1atic structure and its immanence
in local solutions. As Lautman puts it:
Insofar as posed problems , relating to connections that are likely to
support certain dialectical notions , the Ideas of this Dialectic are cer
tainly transcendent (in the usual sense) with respect to mathematics. On
the other har as any effort to provide a response to the problem of
these connections is , by the very nature of things , constitution of effective mathematical theories , it is justified to interpret the 0rerall structure
of these theories in terms of immanence for the logical schema of the
solution sought afte r. 51
The very status of dialectial Ideas as being both transcendent and

in1manent with respect to n1athematical theories resonates with
Deleuze's claim of the transcendence and imrnanence of problems
with respect to their solutions. Furthermore , Lautman as well as
Deleuze both insist on the 'ontological distinction' between dialectical ldeas-problems and their solutions and deterrrl ne the nature
of the relation between them as an intrinsic genesis. lnterestingly ,
Lautman had already used the tern1 virtual' to characterise the ideal
reality of dialectial ldeas and described the moven1ent of effective
creation as a transforrnation fron1 the virtual into the rea l. However ,
despite the obvious parallels between Lautman and Deleuze , there
are considerable differences as wel l. First of all , as Duffy has pointed
out , Ideas for Deleuze are more Kantian than Platonic in character.
They ha ve to be understood as purel) problelnatic. 52 Furthermore ,
we would argue that Lautman and Deleuze differ in the way in
which the structure of ldeas is conceived: while Lautlnan defnes
Ideas as structures constituted by contrary concepts , i.e. 'couples
of opposites which act as poles of tension within the structure ,
Deleuze conceives the structure of Ideas rather as an open and differential multiplicity. In this respect , we maintain that Deleuze is
more influenced by 1ain10n and Bergson than by Lautman. 53 A final
185
point needs to be mentioned which is perhaps even rnore irnportant

than the similarities and differences between Lautrrlan and Deleuze
that we have discussed so fa r. Duffy points to the fact that Deleuze
draws from Lautman 'his strategy of engagement with a range of discourses throughout his work 54 According to Lautl11an , mathematics
plays the role of a n10del for a theory of problellls: 1athematics
thus plays with respect to the other domains of incarnation , physical
reality , social realit) hU l1lan reality , the role of model in which the
way that things come into existence is observed.'55 In an earlier text ,
Lautman further explains: 'Mathematicallogic does not enjoy in this
respect any special privilege. It is only one theory among others and
the problems that it raises or that it solves are found almost identically elsewhere.'56 Deleuze completely aligns himself with Lautman
in claiming that mathematics is not in any way privileged (in the
sense that mathematics would provide the 'Platonic evidence' for a
superior dialectic). Mathernatics is just one domain an10ng others , in
which Ideas express then1selves technically. Ideas-problen1s belong to
a realm of their own , that is a dialectical realm. Problems are always
dialectical and
each dialectical problem is duplicated by a symbolic feld in which it
is expressed. hat is why it must be said that there are mathematical ,
physical , biological , psychical and sociological problems , even though
every problem is dialectical by nature and there are no non-dialectical
problems. (DR 179/232)
Deleuze takes up Lautman's suggestion and develops a theory of
Ideas-problems , the application of which is not only shown relative to mathematics but also in relation to other sciences (physics ,
biology) and cultural studies.
However , it cannot be denied that I1lathen1atics does play a particular role for Deleuze's theory of problems. This becon1es most apparent in the way that Deleuze uses mathematical concepts to expound
the 'calculus of problems' (concepts such as singularity , differential
relation , multiplicity adjunction of elds solution curves , etc.). We
will therefore have to examine the relation between philosophy and
rnathematics for Deleuze lllore closely. Does Deleuze c011ceive phi
losophy as a reflection 011 mathematics , as a philosophy of science?
Or does 111athematics s lllply provide examples that serve as illustrations for contentious philosophical issues? Is the use of rnathematical
termsrnerely' 111etaphorical? Or , on the contrary , does mathen1atics
play a foundational role for philosophy? Keeping these questions
186
Ideas as Problen1s
in rnind we will continue with a brief sketch of the rnathernatical
sources that Deleuze draws upon to develop his theory of Ideasproblerns. These rnathelnatical rnodels will prirnarily be differential
calculus and the theory of dynarnical systerns , the Riemannian notion
of multiplicity and Galois' theory of polynomial equations. For our
purposes it suffices to deal with the first two , and to pursue some of
the reflections undertaken in the previous chapte r. 57
DELEUZE'S MATHEMATICAL SOURCES: WEIERSTRASS , POINCAR ,
RIEMANN
For Deleuze , Ideas are essentially problems. Traditionally , problems

have been evaluated by the logical possibility of their being solved:
the truth of a problem consists only in the possibility that it receive a
solution' (DR 159/207). Under the heading of the seventh postulate
of the dogmatic irnage of thought , Deleuze describes this tendency
of basing the 'truth' of a problem upon its solvability (that is upon a
merely extrinsic criterion) as a philosophical illusion' (DR 159/207) ,
which he diagnoses in Aristotle and Kant among others. Fr Oln there ,
a fatal circle arises:
A problem is solvable only to the extent that it is true' , but we always
tend to defne the truth of a problem in terms of its solvability. Instead

of basing the extrinsic criterion of solvability upon the internal character
of the problem (Idea) , we make the nternal character depend upon the
simple external crterion. (DR 179/233)
Deleuze demands that the 'solvability' of a problem must depend on
its internal characteristic or structure: It lnust be detern1ined by the
conditions of the probleln , engendered in and by the problem along
with the real solutions' (DR 162/210). To use another Deleuzian
phrase: the solutions to a problem are intrinsically determined by the
distribution of singularities' .
In the previous chapter , in the discussion of Maimon , we have
already introduced some basic mathematical concepts and properties
of the differential calculus. We came across the intrinsic genetic force
of the differenti that is the possibility of generating a function from
a given differential relation. It was Karl 'eierstrass who developed
a method of local integration known as Weierstrass's theorern on
the approxirnation of analytic functions. According to this method ,
the successive derivations taken at a singular point can be written in
an infnite sum , a so-called Taylor series or power series expansion.
187
The sum of such a power series approximates the expanded func

tion in the neighbourhood of the singular point (provided that any
remainder approaches zero as the polynomial becomes an infinite
series). Furthermore , Weierstrass developed a method using circles
of convergence' , a circle of convergence being defined as the domain
around a singular point, in which an analytic function (representing
the expanded power series) is differentiable at each of its points.
Through the successive adjunction of more and more circles of con
vergence we can gradually construct the whole domain over which
the analytic function is continuous. 58 Weierstrass thereby showed
that the function generated by a first power series could be analytically extended over every path that does not encounter a pole that
is a particular kind of singularity at which the continuity breaks
down and the power series obtained diverge.
Lautman emphasises the specific role of singular points (c f. DR
324/230 , note 911) in this method: singularities do not only divide up
the domains in which the generated analytic function can be defined ,
they also allow the passage from the locally defined analytic function
to its global characterisation. Thanks to Weierstrass it was ultimately
possible to represent an entire analytic function that is continuous
over its whole domain. However , Weierstrass could not solve the
problem of the representation of functions whose graph consists of
curves with infinite branches , that is to say where the power series
obtained do not converge to the function but diverge. Only in the late
1800s with Poincar's 'qualitative theory of non-linear differential
equations' ,59 was it possible to represent such con1plex functions in
a plane. Poincar extended the geometry of curves to a study of surfaces and their functions in an effort to understand the behaviour of
dynamic systems (such as the planetary system). Duffy explains at
Poincar discovered a new kind of singularity , so-called essential
singularities' , where the values of the function close to the singularity
luctuate through a range of different possibilities without stabilizing'.60 According to Poincar , we need to distinguish four types of
singularities situated in a vector field: saddle points nodes foci and
centres.
The first type of singularity is the saddle point or dip (col) , through
which only two solution curves pass acting as asymptotes for neighbouring curves. A saddle point is neither a maximum nor mll11mUm, since
the value of the function either increases or decreases depending on the
direction of movement away from it. The second kind of singularity is
the node (nud) which is a point through which an nfinite number of
188
ldeas as Problems
curves pass. he third type of singularity is the point of focus (~yer)
around which the solution curves turn and towards which they approach
in the same way as logarithmic spirals. And the fourth , called a centre , is
a point around which the curves are closed , concentric with one another
and the centre. 61
These singularities determine the topological behaviour of the
solution curves , that is their local trajectories in the immediate
neighbourhood. In the language of physics , the singularities function as attractors' determining the trajectories of the curves that fall
within their sphere of influence' .62 Smith describes the application
of Poincar's matherrlatical theory to a dynamic system in weather
forecasting.
Non linear
equations can thus be used to model objectively problematic

(or indeterminate) physical systems , such as the weather (Lorenz): e
equations can define the virtual attractors' of the system (the intrinsic
singularities toward which the trajectories will tend in the long term) ,
but they cannot say in advance which trajectory will be actualised (the
equation cannot be solved) , making accurate prediction impossible.
A problem , in other words , has an objectively determined structure
(virtuality) , apart from its solutions (actuality).63
On the basis of these (matherrlatical and physical) examples we can
clearly discern a qualitative difference between the (virtual) problemelement and the (actua l) solutiorelement. Singularities have to be
understood as the problematising function that deterrnines a fundamental system of solutions , whereby not all of the solutions will
become actualised.
We have come to the point where we can cast a first glance at
Deleuze's forrrlal characterisation of problelns and the way they
generate cases of solutions. Deleuze defines Ideas as virtual systems
of di erential relations between reciprocally determined genetic
elements , internally structured by the distribution of singularities
and ordinary points that determine the conditions of the problem. 64
However , before we proceed further with Deleuze we need to introduce another very lnportant notion which Deleuze uses in order to
characterise Ideas: the notion of 'multiplicity'. As Deleuze says , 'an
Idea is a complex theme' , an internallnultiplicity - in other words ,
a system of rnultiple , non-localisable connections between differential elements which is incarnated in real relations and actual tenns'
(DR 183/237). The notion of multiplicity is deeply influenced by
189
Riemann's notion of nultiplicitor rnanifold' (Mannigfaltigkeit)

(though Deleuze equally draws upon Bergson's theories of multi
plicities 65 ). Let us therefore turn to a very brief sketch of Riemann's
mathematical achievement. Bernhard Riemann (1826-66) revolu
tionised rnodern mathernatics in alrrlost every 111ain field: algebra ,
analysis geometry and topology and his work provided the foundations for the new disciplines of the differential geometry of surfaces
and topology.66 Central to the discussion here are Riemann's radical
ideas concerning spatiality' and multiplicity' and in particular the
distinction between discrete and continuous multiplicities that he
introduced in his inaugural lecture of 1854 'Ober die Hypothesen ,
welche der Geometrie zu Grunde liegen'. 67 While Riemann considers
discrete multiplicities as fonned by isolated rather than continuously
connected elements where the relations subsisting between them are
imposed from outside , continuous multiplicities can be intrinsically
defined by the relations that exist between the open sets' of points
0 1' open spaces'. The concept of continuous multiplicity allows
Riemann to imagine the nature of spatiality as a conglo111erate of
local spaces and networks of relationships an10ng thelTI in a way that
is not biased by any extrinsic determination. Thus , for instance , a
surface can be conceived of as a space in itself without imposing on it
the unity of a three-din1ensional Euclidean space , in which it would
be submerged. In the absence of any external reference to a universal
arrlbient space with fixed dimensions , the multiplicities can form
'n-dimensional spaces being the variable nun1ber of coordinates
assigning values to each point in such a multiplicity. In defining
multiplicities by way of their internal properties , Riemann is able to
wo1' k with spaces of any dimension , even with infinite-din1ensional
spaces. This does not mean that the Euclidean space is c0111pletely
annihilated in the Riemannian type of differential geometry. ln fact ,
a given local space (that is the accumulation of points in a neighbourhood) can be covered by a Euclidean or Cartesian coordinate-111ap.
In Lautman's words:
Each neighborhood is therefore like a small bit of Euclidean space , but

the connection from one neighborhood to the next neighborhood is not
de ned and can be done in an infinity of ways. The most general Riemann
space s thus presented as an amorphous collection of juxtaposed pieces
that aren't attached to one anothe r. 6g
To which Deleuze and Guattari respond in A Thousand Plateaus
y saying: 'if we fo l1 ow Lalltman's fine descriptiol Riemannian
190
Ideas as Problems
space is pure patchwork' (ATP 485/606). Riemannian space is the
precursor for Deleuze's and Guattari's concept of smooth space' (in
contrast to striated space') which appears in A Thousand Plateaus.
However , Riemann already plays an irrlportant role in Difference
and Repetition to the extent that he provides one of the main sources
for the formal characterisation of Ideas as multiplicities of differen
tial relations and corresponding singularities. Ideas are structural
multiplicities defined by their internal properties and the relations
that subsist among them. They defy the use of any external categories , such as the in1position of the concepts of the understanding of
the one and the many:
Multiplicity must not designate a combination of the many and the one ,
but rather an organisation be1 0nging to the many as such , which has no
need whatsoever of unity in order to form a system. The one and the
many are concepts of the understanding which make up the over1 y loose
mesh of a distorted dialectic which proceeds by opposition. The biggest
fish pass through. [...] Instead of the enormous opposition between
the one and the man) there is only the variety of multiplicity - in other
words , difference. (DR 182/236)
Deleuze defines the Idea as an 'n-dimensional , continuous , defined
multiplicity' (DR 182/236). As an example he mentions the Idea
of colour , which for hin1 manifests itself in white light (insofar
as all the wavelengths of colours are seen together).69 The Idea of
colour is a t-din1ensional multiplicity. In fact , we can discern
three dimensions or variables with respect to colour: narnely hue ,
lightness and saturation. These d iInensions are like singular points ,
which in the processes of actualisation generate as solutions' the
diversity of particular colours with a defned degree of hue , lightness and saturation elnbodied in sensuous objects. By continuity
Deleuze n1ay refer to the continuous variation of the dimensions ,
that is their increase or decrease in power (the intensity of a spectral
colour) , or the continuity in the colour spectrun1 itself (wavelength
or frequency).
Having now outlined Deleuze's main mathelnatical sources froln
which he draws his definition of problen1atic Ideas as multiplici
ties or complexes of relations and corresponding singularities' (DR
163/212) that generate cases of solutions (as in the case of solution
curves that fluctuate around a singular point without stabilising) , we
can proceed with describing further formal characteristics and the
functioning of Deleuze's calculus of problemc.:ic Ideas.
191

THE FORMAL CHARACTERISTICS OF IDEAs-PRoBLEMs
In Chapter Four of Difference and Repetition , Deleuze sun1marises
the chief fonnal characteristics of Ideas-problems as follows:
Ideas always have an element of quantitability , qualitability and potentiality; there are always processes of determinability , of reciprocal determination and complete determination; always distributions of distinctive
and ordinary points; always adjunct fields which form the synthetic
progression of a sufcient reason. There is no metaphor here , except the
metaphor consubstantial with the notion of Ideas , that of the dialectical
transport or 'diaphor . (DR 18 1/235)
As can be seen from this quotation , Deleuze primarily employs ITlathematical terms to characterise Ideas-problems and he insists that he
is not using these terrns n1etaphorically. What occurs , according to
Deleuze , is a iiaphora' that is a dialectical transport of notions from
their original mathematical domain to a new meta-mathen1atical ,
philosophical context in which they serve a new function. This
new function , however , is not con1pletely unrelated to their original
use in the mathen1atical field. For this reason , we wiU try to clarify
Deleuze's usage of terms by linking it to their mathematical sources.
In the previous section we have learned that , for Deleuze , Ideas are
multiplicities composed of differential genetic elements (singularities and ordinary points). He now describes the internal topological
processes that take place within the Idea: processes of detern1inability , reciprocal determination and cOlnplete deterrnination. he ele
lnents of quantitability , qualitability and potentiality designate pure
potentials of the Idea to engage in these processes according to the
principles of detenninability , reciprocal detern1ination and complete
determination. Let us look at each of these three elements (or pure
potentials) and corresponding principles that govern the internal
topology of problematic Ideas:
1.he
first thing to point out is that the differental genetic elements (dx , dy) of the Idea are wholly undetern1ined. They have
no prior identity , no defining property , no sensible fonn , that is
no fixed quantity or particular value. Mathematically , a differential dx is an in nitesirnal variation of SOIIle variable x. Now ,
Deleuze says , the di erential can be called 'th
192
Ids
as Prohlems
salient trait that 1I1akes the reciprocal determination possible in

the first place is the determinability of differentials. Although the
differential elements dx and dy are undetermined , they are perfectly determinable in relation to one anothe r. Thus the principle
that corresponds to them is the principle of determinability (DR
172/223)
2. It is important to emphasise that the differential genetic elements

do not exist independently fron1 one another , but always in a
reciprocally determined relation (dy/dx). This means that the
differential relation does not externally relate determined quantities. Rather its relata are qualities' which are defined as intensive
differences (where a difference intrinsically relates to a difference). Consequently , there is a difference of kind (an ontological
difference , in fact) between the differential relation constituting a
differential equation and the so-called prin1i 1' ive function. While
1' he differential equation deals with qualities' or intensive differences' (lifferentials) 1' he prin1itive func t on is concerned with
determined quantities given in intuition. This is why Deleuze calls
the determinable form of the differential relation the pure elernent
of qualitability' (DR 173/224). But it also has another reason:
every differential equa 1' ion can itself be differentiated; thus a firstorder differential gives rise to a second-order differential , and
this again to a third-order differential , etc. The differentiation is
taken 1'0 a higher degree insofar as it expresses no longer the differentiation of a 'supposedly constant relation ("variability") but ,
on the con1'rary [. . .] a degree of variation of the relation i1'self
("variety (DR 173/224). For Deleuze this testifies to the power
of Ideas to give rise to Ideas of Ideas - what Deleuze also refers
1' 0 as different orders of Ideas' (differen 1' degrees of differential
relations) - such that an internal variety or lTIul 1'iplicity of Ideas is
created. Thus the pure element of quali 1' ability pertains to the differen 1'ial genetic elements insofar as they are related 1'0 one another
and even give rise to higher orders of differential relations.
3. As we have seen in our previous discussion of the irnportance
of singularities , the existence and distribution of singular points
provide the conditions of the problem and account for its C0111plete determination (DR 177/230). Tl
193
differential relation at a singular point , resulting in a series of functions with increased power. More precisely , while the operation of
taking the first derivative amounts to a depotentialisation' (DR
174/226) (for instance , if y = x 2 + y3 , we differen t ate by writing
dy = 2xdx + 3y 2 dy) , the derived function is increasingly potentialised by the operation of repeated differentiation. By summing these
higher. order derivatives in the order of their increasing power,
that is in the order of their increasing degree (the degree being the
highest exponent, or power , of the function) , we create a power
series expansion of the form a a 1 x x 2 a3x3+... aFI1.It
shows that as the power series successively expands , it undergoes
a corresponding increase in degree , or powe r. Since an expanding
function also increasingly converges with an analytic function ,
the mathematical power becomes an expression of the increasing potential , or capacity for convergence?O hus the gradual
deterrnination of the differential relations (the terms of the power
series) increasingly brings us closer to the 'solution' , the expanded
function. 'In this sense , the differential is indeed pure power , j ust
as the differential relation is a pure element of potentiali (DR
175/227).
By means of this formal characterisation of the processes and principles that govern the problematic Idea , Deleuze shows that the Idea
is only subject to internal constraints and no longer dependent on
extrinsic factors (as Kantian Ideas are). Due to their internal topological structure , Ideas-problems thernselves provide the sufficient reason
that allows for the genesis of cases of solutions. This whole process of
production or genesis Deleuze designates with the complex notion of
different/c iation (DR 209/270) .71 'Differentiation' (written with a )
characterises the internal topological processes generating the proper
space of the problem. In the words of Salanskis:
Deleuze sees differentiation mathematically as the process of the se
determination of the problem at its specific level , a determination which
places some constraints on the actual values , places shapes and properties
on what counts as a solution to the problem. 72
'Differenciation' (written with a ') characterises processes of the
concrete formation of individual solutions within a symbolic field
relative to the Idea-problern. Mathematically speaking , differenciation refers to the integral curves constituted in the neighbourhood of
singularities.
194
Ideas as Problems
In this regard , four terms are synonymous: actualise , differenciate , integrate and solve. For the nature of the virtual is such that , for it , to be actualised is to be differenciated. Each differenciation is a local integration or
a local solution which thennnects with others in the overall solution or
the global integration. (DR 210-1 1/272)
What Deleuze describes here is a process of genesis which goes from
the virtual Idea-problem with its multiple , non-localisable connections of differential elements to its actualisation in real relations and
actual terms. Deleuze calls the type of genesis from the virtual to the
actual a static genesis' (DR 183/238) , which he understands as a
correlate of the notion of passive synthesis\73 In other words , the
static genesis is not actively performed by some kind of agency at a
particular time. Instead , we have to conceive of it as the instantiation
of a structure that takes place in rhythms and with different speeds
(accelerations and interruptions) determined by the respective symbolic field of actualisation. One exarnple can be found in genetics
that has as its object the genetic structure of organisms , that is the
existence and distribution of chromosomes and their function (cf.
DR 185/240). Chromosomes , the carriers of genes , appear as a di
ferential structure or multiplicity , the elements of which are reciprocally deterrnined and non-localisable in terms of absolute space-time
coordinates. Genetics investigates the potentiality of the genetic
structure to control cell construction and function , and to account
for repetition and difference in phylogeny , both on a reproductive
and evolutionary scale. New research (arising in the second half of
the twentieth century) suggests that the genetic structure is not to be
understood as an irnnlutable dictatoria l' systern. Rather it can be
actualised in multiple ways , at different speeds and different times in
response to environmental stimuli. (l n the language of biology , the
thesis is that the so-called developmental plasticity and phenotypic
adaptivity of an organism to its environment bring out previously
unexpressed potentials of hereditary DNA.f4 The example of genetics shows the conlplex interactions between the virtual and the
actual , the realnl of Ideas-problems and the field of solutions. There
is a specific time (differential speeds and rhythms) invo
195
While it rnight seem at first view that the virtual deterrrl nes the
actual in a one-sided process , this view would arnount to a reductionist 'Platonic' version of the relationship between the virtual and
the actua l. But we need to introduce tirne and recognise its true
meaning of creative actualisation' (DR 216/278). Deleuze says that
rhythlns or different times of actualisation measure the passage from
virtual to actua l. There are thus temporal processes along the lines
of differenciation which determine differential relations to become
actualised (DR 246/317). he different t lnes of differenciation play
an important role in the actualisation of the Idea (DR 217/280).
Deleuze's meta-theory or dialectic of problen1atic Ideas as we
have characterised it so far still remains quite abstract due to its
mathernatical expression. It is therefore appropriate to cite son1e of
Deleuze's own examples at this place. In Difference and Repetition ,
Deleuze elaborates three exarnples in considerable detail: (1) the
exarnple of atomism as a physical Idea (DR 184/238 9); (2) the
organism as a biological Idea (DR 184-5/239-4 0); and (3) social
Ideas in the Marxist sense (DR 186/240-1 ).
In the first example , Deleuze refers to Epicurus' ancient theory
of atomism. Deleuze argues that Epicurus considered aton1s , i.e. the
in1perceptible particles of nature , as forming a differentiallnultiplicity , calling the differential relation between the atorns clinamen.
According to this reading , clinamen does not simply designate an
indeterminate change of movement of atorns testifying to the existence of a free will (as Lucretius would have it). By contrast , in his
letter to Herodotus , Epicurus defines clinamen as the product of
a reciprocal determination between atoms that occurs in a time
smaller than the minimum continuous tin1e thinkable' (quoted in DR
184/238). Deleuze emphasises the use of the tenninology of Eudoxos'
method of exhaustion. However , he admits that this example taken
from the ancient theory of atomism is not cOlnpletely satisfacto
since Epicurus' atoms still retain too much independence and only
engage in external spatio-temporal relations.
he second example bears on the famous controversy between the
French naturalists Etienne Geoffroy Saint- Hilaire and George Cuvier
in the first half of the nineteenth century. In a series of public debates
they discussed the question of whether form or function determine
the phenomena of life. Cuvier defended a unctionalist' approach to
biology , clairrl ng that similarities in the anaton1Y of different animal
species , for instance between particular organs , are the product
196
Ids
as Problems
of a similar function. Curier further held that different organisms

evolved independently from one another , following only their
genetic line of the species. Geoffroy Saint-Hilaire , on the contrary ,
belonged to the formalist' school of thought (or , as one might say ,
a transcendental biology'). He believed that the diversity of anirnal
species could be derived from a single underlying form or structure
which consists of abstract anatomical elements and their ideal relations. His comparative studies on vertebrates seemed to prove that
a common structure of bones in the san1e relative positions could be
found in different vertebrates. hus the hyoid of the cat consists of
nine small bones , while that of man consists of five , the other four
being bent towards the skul l. Geoffroy Saint-Hilaire explained the
modifications between the species by n1eans of processes of bending
and deformation determined by environmental factors. In Deleuze's
terms , the actualisation of the structural multiplicity or Idea proceeds
in accordance with reasons and at speeds determined by the environment , with accelerations and interruptions , but independently of any
transformist passage from one actual tenn to another' (DR 185/23940). Geoffroy Saint-Hilaire's theory of a single archetypal structure
is not as obscure and speculative as it l1ght seeln at first glance , and
such a structuralist approach has enjoyed a revival in modern genetic
biology. As we have already discussed in our exarnple given above ,
genes are defined as a differential multiplicity of chromosomes which
appear as loci, that is within a complex structure of non- localisable
relations of proxnity. This genetic structure can be regarded as
a transcendental or archetypal form , provided that by this we do
not mean a dictatorially determining and active form , but rather a
problematic Idea which actualises itself in response to environmental
conditions.
Apart from these two physical and biological examples , Deleuze
provides a further exarnple from the socio-economic , political
sphere. He treats 'the econon1ic' as a problem or Idea that IS never
given properly speaking , but rather designates a differential virtuality
to be interpreted , always covered over by its forn1s of actualisation'
(DR 186/241). he Idea of the econolnic is defined as a multiplicity of relations of production and relations of property , its abstract
element
197
on the contingent, ernpirical and historical conditions which are

given .75 The important thing about the socio-economic Idea as a
n1ultiplicity is that it involves no pre-existing identities. These are
only the product of a common sense that tends to objectify the virtual
multiplicity: 'It mal s society a thing that can be seen , relnembered
and thought. [. . .] comrnon sense reifies sociability; it lisplaces the
practice , the process of the constitution of social relations , with the
product.'76 he capitalist and wage-labourer of capitalist society are
actualisations of virtual , reciprocally deterrnined relations of production and the determination of contingent circumstances. In other
words , actual class relations , political and j uridical relations are cases
of solutions which answer the socio-economic problematic. Although
we can see that Deleuze here makes extensive use of Marxist theory
and terminology , he also rnodifies Marx or at least Marxism to a
certain extent. First of all , Deleuze does not consider actual econon1ic
relationships as the essence of society as a whole , rather
those actual relationships , and all social relationships are the incarnation
of economic relations as differential virtualities that may be actualised in
different ways. 50 we have something like the priority of the economic
as found in Marx, without the economic essentialism as found in certain
forms of Marxism .77
Furthermore , Deleuze does not regard capitalist society as the
culn1ination of a dialectical development of social formations. The
actualisation of the socio-econornic Idea does not follow any universallaw or dialectic of contradictions. Instead of 1 ing an abstract
transcendent principle , the Idea is not larger than what it deterrnines ,
that is it is itself determined by the contingent en1pirical circurn
stances. herefore the series of social formations as solutions to
the Idea of the economic reselnbles rather a discontinuous line with
cuts and limiting poles. In Anti-Oedius , Deleuze and Guattari will
reconceptualise Marx's universal history so that 'universal history
is the history of contingencies , and not the history of necessity.
Ruptures and limits , and not continuity' (AO 140/163).
RELATION BETWEEN MATHEMATICS AND PHILOSOPHY
Now let us return to sorne of the questions we raised before concern

ing the status of n1athematical theories and notions for philosophical thinking. How does Deleuze conceive of the relation between
mathernatics and philosophy? First we can rule out the thesis that
198
Ids
as Problems
philosophy be understood as a reflection on mathematics. As Deleuze

and Guattari say in What Is Philosophy? , one does not gain anything
for philosophy if one tries to define it in terms of reflection. On the
contrary , the object of philosophy is lost. 78 Besides , mathematicians or scientists have no need for philosophers to reflect on their
theories and never have had such a need. 'e may describe Deleuze's
engagelnent with mathelnatics (and the history of mathematics) as
the creation of an encounter between two disciplines which can generate new n10vements of thought in philosophy , that is give rise to
new philosophical concepts displacing traditional ones. We already
came across some of the new philosophical concepts that are created
through the encounter with mathematics , such as the differential' or
the singular' (as opposed to the ordinary or regular , and no longer to
the universal or general). Now Deleuze mobilises mathematics for the
conception of Ideas-problems fascinated by the way how problems
are expressed relative to the field of solution that they determine (for
example , how singularities of vector fields determine the topological
behaviour of solution curves). Deleuze recognises that the distinction between the problem as a genetic differential structure and its
fundarnental systerrl of solutions has the potential to lead to new conceptualisations of the traditional philosophical concepts of problem
and solution. Is Deleuze then applying mathematics to philosophy?
Deleuze would clearly deny this. Following Lautma Deleuze
says that mathematics , for instance the differential calculus , is just
one syrrlbolic field in which dialectical Ideas are incarnated , and it
rrght not even necessarily represent the most complete form of the
expression of problems and the constitution of their corresponding
solutions (DR 1811235). We have to rernember that the dialectic is
ontologically anterio r. It is the question posed beforehand , although
the Idea of the question does not come to mind until after the study
of the exemplary domain. his means that by virtue of the study of
differential calculus and the problems it poses in mathematics we
can reach the lniversality of the dialectic' , a mathesis universalis of
what it means to think (DR 1811235). In fact , one could say that the
dialectic establishes for its Ideas-problems a differential calculus not
only in mat
Each engendered domain , in which dialectical Ideas of this or that order
are incarnated , possesses its own calculus. [...] lt is not mathematics
which is applied to other domains but the dialectic which establishes
for its problems , by virtue of their order and their conditions , the direct
199
differential calculus corresponding or appropriate to the domain under

consideration. (DR 1811235)
Differential calculus , in the particular sense in which it is deployed
in mathematics , is a specific mathematical tool that entirely belongs
to mathematics , even at the very moment when it finds its sense in
the revelation of a dialectic which points beyond rnathen1atics' (DR
179/232).
Without doubt Deleuze is here rnanoeuvring on dangerous ground
which will not be approved by mathematicians or scientists thinking
in terms of rigorous exact definitions. Deleuze is , however , aware of
the danger:
Of course , we realize the dangers : citing scientific propositions outside
their own sphere. It is the danger of arbitrary metaphor or of forced
application. But perhaps these dangers are averted if we restrict ourselves
to taking from scientific operators a pa 1'ticular conceptualizable cha 1' acter which itself refers to non-scientific areas , and converges with science
without applying it or making it a metapho r. (CIT 1251169)
Deleuze denies at several places that his deployment of rnathernatical
or scientific notions involves the use of metaphor (DR 1811235 and
190/246). If we take this claim at face ralue how then can mathematical or scientic concepts be put to use in philosophy? Deleuze
first restricts this usage to a sub-set of 111athernatical or scientific
concepts , those that are essentially inexact\ hey can be given a
philosophical (or even artistic) conceptual din1ension and n1ade rigorous in a way that is not directly scientific. As Deleuze explains in
an lntervlew:
There are two sorts of scientific concepts. Even though they get mixed up
in particula 1' cases. There a 1'e concepts that are exact in natu 1'e, quantitative , defined by equations , and whose very meaning lies in thei 1' exactness:
a philosopher 0 1' writer can use these only metaphorically , and that's quite
wrong , because they belong to exact science. But there are also essentially
inexact yet completely rigo 1'ous concepts that scientists can't do without,
which belong equally to scientists, philosophers, and artists. They have
to be made rigorous in a way that's not directly scientific , so that when a
scientist manages to do this he becomes a philosophe an artist , too. This
sort of concept's not unspecific because something's missing but because
of its nature and content .7 9
It is not altogether clear whether one should indeed say that scientists
becorne philosophers (or artists) when they re f1 ect on their proper
creations , as is suggested in this quotation. 1n fact , Deleuze renounces
200
1deas as Problen1s
this claim in What ls Philosophy? as a bad joke (WP 6/11 , quoted

in note 78). What is rrlore interesting with respect to this quotation is that Deleuze appeals to the potential of essentially inexact'
matherrlatical or scientific concepts. 80 These inexact concepts seem to
possess an excess of sense , which can be mobilised in contexts that
are distinct from the one in which they are usually (and justifiably)
used. They are essentially mobile' concepts due to their vagueness'
or inexactitude'.81 As such they are capable of generating a movement of thought that transcends their usual sphere of application
and arouses a synthesis with new conceptual components in another
sphere. Why should we Toluntarily block the movement of thought by
clinging to concepts with exact and rigorous definitions within their
traditional context? In his article Bernhard Riemann's Conceptual
Mathen1atics and the Idea of Space' , Plotnitsky argues in favour of
interactions taking place between philosophical and mathematical
thinking , and between both and artistic thinking. The Riemannian
concept of multiplicity is just one example that has proved to be very
inspiring not only for Deleuze but also for Bergson and Husser l. 82
'e do not believe that Deleuze has recourse to mathematical
language because of its being the best tool to express a dogmatic
luetaphysics' (in the Kantian use of the term) as Sa]anskis suggests. 83
Besides , Deleuzian 'metaphysics if we want to use this expression ,
should not be equated with the dogmatic metaphysics that Kant
seeks to abandon , but on the contrary with a type of post-Kantian
transcenden talisI As Deleuze says , the discovery of the problematic' has to be seen in a transcendental horizon , 'as the transcendental
element which belongs "essentially" to beings things and events'
(DR 195/252). Deleuze explicitly refers to the problerrlatic as a
transcenden tal eld and it is this concept of transcendent 'that we
aim to track down in this book. Furthermore , we would argue that
Deleuze's understanding of the relation of philosophy and mathen1atics essentially differs frOI l1 the style of philosophy which Badiou
champions. 84 1hile for adiou the 'little style' of philosophy would
be a philosophy of n1atheluatics belonging to the genre of episten101ogy or philosophy of science , the grand style which he considers
the pred0111inant task of philos
201
atternpts to draw from mathernatics those concepts that give us food

for thought or force us to think. How do we understand the nature
of problerns , their truth and relation to cases of solution? Can we
conceive of a multiplicity that is intrinsically defined and not subject
to external conditioning or totalisation? It is Deleuze's interest in
the irreducibility of problems and their status as problematic Ideas
endowed with a genetic power that motivates his engagement with
mathematics , in particular with the differential calculus and the
Riernannian notion of a structural multiplicity.
Notes
1.
2.
3.
4.
5.
6.
7.
DR 140/182 , 1411184 , 194/25 1.

DR 140/183 , 1411183.
DR 1411183-4 , 194/251.
With regard to Deleuze's discussion of Kant , we will mainly refer to his
book Kant's Critical Philosophy: The Doctrine of the Faculties (1963)
and his seminal essay The Idea of Genesis in Kant's Esthetics' (first
published in 1963). Deleuze's involvement with Lautman's philosophy
is rather hidden , but significant traces can be found in Chapter Four of
Dfference and Repetition.
Cf. CPR A 568/B 596: 1hat is an ideal to us , was to Plato an idea in
the divine understanding , an individual object in that understanding's
pure intuition , the most perfect thing of each species of possible beings
and the original ground of all its copies in appearance.'
See the explanatory note by the editors and translators Paul Guyer and
Allen 1. Wood in CPR p. 746 , note 86.
Plato's claim that Ideas are causes (aitiai) of natural things can be found
in Phaedo 100c 102a Republic 508e 'imaeus 29a 30b.
C f. Allison , 'Is the Critique ofJudgment "Post-Critical"?' , pp. 78-92.1n
this article , Allison argues that Kant's third Critique , which appeals to
the purposiveness of nature understood as a transcendental principle of
a specific form of systematic unity , does not entail the abandonment of
the basic commitments and principles of the first Critique. For Allison ,
the third Critique has to be understood as building upon Kant's initial
critical position. His main argument is that in the Appendix to the
Transcendental Dialectic Kant ascribes not only a regulative function
to reason's Ideas of systematicity and purposiveness , but also already
a transcendental role , that is the role of a condition of the possibility
of coherent experience. In his book on Kant and Deleuze , Christian
Kerslake argues as well that the Appendix to the Transcendental
Dialectic contains a transcendental deduction of Ideas of reason , cf.
Kerslake , Immanence and the Vertigo of Phlosophy , p. 191.
8.
202
Ideas as Problems
9. Allison , Is the Critique of Judgment "Post-Critical p. 8 1. For

Allison , there is no contradiction involved in both the claim that the
Idea of systematicity has a regulative function as a logical principle
and the claim that this logical principle presupposes a transcendental
principle.
10. Deleu H as in "History of Philosophy" in L'Abcdaire.
1 1. Kerslake , Immanence and the Vertigo of Philosophy , p. 19 1.
12. Ibid.
13. Ibid. , p. 192.
14. In this paragraph , CPR A 673/B 701 , Kant explicitly exempts cosmological Ideas from this use , for they produce antithetical world-concepts
(for instance , they equally give rise to both the concept of a world
which is infinite in past time and consists of infinitely many parts and
the concept of a world which has a definite beginning in time and a
finite spatial extension). In comparison , psychological and theological
Ideas do not induce reason to 1' un up against these antinomies and are
therefore suitable as action-guiding rules. However , on another occasion Kant indiscriminately refrs to all psychological , theological and
cosmological Ideas as 'regulative principles for the systematic unity of
the manifold of empi 1' ical cognition in general' (CPR A 6711B 699).
15. Allison explains that Kant is not claiming that systematicity is , of itself,
a sufficient crite 1'ion of empirical truth , as if the systematic embeddedness of an empirical gene 1' alization or "law" in an ove 1'arching theory
0 1' set of laws we 1'e sufficient to account for its truth. he claim is rather
that systematicity is necessary in order to have a sucient criterion of
empirical truth , and the 1' efore , a coherent employment of the understanding , or as the second passage [CPR A 654/B 682] suggests , virtu-'
a11y any valid employment of the understanding at a11' (in Allison , Is
the Critique of Judgment "Post Critical" ? p.82).
16. C f. CJ 5: 180 and 5: 18 1. It can be argued , by the way , that these
passages testify to Maimon's influence on Kant , namely the impact
of laimon's idea of an infinite understanding. Alough as we have
seen , Kant already assumed in the Critique of Pure Reason a being of
reason' as a ground of systematiciq he did not yet conceive of it in
terms of an understanding.
17. Just for this reason we are also j ustified in thinking of the world-cause
in the idea not only according to a subtle anthropomorphism (without
which nothing at
203
20.
21.
prnciple of purposveness. 1n the case of teleologcal judgements , purposveness s taken as objective, material, implying ends , while in the
case of aesthetic j udgements purposiveness is clearly subjective, formal ,
excluding any end (cf. KCP 54/92-3).
See Kant's letter to Reinhold , 28-31 December 1787, in Ka t's
Briefwechsel, pp. 51315. Selections are translated in Zweig (ed.) ,
Kant's Philosophical Correspondence , pp. 127-8.
For more details of the evolution of Kant's third Critique , see Zammito ,
The Genesis of Kant's Critique of}udgment, pp. 5 , 7 and 275 6.
Zammito , The Genesis of Kant's Critique of ]udgment , p. 166.
Zammito assumes that the reformulation of the whole issue in terms
of the faculty of judgement and not of reason is due to Kan t' s high
estimation of reason: 'J udgment as a faculty has none of the prestige
and dignity which Kant cannot help but invest in reason. 1ndeed, he
finds the occasion to 1 the blame for a11 error squarely at the door of
the faculty of judgment' (Zammto , The Genesis of I(ant's Critique of
]udgment , p. 167).
See C] 5: 168. See also Frst 1ntroduction (1790) 20: 201 , 20: 208 and
20: 245 , and Second Introduction (1793) 5: 177 and 5: 197.
In C], ~57 Kant wi11 explicitly claim a necessary relation of aesthetic
j udgements of taste to an indeterminate concept , which s an 1dea of
reason.
In the First Introducton to the Critique of the Power of ]udgment,
Kant ca11s precsely this capacity of representing sublmty in objects a
feeling of spirit [Geistesge;171] \
CJ 5: 270. Kant notes that the most suitable examples for pure aesthetic judgements of the sublime can be found in raw nature'. Products
of art (such as buildings , columns , etc.) cannot adequately illustrate
the sublme , since there is always a human end attached to them that
determines the form as well as the magnitude. Natural things (such as
animal organisms) whose concepts evoke a determinate end are equally
improper examples; see C] 5: 252-3.
CJ 5: 265: He [the unrefined person] wi I1 see in the proofs of the
dominion of nature given by its destructiveness and in the enormous
measure of its power , against which his own vanshes away to nothing ,
only the distress , danger , and need that would surround the person
who was banshed thereto.'
Kant's second example that uses the analogy of the relation between
soul and body is not part
22.
23.
24.
25.
26.
27.
28.
29.
204
Ideas as Problems
3 1.
32.
33.
34.
35.
36.
which perhaps no intuition can ever directly correspond" [C] 5: 353] is

in effect a most complex and obscure procedure , a truly creative , inventive and autonomous act of the imagination' (Schaub , Gilles Deleuze
im Wunderland , p. 74; my translation , D. V.).
We do not wish to say that geometrical construction is not a very inventive activity , but it surely involves a different kind of creativity than
the one in symbolisation. Furthermore , to Kant's eyes , the construction rule of a geometrical figure , say a triangle , already attaches to the
concept, albeit obscurely and in a hidden way , and therefore simply'
requires an a priori presentation in (pure or empirical) intuition. See
CPR A 713/B 741 and B 17.
On behalf of his own philosophicaI concept of Idea , Deleuze will retain
the relation of Ideas and events , but he will define events in terms of the
virtual (not spiritual or mental).
Smith , Deleuze , Kant , and the Theory of Immanent Ideas' , p. 50.
Smith , Deleuze , Hegel , and the Post-Kantian Tradition' , pp. 126-38.
Lautman , New Research' , p. 199. Lautman's collected work was rst
published in 1977 under the title Essai sur !'unit des mathmatiques
et divers crits by Union Gnrale dditions. This rst collection of his
work was reissued in 2006 by Vrin under the title Les Mathmatiques ,
les ides, et le rel physique. In quoting texts from Lautman , we will
use Simon Duffy's recently published English translation which relies
on the V rin edition.
Alunni notes that Lautman will rather situate himself on the side of a
radical questioning of this "traditional metaphysics" which was initiated by Heidegger through what he call "the going beyond" and
Destruion. Such an alignment [. ..] should radically disqualify his
all-too famous Platonism or at least in the sense that it is traditionally ,
and offhandedly , attributed to him' (Alunni , Continental Genealogies' ,
p.73) .
37.
Lal
ltman
38. hus Lltman
writes in the Introduction to his thesis Essay on the

Notions of Structure and Existence in Mathematics' (1938): The
method that we follow is essentially a method of descriptive analysis ,
mathematical theories constitute for us a given within which we try to
identify the ideal reality with which this matter is involved' (p. 92).
39. Lautman , New Research' , p. 204.
40. Ibid. , p. 204.
41. Lautma Essay on the Not
205
42.
43.
44.
45.
46.
47.
48.
49.
50.
5 1.
52.
53.
54.
55.
56.
57.
Lautman , Essay on the Notions of Structure and Existence' , p. 188.

Cf. Lautm
maI
1
Duffy , Albert Lautman' , p. 373 , see also pp. 358-9.

Lautman , 'New Research' , p. 206.
Ibid. , p. 200.
Ibid. , p. 203.
Duffy , Albert Lautman' , p. 367.
Lautman , Essay on the Notions of Structure and Existence' , p. 189.
Lal
1tm
laI
1
Duffy , Albert Lautman' , p. 370: 'There is no ideal reality associated

with ideas in Deleuze but rather ideas are constituted by the purely
problematic relation between conceptual couples.'
Simon Duffy does not point to this difference. Instead he contends that
'For Deleuze , it is the problematic nature of the relations between con
cetual coules that incarnate problematic ideas and which govern the
kinds of solutions that can be offered to them' (Duffy , Albert Lautman' ,
p. 370 , my emphasis , D. V.). Contrary to Duffy , we hold that Deleuze
does not conceive the structure of Ideas in terms of binary conceptual
oppositions , but rather in terms of a differential multiplicity.
Lltman New Research' , p. 203.
Lautman , 'On the Reality Inherent to Mathematical Theories' , p. 28.
For a brief account of Deleuze's engagement with Abel and Galois , see
Salanskis , 'Mathematics , Metaphysics , Philosophy' , pp. 52--3; Smith ,
Axiomatics and Problematics' , pp. 159-61; and Bowden , The Priority
f Events , pp. 113 14.
Lautman describes Weierstrass's local method of constructing the analytic continuity of a function within a whole domain as follows: 'Take
as a new center a point inside the frst circle , both a new series and
a new circle of convergence is thus obtained that can extend beyond
the rst. he new series extends the rst if their values coincide in the
common part of the two circles. The series can thus be extended in a l1
directions up to the points in the immediate neighborhood in which the
series obtained diverge. Thus we see that in this method the domain
is not circumscribed in advance , but rather results from the infnite
succession of local operations' , in Lautman , Essay on the Notions of
Structure and Existence , pp. 96-7.
Kline , Mathematical Thought , p. 732.
Du y The Logic of Expressio l1, p. 82.
Ibid. , p. 82. C f. also Kline , Mathematical Thought , p. 733.
C f. Durie: 'In this conte
58.
59.
60.
6 1.
62.
206
Ideas as Problems
of attraction , then it will inevitably tend towards the attractor' (Durie ,
Problems in the Relation between Maths and Philosophy' , p. 175).
63. Smith , Axiomatics and Problematics' , pp. 161 2.

64. See , for instance , DR 163/212 , 173-4/225 , 1811234 , 244 /3 15.
65. Cf. B 39/3 1-2, ATP 483 -4 /604. 1n fact , Bergson's philosophical ideas
concerning space and time are based on Riemann's geometry and
concept of manifold. Plotnitsky says: Bergson's duration may be seen
[. ..] as an extraction or dist i1l ation of an inexact, qualitative , nonnumerical concept of multiplicity or manifoldness from Riemann's
concept of manifold , a concept that is juxtaposed to "metric manifoldness or the manifoldness of magnitude" , a (mathematical) exact ,
numerical counterpart of it in Riemann's 0rerall conceptual architecture of manifold' (Plotnitsky , 'Manifolds: On the Concept of Space in
Riemann and Deleuze' , p. 191).
66. For more details , see Plotnitsky, Bernhard Riemann's Conceptual
Mathematics and the 1dea of Space' , pp. 105-30.
67. For an English translation , see: Riemann On the Hypotheses which lie
at the Foundation of Geometry' , in Ewald (ed.) , From Kant to Hilbert ,
pp. 652-6 1.
68. Lautman , Essay on the Notions of Structure and Existence' , p. 98.
69. Cf. DR 206/266--7 and 182/236.
70. C f. Duffy's very helpful discussion of this and related issues in The
Logic of Expression , ch. 3 , in particular pp. 77 8.
7 1.he full formula is actua l1 y indi-drama-differenti/ciation' (D R 24 6/317)
which also takes account of spatio-temporal dynamisms involved in the
dramatisation' of 1deas , that is of impersonal' processes of individ
ation. Deleuze argues that there are intensive' processes that function
as movements or conditions of actualisation (c f. Chapter Five of DR).
For more on dramatisation' or intensive individuation' , including a
comprehensive account of Gilbert Simondon's theory of individuation ,
see Bowden , The Priority of Events, in particular sections 3.3 and 3 .4.
72. Salanskis 1athematics Metaphysics , Philosophy' , p. 52.
73. In The Logic of 5ense , Deleuze introduces the notion of a dynamic
genesis' (L5 186/217) in order to account for the counter-process of
a static genesis , namely the movement which goes from the actual
world of physical mixtures and sensations to the virtuality of 1deas. ]oe
Hughes argues that the dynamic genesis
207
74.
75.
76.
77.
78.
allows for no variance in this process. However , it is not clear what he

makes of processes which begin with the problem of a philosophical
concept or a scienti f c formula. Moreover , Hughes interprets the faculty
of thought as the faculty where Ideas are born (ibid. , p. 86). But we
learned from Deleuze and Lautman that Ideas subsist in a virtual realm
which is not immanent to a faculty or to a thinking subjec t. Ideas rather
constitute a plane of immanence themselves and their actualisation is
elicited by internal processes of differentiation. Finally , we very much
doubt Hughes' claim that Dirence and Repetition is comparable with
the Phenomenology of Spirit or the Critique of Pure Reason; he argues
that knowledge , "the supposedly simple result" is the f nal telos of the
entire genesis - of the entire book' (Hughes , Deleuze's 'Dierence and
Repetition' , p. 86). We would argue that instead of there being a telos'
of the genesis of thought, there is rather a suff cient reason' of thought
which forces us to think (we have no choice). The genesis of thought is
a dice-throw and we have no guarantee that the resulting thought is in
any way remarkable , notable or interesting.
We are indebted to John Protevi for his explanations on this issue , given
in a lecture series at the 2 nd Deleuze Camp in CardifflWales in August
2008. See Protevi's series of lecture courses on Deleuze and Biology' ,
a vailab le at <http://www.protevi.com/john/Deleuze_biology_lectures.
pd f> (l ast accessed 22 August 2012).
The emergence of the capitalist society , for instance is conditioned by
the contingent conjunction of two flows , as Deleuze and Guattari will
later explain in Anti~Oedipus: the flow of labourers from country to
town , and the f1 0w of capital which no longer functions only in a local
setting as an equivalent for goods , but on a larger scale as a f1 0w of
f nance and credits (cf. AO 225/266-7).
Read Fetish is Always Actual , Rerolution is Always Virtual p.83.
Choat, Deleuze , Marx and the Politicisation of Philosophy' , p. 19.
C f. WP 6/1 1: [Philosophy] is not re f1 ection , because no one needs
philosophy to re f1 ect on anything. It is thought that philosophy is
being given a great deal by being turned into the art of re f1 ection , but
actually it loses everything. Mathematicians , as mathematicians , have
never waited for philosophers before re f1 ecting on matl
208
Ideas as Problems
8 1. See Patton's excellent article 'Mobile Concepts , Metaphor , and the
Problem of Referentiality in Deleuze and Guattari' , pp. 27-4 6.
82. Plotnitsky also suggests that Jackson Pollock's paintings can be understood as a realisation of Riemannian space in art. Plotnitsky, Bernhard
Riemann's Conceptual Mathematics' , p. 109.
83. C f. Salanskis:Mathematics plays the central part in the new [Deleuzian]
philosophy of nature , which is at the same time and in the same way
philosophy of culture , but it does so in the qualitative , absolute and
immediate manner which characterises for Kant dogmatic metaphysics:
metaphysics which claims to grasp being and becoming as such with
purely conceptual tools. he only difference , but it is an important or
is that the conceptual key is mathematical , and not logical' (Salanskis ,
Mathematics , Metaphysics , Philosophy' , p. 54; see also p. 51).
84. Badiou , 'Mathematics and Philosophy pp.12-30.
209
lme a d
the Split
SuJect
The Kantian Revolution in the Philosophy of Time

he
concern of this chapter will be to show that the genesis of thought

presupposes a fracture or rift in the thinking subject. This fracture ,
as we will see , is caused by time. Deleuze takes his inspiration fron1
Kant , that is the Kantian notion of time as a pure and empty form ,
a form of interiority , but he does not follow all the implications that
are drawn by Kant. Quite to the contrary , Deleuze criticises Kant for
atten1pting to reconcile the two unequal halves of the split subjec
the active faculty of synthesis and the passive faculty of receptivity ,
and to mediate them in a new form of identity: 'the fracture is quickly
fiUed by a new form of identity - nan1ely active synthetic identity;
whereas the passive self [moi] is defined only by receptivity and , as
such , endowed with no power of synthesis' (DR 87 /1 17).1
Deleuze shifts the attention from the transcendental subject to the
transcendental and genetic conditions. He wants to give a genetic
account of the act of thinking within thought and of the subject itself.
Deleuze assigns to time a very special role with regard to the constitution of the subject of thought and creation. herefore it will be
necessary to examine Deleuze's three passive syntheses of time. The
first synthesis of time is constitutive of the subject , the second synthesis is a constitutive condition for acting and thinking. However , the
third synthesis of time is the most important one: Deleuze equates it
with the Nietzschean eternal return , which enacts a power of repetition. Through the repetition of the eternal return the identity of the
tinking subject is dissolved and turned into a series of little selves or
simulacra. Thought occurs only at this extreme point of the fractured
1. The transcendental and genetic conditions of thought have thus to
be considered in relation to Deleuze's con1plex theory of tIne.
As the above outline of the problenl of this chapter suggests , the
point of departure for Deleuze's re f1 ections on tirne and the split
subject is Kant's critical philosophy. The Kantian critique of traditional n1etaphysics , l1 particular of the three branches of rational
210
Time and the Split Subject

theology , rational cosrnology and rational psychology , contested the
truth of established speculative Ideas such as the existence of God or
the existence of an imrrlortal sou l. Hitherto God had played the role
of the guarantor God , who constitutes the identity of the subject by
creating man in his own in1age and who judges the subject in accordance with divine laws. With Kant , man has become autonomous , a
judge of his own affairs , guided by the rules of reason. But man has
also become a finite beI dependent on sensible intuition when he
reaches out for truth. he split of the subject into its intelligible and
its sensible natllre yields a crisis for the unity of the subject. Although
Kant was careful to resurrect God and the soul as regulative Ideas
and to restore the superiority of the intelligible over the sensible , there
nonetheless ensued a significant problem for the subject's identity.
he subject of rationalisn1 - whether Leibniz's conception of a
simple substance or Spinoza's conception of a particular mode of
God's attribute of thought - was conceived as an analytic identity.
With Kant , the conception of the subject undergoes a fundamental
change since the subject is no longer defined as an analytic identity
but as a synthetic identity. The Kantian subject is premised on the
synthesis of two opposed faculties: the active faculty of thought and
the merely passive faculty of receptivity. While the faculty of thought
presupposes the transcendental form of the 1 think' and the a priori
categories of the understanding , the faculty of receptivity provides
sensible intuition given in the forms of space and time. In drawing
this finn distinction Kant made the human subject and its claim to
knowledge and trllth dependent on he touchstone of experience'
(CP R A viii). The human subject has becorne a fnite subject by virtue
of its dependence on the receptivity of intuition. This is a point that
Heidegger strongly emphasises in his selninal book Kant and the
Problem of Metaphysics. Deleuze however questions the importance of this change in the philosophical conception of the subject.
lt is sometimes argued that a considerable philosophical change took
place between pre-and post-Kantianism - the former being defined by the

negative of limitation , the latter by the negative of opposition; the one by
analytic identity , the other by synthetic identity; the one from the point
of view of infinite substance , the other from the point of view of the finite
Sel f. [. . .] However, the importance of such changes is open to q uestion.
(DR 58/81)
According to Deleuze , no fundan1ental change has taken place so

long as the identity of the subject is still preserved. The issue of
211
TRANSCENDENALIDEAS
critique is crucial , but Kant's critique of speculative lTletaphysics did

not go far enough. God is retained inasmuch as the place of God is
kept intact. lt is true that man has been put in the place of God , but
the importance of this change is questionable. he reign of the negative (whether limitation or opposition) and identity (whether analytic
or synthetic) continues. 'That is why the Man-God permutations are
so disappointing , and do not advance matters one step' (DR 58/81).
The Kantian subject has conserved its identity , which grounds the
harmonious accord of the faculties. What Deleuze finds revolutionary in Kant is not the new synthetic conception of the subject which
was adopted by post-Kantian philosophy , but a radical rnoment in
his thought that was all but lost to the philosophy that followed.
Rather than being concerned with what happens before and after Kant
(which amounts to the same thing) , we should be concerned with a precise
moment within Kantianism , a furtive and explosive moment which is not
even continued by Kant , much less by post-Kantianism except, perhaps ,
byH lderlin.
(DR 58/81 2)
Deleuze identifies this great revolutionary mon1ent with Kant's new

conception of time. Time is no longer defined as a cosn1010gical or
psychological time , but as a forn1 of interiority' , a pure and ernpty
fOrITl of tIne.here ensue two in1portant consequences which
Deleuze surnrnarises by means of the somewhat cryptic formulas
'tin1e is out of joint' and '1 is an Other'.2 In the following , we will first
look at Deleuze's reading of the Kantian philosophy of time and then
reconstruct his own theory which is laid out in the three syntheses
of time in Chapter Two of Difference and Repetition. In particular ,
we wiU focus on the third synthesis of tirne , the static synthesis
or synthesis of the future , and exan1ine its heterogeneous literary ,
mathematical and philosophical sources. It wiU be seen that the
third synthesis of tin1e which organises time in a series by means of
a constitutive caesura or cut is necessarily linked with a split subject
which , for Deleuze , is the subject of thought and creation.
'THE IME
1s
OUT OF ]OINT'
Deleuze begins his 1978 lecture course on Kant by declaring that

'all of the creations and novelties that Kantianism will bring to
philosophy turn on a certain problen1 of time\Kant introduces an
entirely new conception of time , a conception of which we can say
that its elaboration y Kant will be decisive for all that happened
212

afterwards' (LK 1, p. 1). Deleuze chooses the forrrlU la he time is out
of joint' uttered by Harnlet in Shakespeare's tragedy Hamlet, Prince
of Denmar to summarise what for Deleuze arnounts to Kant's first
great reversal in the theory of time (LK 1, p. 14).3
In antiquity , the theory of time was bound up with the concept
of an enclosed geocentric cosmos , and time was thought of as the
rneasure of the movement of celestial bodies in their orbits. 4 The
ancient model of the cosmos , the armillary sphere , which was the
predominant model until Copernicus , represented a spherical cosmos
with eight rrloving circles and the Ear at the centre. he outermost
sphere was conceived as a moving shell of fixed stars, and it carried
round with it all the contents of the sphere including the Moon ,
the lunar planets (Mercury and Venus) , the sun and the three outer
planets (Mars , Jupiter and Saturn) , though the planets had movements of their own in counter-rotation to the outermost sphere. he
Earth was at rest in the centre , like a hinge around which the celestial
bodies pivot. However , it should be noted that not all theories held
the Earth to be stationary - Plato , for example , claimed that the earth
revolved on its axis.
Both Plato and Aristotle associated time with number , defining
it as 'the number of moven1ent'. In recounting the myth of creation ,
Plato's Timaeus imagines a den1iurge who creates the universe as an
image of the ideal world and its eternal forms:
50 he bethought him to make a moving image of eternity (aion) , and
while he was ordering the universe he made of eternity (aion) that abides
in unity , an eternal (aionios) image moving according to number , even
that which we have named Time (chronos). (Tim. 37d)
Although Aristo t1 e argued against Plato that a temporal beginning
of time is in1possible (De Caelo 1, 279 he also defined time as 'a
nun1ber of change in respect of the before and after' (Physics IV ,
219b1) , or simply a number of movement' (De Caelo 1, 279a15).5
hus time was conceived as the number of movement of the planets ,
which passed time and again through 'cardinal points that is certain
privileged and xed points. By n1eans of their celestial reTolution the
planets rnarked off regular periods of time , the shortest unit of which
was the period of day-and-nigh t. Plato said in the Timaeus: utnow
day and night , being seen of us and n10nths and the revolution of
the years have created number , and they gave us the notion of time'
(Tim. 47a). And Aristotle noted that the uniforrrl circular motion
of the planets was n10st apt to provide a number of n1easuren1ent
213
i.e. tirne , since there is no uniforrn qualitative change or uniforrn

increase in size or uniforrn corning-to-be , but there is uniforrn locornotion' (Physics IV , 223b12).
Since tirne in ancient cosrnology was the rneasurernent of the circular rnovernents of the planets , it followed that tirne was naturally
thought of as a cycle (cf. Physics IV , 223b12). Furtherrnore , the definition of tirne could not be separated frOlTI the rnovement of physical
bodies. As Aristotle put it: irne is a number of moven1ent but
there is no movement without physical body' (De Caelo 1, 279a15).
This is why Deleuze attributes to the ancient Greeks a concept of time
which s a mode and not a being. No more than number is a being ,
it's a mode in relation to what it quanti es in the same way tirne is
a mode in relation to what it rneasures' (LK II , p. 2). According to
Deleuz time loses its rnodal character and ceases to be circular only
subsequent to the establishrnent of modern science in the sixteenth
and seventeenth centuries (in particular , in the scientific cosmology
of Newton). Tith Kant finally , time becomes purely formal; it is a
pure forr it has unrolled into a straight line In Deleuze's words:
'The time is out of joint' , time is no longer coiled up in such a way that
it is subordinated to the measure of something other than itself, such as ,
for example , astronomical movemen t. Time has ceased to be the number
of nature , time has ceased to be the number of periodical movement.
Everything happens as if, having been coiled up so as to measure the
passage of celestial bodies , time unrolls itself like a sort of serpent it
shakes off all subordination to a movement or a nature , it becomes time
in itself for itself, it becomes pure and empty time. It measures nothing
anymore. Time has taken on its own excessiveness. It is out of its joint
which is to say its subordination to nature; it's now nature which will be
subordinated to it. (LK 1, p. 14)
Deleuze invokes the images of a serpent which unrolls itself, or of a
circle that snaps , a spring that uncoils itself to become a pure straight
line (cf. LK II , p. 2). He further adduces Borges' idea of a labyrinth
composed of a single straight line' that is all the more tortuous since
it is indivisible , incessant' (KCP vii; DR 1111147).6 These images
already sllggest that with the modern conception of time something
fundamental has change1: the cos l11010gical harmony between the
world and the heavens , man and the heavenly gods , has s0111ehow
broken down. It is as though Tirnaeus' prediction has become true:
'Be that as it may , Tirne came into being together with the Heaven ,
in order that , as they were brollght into being together , so they may
214

be dissolved together , if ever their dissolution should come to pass'
(Tim. 38b). The time of antiquity has perished together with the gods
and the heavens , and a new time is born. Pure and empty time is now
the true subjectivity. It has become an infinite , straight line , which
will cut right through the consciousness of the subject. Its effect is
tortuous and inhuman in the sense that it carries the subject to the
border of the liveable and destroys the well-constituted identity of
the subject.
'1 Is
AN OTHER'
Deleuze borrows from the French poet Arthur Rimbaud (1854 91)
the formula by which he characterises the second novelty of the
Kantian theory of time: '1 is an Other' (j e est un autre)J By defining time as a form of interiority , Kant introduced a fundamental
split in the subject. he Kantian subject is torn between the form
of spontaneity , that is the 1 think' which accompanies all concept
production and guarantees the unity of synthesis , and the ernpirical
self which experiences the effects of thought rather than initiating
the act of thought itself. According to Deleuze , Rimbaud's formula
is an Other' is apt to express the alienation to which the Kantian
subject succumbs , notwithstanding that Rimbaud's phrase occurs in
a rather different contex t. Rimbaud puts the forn1ula '1 is an Other'
in an Aristotelian context , referring to Aristotle's distinction between
determining forrrl and indeterminate matter. Sufce here the following quotation in which Rin1baud distinguishes between matter and
form: 'Too bad for the wood which finds itself a violin! If the copper
wakes up a bugle , that is not its ult. g By means of the forn1ula '1
is an Other' , Rimbaud expresses the experience of being forrrled by
thought rather than being the originato r. Thought forms me - 1 am
not the master of thought at al l.
With Kant , the concern is no longer of a forn1 that inforn1s matter
but of 'an infinite n10dulation , no longer a mould' (KCP ix). Thought
works within me - 1 am affected by thought that is both mine and
the thought of an Other. The fracture or crack in the '1' is produced
by the pure and en1pty forn1 of time.his n1eans that 1 experience
n1yself, i.e. n1y feelings , thoughts , actions and bodily sensations ,
etc. , always under the condition of tin1e which is the interior forrn
of receptivity. But the synthesis of all these different representations
within the unity of consciousness is performed by the transcendental 1, or the '1 k' as the transcendental form of apperception.
215
Phrased more precisely , the 1 affects itself under the forrn of time.
The remarkable outcon1e of this kind of auto-affection is that the
difference between being and thought , or matter and form , is inte"
orised. Deleuze refers to this establishment of internal difference as
the moment of discovery of the transcendental , the element of the
Copernican Revolution' (DR 861117). A transcendental difference
separates the a priori syntheses of the 1 think' from the empiric
psychological sel f. In other words , the '1 think' is the transcendental
condition of the en1pirical self. In order to better understand the
impact of this event , it is worth going back to the historical context
of the Cartesian Cogito and Kant's criticism of it.
Descartes' n1ethod of doubt , by means of which he searched for an
absolute certainty that contains its ground within itself famously led
hirn to the formula '1 think , therefore 1 am' (cogito ergo sum). In his
Meditations , Descartes first applies his philosophical doubt to everything concerning his ody bodily sensations (such as pain , pleasure ,
hunger , etc.) and sense lnpressions , and then continues to doubt the
products of his inner faculties (such as n1en10ries , imaginings , etc.).
He concludes that he could doubt everything except the fact of his
doubting. Since , according to Descartes , doubt is a mode of thought ,
of intellectio in general , and cannot exist separately fron1 thinking ,
the act of thought manifest in the '1 think' appears to be an absolute
certainty. he key idea is that the act of thought cannot be doubted
the instant that it is performed , and thus it is in1possible for it not to
be true. Leaving the question of the kind of impossibility involved
aside , Descartes continues that in order to think one must be. he act
of thought necessarily implies an indeterminate existence. Descartes'
line of argument thus goes from '1 doubt' to '1 think' to '1 am'. Now
Descartes immediately applies the determination 1 think' to the indeterminate existence '1 ar" which is to say that the '1' is deterrnined as
a thing that thinks. 1 am a thinking thing (res cogitans).
Kant agrees with Descartes that the determination think' in1plies
an indeterminate existence. As he puts it the proposition "1 think
insofar as it says only that 1 exist thinking , is not a merely logical
functior but rather deterrnines the subject (which is then at the san1e
tin1e an object) in regard to existence.' However , 'this cannot take
place without inner sense , whose intuition always lnakes available
the object not as thing in itself but rnerely as appearance' (CPR B
429). he second part of this quotation is decisive. To 1 sure the
indeterminate existence is to be determined by the determination ,
216

but this does not yet give us the form under which this indeterminate existence is determinable. Descartes simply assurnes that rny
indeterrninate existence is determinable as a substance or thing (res
cogitans) , and draws out irnplications concerning the '1' which now
enjoys all the attributes of a substance: subsistence , simplicity and
identity. Kant rejects all these claims as instances of paralogism in his
criticism of rational psychology. As Kant has famously declared with
his notion of the phenomenon , something can only be an object for
us if it conforms to our conditions of knoyledge. These are first and
foremost the a priori forms of intuition and secondly the categories
of the unders t: anding. The categories can only apply to something
that is determinable in space and tIne. In fact , time is the most
universal form under which anything given to us in perception must
be determinable. This implies that whatever appears to us in outer
space' has rst to be given in the successive flux of our inner sense ,
that is under the condition of time.
50 time is an a riori condition of all appearance in general , and indeed
the immediate condition of the inner intuition (of our souls), and thereby
also the mediate condition of outer appearances. [. . .] From the principle
of inner sense 1 can say entirely generally: all appearances in general , i.e. ,
all objects of the senses , are in time , and necessarily stand in relations of
time. (CPR A 34/B 51)
Time as the condition of all appearances in general is also the form
under which 1 appear to myself. 1 appear to myself under the fonn of
the determinable as a phenomenon in tin1e.his is what Kant offers
in reply to Descartes: the detern1ination '1 think' indeed actively
determines my indeterminate existence , but it can only do so under
the fonn of the determinable (see also CPR B 157-8 , footnote 1).
Hence the determined thing is not sin1ply a thinking thing' but a
passive being in time. This means that for a subject to internally
intuit itself, in order to
seek out (apprehend) that which lies in the mind, it must affect the latter,
and it can only produce an intuition of itself in such a way , whose form ,
however, [. . .] determines the way in which the manifold is together in
the mind in the representation of time; there it then intuits itself not as it
would immediately self-actively represent itself, but in accordance with
the way in which it is affected from within, consequen t1 y as it appears to
itself, not as it is. (CPR B 68-9)9
Time is therefore the form under which the subject affects itself.
According to Deleuze , the idea that time is the fonn of auto-affection
217
plays an increasingly important role in Kant's Opus Postumum (LK

1, p. 15). Let us now consider some of the consequences of Kant's
new definition of time. Most importantly , time ceases to be defined
in terms of succession. Kant thereby contradicts Leibniz who de ned
time as the order of succession. 10 According to Kant , succession is
only a mode of tirne. Equally , coexistence , which for Leibniz was
supposed to define space , is just another mode of time. 'ime he
[Kant] tells us , has three modes , duration or permanence , coexistence and succession. But time cannot be defined by any of the three
because you cannot define a thing through its modes' (LK 1, p. 15).11
Instead , 'tirne is the form of interiority. It 's the form under which we
affect ourselves , it's the fOflTl of auto-affection. T lne is the affection
of self by self' (LK 1, p. 15).
he transcendental difference that Kant introduces into the subject
through the form of time as the form of auto affection is correlated
to a certain distribution of the transcendental and the empirical , and
of activity and passivity. Kant resurrects the identity of the subject in
the transcendental unity of apperception' , which although it is not
a substantial being but a universal forn1 , unites all spontaneity l. e.
synthetic activity of thought in itself. rhe empirical self, on the other
hand , is defined as entirely passive , as pure receptivity , deprived of
the power of synthesis. It is at this point that Deleuze parts c Olnpany
with Kant:
It is impossible to maintain the Kantian distriblltion , which amounts to
a supreme effort to save the world of representation: here , synthesis is
understood as active and as giving rise to a new form of identity in the 1,
while passivity is llnderstood as simple receptivity withollt synthesis. he
Kantian initiative can be taken up , and the form of time can support both
the death of God and the fractllred 1, but in the course of a quite different
understanding of the passive self. (DR 87/118)
Contrary to Kant , Deleuze argues that the passive self is constituted by a passive synthesis of tin1e. ln fact , he analyses three
distinct syntheses of time: (1) the passive synthesis of contemplationcontraction; (2) the passive synthesis of (transcendenta l) n1enlory;
and (3) the static synthesis of empty tin1e. The first synthesis cons ttutes tirne as the n lO ment of the lived present , the second constitutes
the cor ition of the pure past while the third constitutes the pure
order of tirne and opens up the possibility of a future. It produces
a fracture or crack in the 1, such that it opens itself up for the
movement of though t.
218

Deleuze's Passive Syntheses of Time: The Lived Present and
the Pure Past
At rst glance , the notion of a passive synthesis' appears selfcontradictory. Isn't synthesis a conscious activity that depends
on representations? Aren't the capacity to represent and selfconsciousness pre-conditions for synthesis?'12 According to Deleuze ,
a synthesis can be carried out cornpletely unnoticed by the conscious
lnind. It takes place subconsciously , which is to say that it does not
require a series of conscious representations but rather presentations'
in the sense of Maimonian differentials. Inasmuch as there is no selfconscious agent involved , Deleuze speaks of a passive synthesis'.
It is important to note that , for Deleuze , the passive syntheses of
t lne do not occur within a subject, rather they are constitutive of
subjectivity itself. his rneans that there are segments of subjectivity
and agency which are engendered by the passive syntheses of time. In
other words , only under the condition of passive syntheses are partial
identities (i.e. the formation of local selves or egos) and activities
(such as the active syntheses of menlory and understanding) possible.
However , Deleuze refuses any ground on which to build a secure
identity of the subject.Selves are larval subjects; the world of passive
syntheses constitutes the system of the self, under conditions yet to
be determined ut it is the system of a dissolved self' (DR 78/107).
Let us now look at the frst synthesis of time: the passive synthesis
of contemplation-contraction , constituting tirne as a lived , or living ,
present.
HE PASSIVE SYNHESIS OF THE LIVED PRESENT
Deleuze bases the frst synthesis of time on David Hume and partly
on the English novelist Samuel Butler (1835-1902) , in particular his
book Life and Habit (1878).
One of the major concerns that Hurne dealt with was the problem
of custom or habit and how we acquire it. Our psychic life rests to a
great etent on habit. Hume's investigations into causal sequences ,
for instance , convinced hn that the idea of a necessary connection
of events is not drawn from the perception of an objective relation in things themselves but is based on a subjective impression
or internal sensation of the mind. As Hume fanlously stated , the
idea of necessity involved in causal relations arises from a feeling of
determination , acquired by habit. 13 How does a habit then become
219
rnanifest? Let us suppose that two events A and B constantly appear

in conjunction in our mind. Although each pair of AB is distinct and
independent frorn the other , our imagination tends to assimilate the
different cases and regard them as repetitions of the same conj unction. The rnore we experience the constant conjunction of AB , the
more the Inagination is inclined to associate to a given event A its
usual companion B. As Hume says: After a repetition of similar
instances , the mind is carried by habit , upon the appearance of one
event , to expect its usual attendant.'14 Naturally , a single case of AB
yields no expectancy. But if we experience the repetition of similar
pairs of events , something in our mind changes. The repetition of AB
produces the propensity to think of B when A appears.
According to Deleuze's interpretation of Hume , this act of association is possible by virtue of a specific contractile power' (DR 70/96)
of irnagination. 15 This means that the imagination is able to retain
a preceding event when another occurs , and , upon the repeated
contraction of similar pairs of events , to ground them in an internal
qualitative impression endowed with a certain weight' (DR 70/97).
In 15 way a habit is acquired.
As Hume makes clear , the acquisition of a habit is not a cognitive
act , but rather belongs to our sensory nature. The imagination contracts past and present experiences and projects them into the future ,
producing an expectation. Deleuze understands this as a synthesis
of time (DR 70/97). The important fact is that the idea of time is
not derived from a succession of instants. On the contrary , it is only
by ITleanS of the contraction of instants that time is constituted and
assembled in the lived , or living , present.
To it [the present] belong both the past and the future: the past in so
far as the preceding instants are retained in the contraction; the future
because its expectation is anticipated in this same contraction. The past
and the future do not designate instants distinct from a supposed present
instant , but rather the dimensions of the present itself in so far as it is a
contraction of instants. (DR 701/97)
For Deleuze , this fusion or contraction of (past , present and
future) instants forms the passive synthesis of the living presen t. lt
can be expressed in an asymmetrical arrow of time , which goes from
the particularity of the past to the generality of the future.
Hume's exampl
220

level of contraction that constitutes not sirnply the habits that we
have , but the habits that we are. In other words , apart frorn perceptual passive syntheses there are also organic passive syntheses
(DR 73/99). 'What organism is not made of elements and cases of
repetition , of contemplated and contracted water , nitrogen , carbon ,
chlorides and sulphates , thereby intertrining all the habits of which
it is composed (DR 75/102). It might be surprising that Deleuze
speaks of 'habits' when he seems to refer to mere biological , chemical syntheses of elernents , which , to be sure , can be described as
contractions. Understood in this way , the quotation is misleading.
Delellze does not regard the organism frorn an objective , scientific
viewpoint of chemical syntheses , bllt rather as a living thing composed of many conten1plating selves or souls' (cf. DR 75/103). It
is at this point that he refers to Samllel Butler. Blltler described the
corn in the fields as composed of thollsands of contemplating sOllls
which contract water and minerals in anticipation of turning it into
wheat. 16 Every contraction is at the same time contemplation , that is
conternplation of past experiences and of fllture expectations. here
is a past inscribed in the organs , mllscles and nerve cells of an organlS 1 as well as a futllre dimension. 'Need is the rnanner in which this
flltllre appears as the organic f 01
1
appears in the form of cellular heredity' (DR 73 /1 00). ThllS Delellze
concllldes:
A soul must be attributed to the heart , to the muscles , nerves and cells ,
but a contemplative soul whose entire function is to contract a habi t. This
is no mystical or barbarous hypothesis. On the contrary , habit here manifests its full generality: it concerns not only the sensory-motor habits that
we have (psychologically) , but also , before these , the primary habits that
we are; the thousands of passive syntheses of which we are organically
composed. (DR 741101)
Deleuze further enriches the concept of contemplation by linking

it to the Greek myth of Actaeon who contemplates the goddess
Diana while she is taking a bath in the woods , and of Narcisslls who
contemplates his own image in the wate r. Actaeon and Narcisslls
are both taking plaSllre in contemplating. Pleasllre , according to
Delellze , is not jllSt a defined and isolated case in Ollr psyche , bllt
a principle that rllles over Ollr psychic and organic life. 'There is a
beatitllde associated with passive synthesis , ar we are all Narcisslls
in virtue of the pleasure (allto-satisfaction) we experience in contemplating even thOllgh we contemplate things quite apart from
221
ourselves' (DR 7411 02). Deleuze subjoins a sort of Freudian interpretation to the syntheses of time. He associates the first synthesis of
tirne with Eros (the narcissistic libido) , the second synthesis of time
with Mnen10syne , and the third synthesis of time with hanatos
or the death instinct (DR 108-15/143 52). However , we will pass
over this psychoanalytic dimension of the syntheses of time , since it
opens up another level of interpretation that can be quite misleading. By interpreting the syntheses of time in Freudian terms , there
is the danger that they are thought of as inner-psychic syntheses of
an already constituted subject. Yet the Deleuzian syntheses of time
surpass the view of a composed and Wrell.-c
form the transcendental , genetic conditions which first of all constitute the local selves or conten1plating souls that are connected within
the system of a dissolved self. It is important to keep in mind that
Deleuze is not interested in a psychology of the subject but in n1etapsychological , transcendental syntheses of time.
That has been shown is that Deleuze substitutes for the Kantian
passive self a multiplicity of little selves or contemplative souls , each
endowed with a present of a particular duration. he living p 1'esent
of an o 1'ganism is thus an assemblage of va 1' ious p 1'esents 0 1' 1' hythms ,
constituted by the cont1'actions and fatigues of the conternplating
souls. Thus the passive self is no longe 1' defined in te 1'n1S of passivity
0 1' the fo 1' m of 1'eceptivity. It is no longer the receptacle of sensations al 1'eady forn1ed acco 1' ding to space and time and on which the
a p 1'iori catego 1'ies a 1'e supe 1'lnposed. On the cont1' a 1' Y, expe 1'ience
is a product of thousands of passive syntheses perfo 1' med by the
contemplative souls.
Deleuze's notion of passive syntheses of contemplationcont1' action has an impo 1'tant consequence. The conditions of
experience cease to be abstract conditions of possibility in o 1' der
to become genetic conditions of real experience. By means of
the elabo 1'ation of genetic conditions of real experience , Deleuze
achieves a reunification of the aesthetic that Kant had left divided
into the aesthetic of sensibility and the aesthetic of the beautifu l.
Deleuze c1' iticises Kant for dividing aesthetics
into two irreducible domains: that of the theory of the sensible which
captures only the rea l' s conformity with possible experience; and tha
222

The two domains of sensible experience and artistic creation become
one , since both involve the real of experience (the rhythms of sensible
differentials or intensities , the dynarnic forces of contraction and
expansion) as their genetic condition. 17
Thus Deleuze's first synthesis of time , the passive synthesis of contemplation-contraction , already constitutes a kind of Transcendental
Aesthetic' (DR 981130) that goes beyond the Kantian Transcendental
Aesthetic. Although this synthesis of time resolves sorne of the problerns bequeathed by Kant , it also raises new questions and problems.
We have seen that the passive syntheses of contemplation-contraction
constitute time as lived , or living , present. In Deleuze's words , the
first synthesis , that of habit , is truly the foundation (fondation) of
time' (DR 79/108) , that is of the living present. However , the living
present is not to be understood in terms 'of a perpetual present, a
present which is coextensive with time' (DR 7611 05). Quite to the
contrary , the living present passes. Te first synthesis of time is necessarily 'intraternporal' (DR 76/105). This is why Deleuze sees himself
compelled to assume a second synthesis of tirne , one in which the st
synthesis of tirne operates. This second synthesis of time is not to be
understood as the foundation (fondation) of time but rather as its
ground (f ondenzent). The ground is that which is presupposed by the
passing present; it is that which causes the present to pass. According
to Deleuze , it is the (transcendental) memory that grounds time (cf.
DR 7911 08).
HE PASSIVE SYNTHESIS OF THE PURE PAST
With regard to the second synthesis of time , Deleuze draws on

Henri Bergson's notion of a pure past and mobilises examples from
Marcel Proust's /n Search of Lost Time. Both Bergson and Proust ,
although they have quite different concepts of time , adrnit the idea
of a pure past , or in Proust's words ,a bit of time in the pure state'
(un peu de temps l'tat pur).18 This pure past (souvenir pur) is not
to be confused with a recollection-image (i1nage-souvenir) , that is the
psychological or nlental irnage of a former present. The pure past is
tIne in-itself. That is nlost difficult to understand about Bergson's
virtual cone of memory or Proust's essences is that they designate a
pure time in-itself that exists outside consciousness. 19
However , Deleuze is careful to point to a major difference between
Bergson and Proust (B 122/55 , footnote 1611): Proust believes that
tinle in-itself can be lived , that it can traverse us in terms of an
223
involuntary rnemory and be expressed in a work of art. Bergson ,

on the contrary , treats the pure past as something that cannot be
lived directly. It can only be experienced if it materialises itself in
a recollection-image , that is as a nascent sensation (sensati nais
sante) in the present. But then it is no longer pure past. he pure past
coexists with the present as the hidden ground or the condition of
the passage of time. Bergson's pure past is a transcendental memory ,
an 'imn1emorial or ontological Memory' (B 57/52). In great admiration for Bergson , Deleuze declares in Diflerence and Repetition: If
Matter and Memory is a great book , it is perhaps because Bergson
profoundly explored the domain of this transcendental synthesis of a
pure past and discovered all its constitutive paradoxes' (DR 8 1/110).
Yet, by identifying the second passive synthesis of tin1e with the
synthesis of a Bergsonian pure past or of memory , Deleuze risks
falling into a speculative philosophy of a Platonic type. He seen1S to
forsake critical philosophy for an analysis of a pure , a priori being of
time The problem we will have to cope with is how Deleuze can
maintain a transcendental' philosophy that does not fall back into
a transcendent' or pre-critical metaphysics of things in thernselves.
This can only be achieved by emphasising the role of Proust and the
meaning of a 'transcendenta l' aesthetic of signs or Ideas.
Deleuze introduces the passive synthesis of the pure past on account
of the 'paradox of the present: to constitute t n1e while passing in the
tirne constituted. We cannot avoid the necessary conclusion - that
there must be another tirne in which the frst synthesis of time can
occur. his refers us to the second synthesis' (DR 7911 08). The second
synthesis constitutes the pure past , not as a dimension of the present
but as a past in general' , 'of which the present and the future are
only dimensions' (DR 82/111). This reasoning appears very biased
by spatial metaphors , for it seems to suggest that the past in general is
a kind of container in which the present passes. To be sure , Bergson
would have opposed such an Inpression. He declares that the categories of containing and being contained only apply to bodies , which
are 1T1O mentarily perceived in space. 21 We rnust not conceive of the
existence of the pure past in this way. On the other hand , we cannot
simply define the past in tenns of a n10dification of the presen as
what the present becomes. AcC:c
ording
to our imrlTlediate intuition
1
we aSSLlln1e
t ha
t in be
co
rr
I
rneta
morphosis rnaybe a decrease in intensity. It SeeIT1S natural to
suppose that the past is constituted after haring been present , and
?
224
Tinle and the Split Subject

that the arrival of a new present , i.e. a new sensation , changes the
former present into a (l ess intensive) past. However , this way of
talking about time is no less problematic. At what 1110ITlent does a
present become past? How can we distinguish one frorn the other?
If there is only a difference of degree between both , say , a difference
in intensity , a distinction is ultin1ately not possible. For instance ,
Bergson asks: how could we possibly have the melTIOry of a strong,
intense pain? Under the assumption of a mere difference of degree
between perception and memory , the mernory can by definition only
be actualised in a weak sensation. Consequent it will be impossible
for us to distinguish it om a weak sensation that we experience , or a
weak sensation that we imagine. This weak sensation can never give
us the lTIemory of a strong pain. 22 According to Bergson , there must
be a qualitative difference , a difference of kind between present and
past , or , as he says , between perception and lTIemory.23 he present
is what interests us and which extends itself into sensory-motor
action , while the past is essentially impotent (i mpuissant) and of no
interest for our body (unless it becomes actualised in a sensation).24
he difference between present and past is marked by the criterion of
usefulness for our actual body.
Bergson revolutionises our understanding of time , in particular
the passing of time. According to Bergson , present and past are not
successive instants. One cannot pass frOlTI a present to a past. A
present would always be present and there would be no way for
a new present to arrive. Bergson argues that past and present exist
sin1ultaneously. This implies that whatever happens , happens at least
twice: in t
mOlTI
lent that has already been lived. This leads us to ergson's first
paradox of time: the contemporaneity of the past with the present
that it was' (DR 81/11 1) .25
The past and the present do not denote two successive moments , but two
elements which coexist: One is the present , which does not cease to pass ,
and the other 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 general': The past does not follow the present , but on the contrary , is
presupposed by it as the pure condition without which it would not pass.
(B 59/54)
For Bergson , the de ning characteristic of time is not the succession of instants , but the coexistence of different sheets of time.his
means that duration 0 1' lived ti lTle is indeed a succession but it is
225
so only beca l1 se of its virtual coexistence' with the totality of the

past (B 60/56). This implies that each present is not only always
contemporaneo l1 s with its correlative past , it also coexists with the
whole of the past in genera l.he p l1 re past is like a gigantic cone
constit l1 ted of distinct levels or circles of mernory. The larger these
circles are , the lesser is their degree of contraction. This means that
the base AB of the cone contains an indefinite n l1 mber of memories
in all their various details , while at the level of A'B' the memories
are comparatively more contracted , that is , differences becorne
assimilated. The level of contraction and assimilation of differences
increases until the present 5, which we incarnate with our bodies. 5 is
the most contracted degree of the past and the greatest simplification
of O l1 r mentallife , in other words , the prolongation in sensory-motor
reactions. 26 he present 5 coexists with the past AB , but also with
A'B' , and A" and so on to infinity. Thus , whatever happens at the
present 5 is in effect, repeated at all the levels of the pure past sirnultaneo l1 s1y , in greater or lesser detail depending on the respective
distance to the base or the peak of the gigantic cone of melnory. 'The
same psychological life , therefore , must be supposed to be repeated
an endless number of times on the different storeys of memory , and
the same act of the mind may be performed at varying heights.'27
Bergson's metaphor of the cone illustrates his second paradox of
time: the paradox of coexistence of the entire past with the present
(cf. DR 8 1/11 1).
The third paradox is the pre-existence of the pure past. he pure
past has never been present. It never passes or comes 'after'. It is
preserved in itself. The rnode of being of the pure past is not one
of actual existence , but rather of 'insistence': it does not exist , but
it insists , it consists , it is' (DR 82/111). Deleuze opposes the terms
'present' ,actual' and existence' to the terms pure past' , 'virtual'
and insistence\
The question that now arises is: can we somehow penetrate this
virtual being of the pure past? Is our active synthesis of mel110ry
capable of recovering an instant of the pure past? Deleuze warns us
against an interpretation of Bergson that uses a psychological n10del
of recollection.
One must avoid an overly psychological interpretation of the text. Bergson
does indeed speak of a psychological act; but if this act s 'sui generis ,' this
is because it has made a genuine leap. 1 e place ourselves at once in the
past; we leap into thepast as into a proper element. (B 56/51
226

We have to jl1 mp into a partic111ar circle of the p l1 re past. As Dele l1 ze
states: Mernory is not in l1 S; it is we who move in a Being-memory ,
a world-lllemory' (CIT 95 /1 29-30). At this point, Dele l1 ze brings
in Pro l1 st and the Pro l1 stian examples of invol l1 ntary memory in In
Search of Lost Time. It is not by means of an active synthesis associated with rol l1 ntary mernory that the hero of the novel recovers his
past childhood , his hornetown Combray , his dead grandmother , etc.
'It is within Forgetting , as tho l1 gh immemorial , that Combray reappears in the form of a past which was never present: the in-itself of
Cornbray' (DR 85/115). A n l1 mber of threads come together here.
Dele l1 ze paradoxically connects the notion of forgetting with the
Platonic term reminiscence' in order to describe this s l1 dden j l1 rnp
into the p l1 re past , the invol l1 ntary memory of Pro l1 st. In Pro l1 st, it is
not the effort of recollection b l1 t a sens l1 al trigger (s l1 ch as the taste of
the madeleine , the st l1 mbling over a cobblestone , etc.) that s l1 ddenly
brings back the menlory of Combray. It is tl1 S in a state of forgetf111ness , when the cognitive effort of active conscio l1 sness ceases ,
that reminiscence OCC l1 rs. Rerniniscence recovers the p l1 re essence of
Combray , 'not as it was or as it c0111d be , b l1 t in a splendo l1 r which
was never lived' (DR 85 /1 15). Reminiscence is therefore very different from the active synthesis of memory which is sinlply a tool of representation and reprod l1 ces a fornler present. According to Dele l1 ze ,
the fragment of the p l1 re past is irred l1cible to both the former present
that it was and the present present in which it might reappear as a
recollection. It telescopes these two presents together and insists only
in the 'q l1 alitative difference' between the two (DR 122/1 60).
It might be tempting to read Dele l1 ze as if he were here s l1ggesting
some sort of Platonic forms or essences , b l1 t this interpretation is
inc Olnpatible with Dele l1 ze's call for a reversal of Platonism' (see , for
instance , DR 126-8/1 65-8 and Appendix 1 in LS 253-66/292-306).
Dele l1 ze rejects Plato's hierarchy of transcendent eternal forrns and
the m111tit l1 de of copies and sIn111acra spreading O l1 t on the sl1 rface of
the earth. There are no originals , eternal forrns or essences of which
we can beconlereminiscent\Instead Dele l1 ze rather specifies 'reminiscence' as being sim111taneo l1 s1y an act of creation. In c it is art
that prod l1 ces p l1 re essences , reveals t
227
be. What is meant is perhaps a kind of contrac)n or condensation of all possible experiences of Comb 1' ay in a complex intensive
impression. More precisely , the a 1' tist does not rnerely represent his
own lived experiences , his own perceptions and affections in the
artwo 1' k , but he also contracts those of the lives of others. Properly
speaking , the lnatte 1' is no longe1' of pe1'ceptions and affections , but
of universal percepts' and affects' terms that Deleuze introduces in
his lateuvre. We thus deal with a kind of de-subjectified sensation ,
in Deleuze's words , a 'sensation in itself' (cf. WP 164/155) or a bloc
of sensations' (WP 164/1 54). As Deleuze says:
We write not with childhood memories but through blocs of childhood
that are the becoming-child of the present. [. . .] We attain to the percept
and the affect only as to autonomous and sufcient beings that no
longer owe anything to those who experience or have experienced them:
Combray like it never was , is or will be lived; Combray as cathedral or
monument. (WP 168 /1 58)
In his early work of the 1960s, Deleuze has not yet found these
concepts. With rega 1' d to the example of Combray in itself' , Deleuze
1' ather uses the P1' oustian exp 1' ession of a pu 1' e essence or a shred of
pure past ('a bit of time in the pure state').29 It is important to keep
in mind that Deleuze does not have recourse to a type of mystic
experience , a kind of fusion with the pure past , a relniniscence of
pure essences , but that he invokes the necessity of an act of artistic
creation. It is true however that the artist, for Deleuze , does not
act as a God-like creator or originato r. On the contrary , the artist
creates only under the condition of the ressure of the work of art.
He is forced to create; he is not free to choose the conditions of his
creation. We may perhaps say that there are intensive forces which
surpass the artist but which nonetheless den1and to be given expression in the work of ar where they becon1e re-created in a bloc of
sensation'. lndeed , we must understand Deleuze's interp 1'etation of
a pure past not as an independent transcendent realn1 , but rather an
irnrrlanent field of intensive forces that require a recreation in the
work of the artist.hey are the condi
tion of any great work
0 fa
rt(0r
in
de
ed any excessive act) bL
lt
a t the san1time they bec:
om
le l'e-c1
in the
r esu
lt whi
ich is ca
iled a pure essence or the in-itself.
Deleuze follows up the idea of artistic creation , 0 1' rather creation
per se: creation can also nlean the performance of a transgressing ,
excessive act. 1 e will see that this issue becornes of crucial impol
tance in the third synthesis of tinle. It should be noted that between
228

the second and the third syntheses of tirne prevails an obvious
affinity.
he
Proustian formula a little time in its pure state' refers frst to the
pure past, the in-itself of the past [. . .], but more profoundly to the pure
and empty form of time, the ultimate synthesis, that of the death instinct
which leads to the eternity of the return in time. (DR 122/160)
We would argue that the passive synthesis of the pure past becomes
ultimately absorbed in the third synthesis of pure time. A rst indication can be found in The Logic of Sense , where Deleuze abandons the
model of three syntheses of tirne and simplifies his account of time
by turning to the dual rnodel of Chronos and Aion. It is obvious that
Chronos has to be equated with the first synthesis of tIne , that of
perceptual syntheses and habit. Aion bears a strong resemblance to
the third synthesis of time , that of empty time which is constituted by
the cut' , that is an excessive act or event , dividing past and future.
Since the second synthesis of the pure past or Memory signifies an
event - the involuntary rnemory and the artistic creation of a pure
essence ( a bit of time in the pure state') it should be counted on
the side of Aion. 30 Evidence for this thesis can be found in Deleuze's
treatn1ent of Bergson's notion of the pure past in Cinema 2: The
Time-Image. Here Deleuze suggests a third model of Bergson's cone
of Memory: a line , which at one point splits itself into two distinct
arrows. 31 As Deleuze explains:
ime
has to split itself in two at each moment as present and past, which
differ from each other in nature , or what amounts to the same thing , it
has to split the present in two heterogeneous directions , one of which
is launched towards the future while the other falls into the past. (CIT
79/108-9)
We will come across this split into a past and a future , a before
and an after , with regard to the third synthesis of time. Moreover ,
Deleuze explicitly brings together Bergson and Kant (although with
some reservation):
Time is not the interior in us , but just the opposite the interiority in
which we are, in which we move, live and change. Bergson is much closer
to Kant than he himself thinks: Kant defned time as the form of interiority , in the sense that we are internal to time (but Bergson conceives this
forrn quite differently from Kant). (CIT 80/110)
I11 both Bergson and Kant , Deleuze recognises a de-psychologisation

of time: time is no longer defined as the suc
229
but as a 'virtual coexistence' of sheets of time , or respectively as

the fonTl of empty tirrle which conditions the err irical subject as a
ten1poral being. Although in a very different manner , both Bergson
and Kant conceive of time as a condition that first of all grounds
perceptual syntheses or active syntheses of men10ry and thus allows
for processes of subject forrnation. Speaking boldly , tin1e can be
called the true agent , the true subjectivity on which we rely as the
constitutive ground. As Deleuze states with reference to Bergson:
the only subjectivity is time , non-chronological tirrle grasped in its
foundation , and it is we who are internal to tin1e , not the other way
round' (CIT 80 /1 10). It ren1ains to be seen how Deleuze hirrlself will
specify the transcendental aspect of time. The transcendental is surely
not to be understood in terms of a ground (fondement). As Anne
Sauvagnargues aptly says:
1n order to express this new relationship between time and subjectivity , Deleuze creates the neologism 'effondement' [ungrounding] , from
'effondrement' [collapse] and absence of fondement' [ground] , in order
to define the non-original character of time , and to indicate his polemic
hostility with respect to phenomenologies of foundation and origins. 32
Deleuze's Static Synthesis ofTime

As we have seen , Deleuze credits Kant for revoIutionising the concept
of time by defining it independently from Inoven1ent as a pure and
empty form , that is the form of interiority or inner sense. He particularly points out the revolutionary mon1ent , when Kant introduced
hat schizophrenia in principle' (DR 58/82) , that is the fractured 1
or split subject. However , Deleuze deviates from Kant in an in1portant sense: the dissolved self as Deleuze envisages it is not congruent
with the Kantian empirico-transcendental subject. Deleuze dees
Kant's reintroduction of a synthetic identity and the attribution of
all power of synthesis to the transcendental 1. His n10del of the three
syntheses of time in Difference and Repetition challenges the Kantian
account of subjectivity with its strict distribution of the empirical
and the transcendental , of passivity and activity. le have seen that
Deleuze's first synthesis of tin1e , that of the living present , constitutes
subjects as forn1ations of habit. The second synthesis of t lne , that
of the pure past , establishes the coexistence of several levels of the
past. lt shows that the individual life in fact presupposes the whole
of the past , that is not only its own past experiences and lTlemories ,
but also those of others. Thus by lneans of the first two syntheses ,
230
Deleuze attempts to account for processes of habituation , subjectification and artistic creation. Now , by rneans of the third synthesis
of time , that is the tirne of the future , Deleuze envisages a process of
dissolution of subjective structures and of becorning. Deleuze's third
synthesis of time is arguably the most obscure part of his tripartite
theory , as Deleuze mixes different theoretical concepts drawn from
philosophy , Greek drama theory and matherrlatics. Of central irnportance is the notion of the caesura or cut, which is constitutive of the
third synthesis of time defined as an a priori ordered terrlporal series
separated unequally into a before and an after. This ordinal definition of time , we will argue , is heavily inspired by Kant's definition of
time as pure and empty form , H lderlin's notion of caesura' drawn
from his Remarks on Oedipus' (1803) and Dedekind's method of
cuts as developed in his pioneering essay Continuity and Irrational
Numbers' (1872). In the last section of this chapter , we will then
see how Deleuze ties together the conceptions of the Kantian empty
form of time and the Nietzschean eternal return , both of which are
essentially related to a fractured 1 or dissolved self.
HLDERLIN'S CAESURA
At several places, Deleuze cites Hlderlin as the true descendant of

Kant (DR 58/82; DR 87/1 18). Hlderlin is one of Kant's best disciples' (LK 1, p. 14) , who poses the problem of the Kantian theory of
time on the level of Greek tragedy and makes the effect of pure and
empty time palpable. Time becomes a desert , the straight line which
Oedipus wanders (LK 1, p. 14).
The two core texts which Deleuze refers to are Hlderlin's cryptic
Remars on Oedipus (presumably written around September 1803 )33
and Jean Beaufret's small book Hlderlin et Sophocle (1965) , which
gives a very Kantian inspired interpretation of H lderlin's Remas
on Oedipus. 34
Hlderlin interprets Sophocles' OediPus Rex as the undoing
of the coupling between rnan and god: lnan and god become 'an
unlimited One' (das grenzenlose Eineswerden) , which , however , can
e purified' only through an 'unlin1ted separation' (grnzenloses
Scheiden).35 Deleuze describes this unlimited separation as a double
deviation: 'God turns away from man who turns away from God'
(LK II , p. 4). While in the tragedies of Aeschylus or Euripides the gods
still ensured j ustice , they punished and pardoned according to their
j udgement , Sopho c1 es' tragedy marks a significant char The bond
231
between rnan and the gods has broken. The gods turn away frorn
rnankind. Sophocles' tragedy Oedipus Rex shows how Oedipus ,
raging against the di vine betra yal ( gttliche Untreue' in H lderlin's
words) , searches desperately for who he is and tries to recover his
identity. As H lderlin says , at such rnornents rnan forgets hirnself
and the god and turns around like a traito r. '36 Sophocls designated
Oedipus as atheos , which not only rneans being a non-believer but
literally being separated frorn GOd. 37 Even when Oedipus' crirne
lS nally discovered , that is when the blind seer Tiresias reveals to
Oedipus that he had killed his own father Laius and rnarried his own
rnother Jocasta of Thebe the gods do not punish Oedipus through
an irnrnediate and brutal death. Instead , Oedipus' death is his long
and lonesorne wandering with no airn and no end in sight. According
to Beaufret's interpretation , rnan has to learn to endure the absence
of God and to accept the abandonrnent. 38 his is , in effect , the
essence of tragedy (in Gerrnan Trauer-spiel , a 'play of rnourning').
Heaven has becorne a transcendent realrn (the Kantian 'starry
heavens' above the head) , whence follows the unlirnited separation
of heaven and earth.
As we have already rnentioned Sophocles' tragedy Oedipus Rex
clearly breaks with the forrn of tragedy in Aeschylus and Euripides ,
where the unity of the cosrnos was still intact, and where the divine
law revealed itself in the order of the universe , the course of nature
and hurnan fate.his rneans that in the tragedies prior to Sophocles ,
the destiny of the characters was settled frorn the beginning. The
gods ensured that justice was done and atonernent lnade for every
excessive act that violated the divine law. For instance , in Aeschylus'
Agamemnon , we can distinguish three unequal episodes that together
forrn a 'cycle of lirnitation , of transgression and of atonerner (LK II ,
p. 3). In the rst episode the great Agarnernnon rules over Mycenae ,
the rnost powerful kingdorn in Greece , a realrn of order , law and
lirnits. In the second episode occurs the excessive act , the act of inj ustice: upon Agarnernnon's return to Greece fron1 the rOJan 1a his
wife Clyten1nestra assassinates hirn. This is the lnornent of transgression or violation of the lirnit. In the third and last episode , Orestes ,
the son of Agan1ernnon and Clyternnestra , avenges his father by
killing his n10ther and new husband Aegisthus. Thus the atonen1ent
for
232
When Agarnernnon enters the palace , Cassandra , princess of roy

and abdllcted as Agarnernnon's concubine , has a vision of both the
crime and also Orestes' subsequent revenge. Bllt her prophecy has
no effect on the course of the events at al l. Her vision is nothing
bllt a confirmation of what is set frorn the beginning. Picking IIp
H lderlin's phrasing , both Beallfret and Delellze say that the beginning and end of the tragedy are bent in such a way that they correspond or rhyme' with each other. Time is conceived as a cllrve.
Sophocles , on the other hand , suggests a new conception of tirne. He
lln-curves (dcourber) time , turns it into a straight line , and Cllts it
by a caesllra which will produce a before' and an after' which no
longer rhyn1e together. 40 Indeed , Tiresias' revelations - that Oedipus
is a child of Thebes at he is brother and father to his own children ,
and son and husband to his own mother - forrn a caesllra. Hlderlin
describes the moment of the caesllra as follows:
In the lltmost form of sllffering [. . .] there exists nothing bllt the conditions of time and space. Inside it , man forgets himself becallse he exists
entirely for the moment , the god [forgets himsel f] becallse he is nothing
bllt time; and either one is llnfaithflll , time , becallse it is reversed categorically at sllch a moment , and beginning and end no longer rhyme [und
Anfang und Ende sich in ihr schlechterdings nicht rein1 lt]; man
becallse at this moment of categorical reversal he has to follow and thllS
can no longer resemble the beginning in what follows. 41
At the lnoment of extreme suffering, Oedipus is left with the pure
form of tIne , which is emptied of all meaningflll content and
annollnces neither punishn1ent nor relief from the interminable ,
incessant suffering. As H lderlin says in very Kantian terms ,there
exists nothing bllt the conditions of time and space'.42 How are we to
llnderstand this clairn? Beallfret exlains that after the turning-away
of God , Oediplls has to face the immensity of an empty heaven
without ground. God is fron1 then on no longer a father , a friend ,
not even an adversary to con1bat. 3 God is nothing but empty time.
his is why man is ultimately thrown back on himself. Bllt for
Beallfret this means that he has to find the law in his own natllre ,
to reconstitute himself on the basis of the n10ral nature of reason ,
in other words it is in Kant's rnoral philosophy (the categorical
imperative) that Beallfret finds the solution for man abandoned by
GOd. 44 Contrary to Beaufret , Deleuze draws a very different conclusion from the tllrning away of God. Man cannot be thrown back on
hin1sellbecallse that self is shattered. Kant breaks tin1e and therefore
LJJ
CONDITIONS OF
HOUGHT:
the self, but the Kantian rnoral philosophy is not the solution to
this but its covering ove r. Hence Deleuze identifies Nietzsche rather
than Kant as the philosopher who had the courage to face up to the
consequences of the Kantian theory of time. The explosive moment
opens the possibility of the dissolution of the self and the liberation
frOITllaw and judgernent imposed by the gods or hurnan reason. The
caesura marks this explosive moment , the final brea1 up , the appearance of the fracture or crack in the I. The story of Oedipus , then ,
goes as follows: Oedipus is trapped in the pure instant of time , from
which a past and a future wiU be produced on the straight line , which
is to say a before and an after which no longer rhyme together' (LK
II , p. 4). Tiresias' intervention has put before Oedipus the thought
that he rnight not be the son of King Polybus of Corinth and his wife
Merope who raised him. This is a thought which is almost ll1possible
to think. All of the personal n1emories in which Oedipus has believed
so far , together with his future expectations are eliminated , destroyed
at a single blow. He can no longer be and resen1ble what he has been
before. In fact the caesura is not only a break in tin1e , but also a
split of Oedipus' self. Oedipus is other to hin1sel f. He experiences
this internal difference in the pure present the 'pureness' of which
signifies that it occurs like a cut. The series of fonner presents do not
converge with this present mon1ent.
Deleuze compares Oedipus' experience of splitting with that of
Halnlet who , just like Oedipus , is brought to a state of internal difference with himself. Through the apparition of the ghost of his father ,
Hamlet learns that his fathe King Harnlet, was murdered and he
swears to exact revenge on the rnurderer , his uncle Claudius , who
has meanwhile married Hamlet's mother Queen Gertrude. However ,
Hamlet hesitates for a long time in his task of avenging the fathe r.
Only when he is sent on a sea voyage to England by Claudius , who
conspires to have him killed on this journey , does Hamlet finally find
himself capable of committing the act of vengeance. In projecting
an ideal self, that is the future agent of the excessive act , Hamlet
detaches himself fro111 his past. The time at which the irnagined act
appeared to be o big' is gone.
In both cases Oedipus and Hamlet , Deleuze recognises an a priori
order of time , determined by the caesur The caesura must be
understood
234

throw time Ollt of joint, to make the slln explode , to throw
oneself into the volcano , to kill God or the father' (DR 891120). The
symbolic act of killing God recalls Nietzsche's farnolls description in
The Gay Science.
ways:to
We have illed him - you and I! We are all his murderers. But how did
we do this? How were we able to drink up the sea? 1ho gave us the
sponge to wipe away the entire horizon? What were we doing when
we unchained this earth from its sun? Where is it moving to now?
Where are we moving to? Away from all suns? Are we not continually
falling? And backwards , sidewards , forwards , in all directions? Is there
still an up and down? Aren't we straying as though through an innite
nothing? Isn't empty space breathing at us? Hasn't it got colder? Isn't
night and more night coming again and again? Don't lanterns have to
be lit in the morning? Do we still hear nothing of the noise of the gravediggers who are burying God? Do we still smell nothing of the divine
decomposition? 5
Nietzsd 's
rnadman goes on to ask: 'Is the rnagnitllde of this deed

not too great for llS46 When he only encollnters silence and incomprehension from the crowd of atheists in the marketplace , he concllldes that , indeed , the excessive act of killing God is too big for llS
the trernendolls event' has not yet reached the eyes and ears of lnen.
Speaking philosophically , the event has not yet become effective. he
rnoment the event does becorne effective , it will destroy the sllbject.
The traditional sllbject, whose identity is gllaranteed by God , will be
dissolved.
Deleuze eqllates the excessive act , in whatever image it rnay be
symbolised , with the caesllra which clearly separates a before and
an after. Thus the past is defined a priori as the before , that is the
time where one is not yet capable of the act. The pllre present ,
which refers to the caesllra itse lS he present of metamorphosis ,
a becom
llng-.
projection of an idea1 self in the ln
lag
e of the act (thiS
1 s marked by
Ham
lle
t's sea voyage and by the olltcome of Oedi
pll1
s's enq lliry: the
hero becom1es "capable" of the act)' (DR 89/121). Delellze remarks
that it lnatters little whether the act has been actllally performed or
not - 'Oedipus has already carried out the act , Halnlet has not yet
done so' (DR 891121). In the case ofNietzsche's aphorism ~125 the
ITladman accuses mankind of having already murdered God (God
'has bled to death under ou
235
done it themselves!').47 In the end , the empirical incarnation of the

syrnbolic act does not count. It is not the er
mplrical content of tim
le
that
dist1
determ
1 ned
:
the past is the

e time before the caesura; the pure present!
is the becoming
e qua
1 to the act and the experience of internal difference (between the past self and the ideal self in the in1age of the act);
nally the future is the tin1e after the caesura. he future n1arks the
time when the excessive act turns back against the subject , destroying
its identity and dispersing it in a discrete multiplicity of little selves ,
of egos with many nalnes or , what an10unts to the san1e thing , a
universal ego with no name at all (cf. DR 90/121).he theme of
the dissolution of the identity of the self in favour of a system of
little selves , of the n1an with no narne and no qualities , the universal
people or of diverse becomings' (becon1ing-plant , becoming-tVoman
becoming-animal , becoming-child , becon1ing-n101ecule , becorningother , becolning-imperceptible , etc.) is often repeated in Deleuze's
uvre. 48 We will con1e back to this point later in relation to
Nietzsche and Klossowski.
DEDEKIND'S CUT
Deleuze reaches for a new conception of time , an ord1 tin1e 'which

ceases to be cardinal and becomes ordinal , a pure order of time' (DR
88/120). According to this new conception , tin1e is no longer en1pirically detennined by means of the changes that occur , the events
that take place within it or the periodical rnovements of planets that
pass through cardinal points. Instead , time is a priori determined by
n1eans of the caesura' which distributes unequally on both sides the
before and the after.
The future and the past here are not empirical and dynamic determinations of time: they are formal and xed characteristics which follow a
riori from the order of time , as though they comprised a static synthesis
of time. The synthesis is necessarily static, since time is no longer subordinated to movement; time is the most radical form of change , but the form
of change does not change. (DR 89/120)
The idea of the form of time , which does not change itself, is
clearly a Kantian thought. Repeatedly , Kant says in the Crtique
of Pure Reason that 'tin1e itself does not alte but only sOlnething
that is within time' (CPR A 4 1/B 58).49 But is it true that Deleuze
still adheres to the traditional philosophical conception of form?50
236

Deleuze's remarks on a static synthesis' should make us suspect
that Deleuze is not sirnply repeating Kant's theory here. Egually ,
the idea of a caesura which constitutes a serial and linear time by
distributing a before and an after , indicates a source other than Kant.
For sure , Deleuze deduces the term caesura' rst and foremost from
Greek drama theory , but in his lecture course on Kant of 21 March
1978 , he does not hesitate to compare it to the mathematical terms
'limit' and cut' (coupure).51 Moreover , the word caesura' derives
from the Latin root caes' and can thus be rendered as cut'. For these
reasons , we suggest that the caesura' can be understood by means
of the concept of cut' in Dedekind's theory of real numbers , which
Deleuze discusses in Chapter Four of Difference and Repetition in
the context of the continuousness of Ideas (DR 172/223).
he
mathernatician Richard Dedekind (1831-1916) is famous for

giving a rigorous arithmetical foundation to differential calculus , and
thereby expunging fron1 calculus undefined geornetric concepts such
as infinitesin1a l' quantities and the limit concept involving the idea of
approaching. 52 Deleuze acknowledges his achievements , in particular
the renewed conception of limit: 'The limit no longer presupposes the
ideas of a continuous variable and nfinite approximation. On the
contrary , the notion of lin1it grounds a new , static and purely ideal
de nition of continuity' (DR 172/223).53 Indeed , it can be argued
that Dedekind invents this new 'static and purely ideal' conception
of continuity. His thoughts on continuity were first published in the
pioneering essay 'Stetigkeit und irrationale Zahlen' ( Continuity and
Irrational Numbers') in 1872 , fourteen years after he developed the
basic ideas on which it relies. The main guestion the essay deals with
is: what is the nature of continuity? As Dedekind states:
An explanation of this continllity is nowhere given; even the most rigorous expositions of the differential calculus do not base their proofs
upon continllity bllt [...] they either appeal to geometric notions or
those suggested by geometry , or depend upon theorems which are never
established in a Pllrely arithmetic manne r. 54
That is , the notion of continuity was either geometrica l1 y explained
as a vague hang-togetherness an unbroken connection in the smallest parts' ,55 or was based on insufficiently founded theorems , such as
that every lTIagnitude which grows continually , but not beyond all
limits , lTIUSt certainly approach a limiting ralue. Dedekind set himself
the task of securing a real definition of the essence of continuity.
237
He first approached the problem by trying to 'map' the geometrical

continuum (the straight line) onto ordered systems of discrete guan
tities (numbers). Comparing the system of rational numbers with
the points of the straight line , he saw that they cannot be put into a
one-to-one-correspondence with each other: Although each rational
numer can be correlated with a point on the line , not every point of
the line can be expressed as a rational nurnber. In fact , in the straight
line there are infinitely rnany points which correspond to no rational
numbe r. To identify these points , which are inexpressible as rational
numbers , Dedekind used a method of division: the Dedekind cuts'
(Schnitte). Any cut divides the points of the line into two classes , such
that all the points of one class are always to the left of all the points
of the other. Furthermore , there is precisely one and only one point
determined by this cut. The cut can correspond to a rational number
or else designate a 'gap' between the rational numbers , that is an
irrational guantity (suc as -12). In the latter case , cuts define a new
type of number , i.e. the irrational numbers.
Dedekind made use of geometrical considerations in order to
introduce the notion of the cut. However , he sought to define cuts
directly in terms of the nurnber system so that any reflections on
ge01netric lines can be put aside. As Robert Bunn says:
Geometry was to serve only as the source of the idea for consrructing an arithmetical foundation. The conrinuous system which was to
be Dedekind's foundarion would be arithmetical in rhe sense that its
operarions would ulrimarely be defined in terms of operations on natural
numbers , and no mention would be made of any geomerrical objecrs. 56
Dedekind demanded that the systern of rational numbers be irnproved
by le creation of new nurnbers such that the domain of numbers
shall gain the sanle cOlnpleteness , or as we may say at once , the
same continuity , as the straight line' .57 As it is , the system of rational
nmnbers is marked by a certain incornpleteness or discontinuity ,
due to the existence of gaps. Thus , in order to render the domain
of rational nunlbers into a continuous systern , Dedekind defined for
,
un
lrr er a '
new ob
Ject'
each cut that is not produced by a rational nl
which he called an irrational num
lber. Thus -1 2 can be defined as the
cut between two classes A and B, where A contains all those numbers
whose sguares are less than two and B those whose squares are
greater than twO. 58
It should be noted that Dedekind does not identify irrational
numbers with cuts , snce every definite cut produces either a definite

rational or irrational numbe r. Rather , Dedekind cuts constitute the
next genus of numbers' (DR 172/223) , narrlely real numbers\59 he
new genus of real nurnbers allows the treatment of both rational
and irrational nllmbers as elen1ents in an encompassing nllmber
system , which forms a continuous and ordered system. It has to be
emphasised that this continuity of the number system is something
quite different from the traditional conception of continuity founded
on the intuition of the way in which geometric quantities arise:
according to an intuitive conception of continuity , a line is consid
ered continuous insofar as it arises from the continuous mOVelTIent
of a point , and a plane from the movement of a line. By contrast ,
Dedekind's new conception of continuity claims not to rely on intui
tion , or any considerations of smooth movement. It claims to contain
nothing en1plflC since it can be deduced from number systems
only. In a way Dedekind's project of arithmetising the conception of
continuity is grounded on an Idea of reason: the Idea of an infinite ,
ordered and dense set of numbers each of which can in principle
be identified by a Dedekind cut. 60 According to Bunn , Dedekind's
continuitcan better be described as the property of the completeness' that characterises certain densely ordered number systerns. 61
Bunn remarks that 'the tern1 "continuous" is not an especially apt
one for the characteristic involved , but it indicated the correlate in
the old system - continuous magnitude.'62 Boyer explains t: hat the
notion of continllity 'specifies only an innite discrete multiplicity
ol elements , satisfying certain conditions - that the set be ordered ,
dense , and perfect'. 63 He goes on to argue that that there is nothing
dynamic in the idea of continui '.64 As Deleuze says , we gain 'a new ,
static and purely ideal definition of continuity' (DR 172/223) , which
is grounded on the Dedekind cut, inasmuch as it constitutes this new
continuous and ordered systerrl of real numbers. Deleuze also calls
the Dedekind cuts the ideal cause of continuity' (DR 172/223).
Now the question to ask is: in what way does Deleuze benefit from
Dedekind's theory of cuts and the new notion of continuity that designa tes not a vague hang-togetherness' but rather an infinite , discrete
lTIultiplicity of elernents whose order is a priori detern1ined? Deleuze
uses Dedekind's ideas in order to construct a time that is not empi"
cally defined through our intuition of a dynan1ic flux of events , but
one that is determined a priori and designates a static state ol aalrs.
This latter time is a 'static synthesis' of divergent series of tin1es
(series of the past , the enduring interval of the present and future
239
becomings) which are distributed by the caesura , i.e. 'a genuine cut
[coupure]' (DR 172/223) , into a before and an afte r. The caesura or
cut is constitutive of this ordered system of time which maps onto
the straight line. Thus Deleuze transforms the Kantian definition of
a purely forrnal time by rneans of mathematical considerations of the
notion of 'cut' and static synthesis'. For Deleuz the third synthesis
of time is not simply an a priori subjective form , but an a priori and
a-subjective static synthesis of a multiplicity of series of t lnes.
However , it should be noted that Deleuze's account of serial time
as a straight line is not as straightforward as it appears. Deleuze does
indeed make use of the Dedekind cut and the idea that it constitutes
a static continuum , yet he takes licence in modifying Dedekind in a
way that is not Dedekindian at al l.he way that Deleuze conceives
the series of time retains the idea of the irrational cut designating a
'gap\In Difference and Repetition , Deleuze says that the irrational
numbers [. . .] differ in kind from the terms of the series of rational
nU111bers' (DR 172/224). hey are 'constructed on the basis of an
essential inequality' in relation to the next-lowest type of numbers ,
i.e. the rational nurnbers. That is to say , they express the lnpossibil
ity of detern1ining a common aliquot part for two quantities , and
thus the impossibility of reducing their relation to even a fractional
number' (DR 232/299). However , they cOlnpensate for their characteristic inequality by their 'limit-equality indicated by a convergent
series of rational numbers' (DR 232/299). Her Deleuze seems to
borrow again fron1 Dedekind , who considered irrationals as limits of
convergent series of rationals. Now the interesting move by Deleuze
is to ascribe an original intensive nature to irrational an implication
of difference or inequality , which is cancelled or covered over as soon
as they are constructed as elements of an extensive plane of rational
numbers. In fact , Deleuze holds that an intensive nature belongs to
every type of number , insofar as they are not explicated , developed
and equalised in an extensity.
Every number is originally intensive and vectorial in so far as it implies a
difference of quantity which cannot properly be cancelled , but extensive
and scalar in so far as it cancels this difference on another plane that it
creates and on which it is explicated. (DR 232/299-300)
Deleuze's reflections on the nature of irrationals show that he regards
the nU111ber line as a ction a spatial irnage which covers over an
intensive depth. The straight line of rational points is but a false
in f nity , a sin1ple unde f nite that includes an in f nity of lacunae;
240
in1e
and the Split Subject
that is why the continuous is a labyrinth that cannot be represented

by a straight line' (FLB 17). We have already come across Borges'
labyrinth of the straight line. hus it seems that Deleuze , when he
speaks of a line of time , does not mean a simple straight line , but one
that is perforated by lacunae. 65 These lacunae or gaps are precisely
designated by the irrational cut , that is the interstice between series
of rational numbers. hey syn1bolise the irruption of the virtual
event within the empirical continuun1 of space and the chronological
succession of instants.
Deleuze is thus not faithful to Dedekind , but he uses Dedekind's
mathematical notion of the cut and the static continullln in order to
get rid of empirical representations of time. We believe that Deleuze's
recourse to the Dedekind cut can serve as a response to the question
that James Willian1s has raised in his recently published book on
Deleuze's philosophy of time. Considering Deleuze's references to
exarnples and voca ulary from drama to introduce and explain the
third synthesis of tirrle , Williams wonders how Deleuze can claim
that the third synthesis involves a formal cut , when in fact it is
deduced from a somewhat narrow dramatic event (the appearance
of a ghost to the Prince of Denn1ark)'. 66 Williarrls states that the
dralnatic event has to be defined formally as a cut that assembles
what comes to either side of it, but he does not make any reference to
Dedekind cuts. He also rightly insists that Deleuze's third synthesis
of tIne is a priori and not dependent on empirical observation. We
suggest that Dedekind's theory of static continuity and his method of
cuts helps to explain Deleuze's a priori and ordinal definition of time ,
which is a point of intersection of various heterogeneous theories.
Deleuze does not shy away from bringing together Kant's empty
form of time with Hlderlin's caesura and Dedekind's cuts.
lt is important to note that Deleuze's method of mixing theoretical
concepts is not to be conceived in terms of assimilating differences
and blending one concept into the other. Rather Deleuze maintains
their heterogeneity and relates them to one another as differences. It
can be regarded as a technique of montage operating by cuts commonly used in cinema. 67 In Deleuze's final co-authored work with
Guattari , What Is Philosophy? , this n1ethod of montage is called
'thought as heterogenesis' (WP 199/188). The authors do not deny
that the different disciplines that are brought together are distinct
with regard to their objects , n1ethods and modes ()f enunciation'
(WP 1271121) - thus , philosophy creates concepts , art erects blocs of
241
sensation , and science constructs functions - and that none of thern

is superior to the other frorn the point of view of thought or creation. 68 However , the respective fields of creation can intersect and
establish a 'rich tissue of correspondences' (rp 199/1 88). Deleuze
and Guattari are aware of the dangers threatening this rnethod
of 'thought as heterogenesis': for example , in doing philosophy
one could be l1listakenly regarded as trying to do science or art ,
or to e111ploy scientific concepts as lllere examples or metaphors.
Nevertheless , speaking from the point of view of philosophy only,
the authors clain1 that philosophy has a fundan1ental need for these
encounters with other disciplines (cf. WP 162/1 53). Events in science
and art can provoke new problems in philosophy. ln fact , philosophical concepts frequently involve references to science and to art , which
is not to say that philosophy inevitably becomes a positivist theory
of science or a kind of poetical thought. According to Deleuze and
Guattari , philosophy has its own specic task: it creates concepts
and brings forth events. However e components of concepts can
incorporate heterogeneous elelllents frolll other fields of creation
and thus further and enrich the development of the new. 'Thought
as heterogenesis' n1eans a becon1ingother (hetero) of thought , and
it does so as a response and resistance to the repressive function
of an academic philosophy that buries its young generation under
the volumes of the history of philosophy and excludes any lateral
encounters. 69 Philosophy is not Sil1 ly an acaden1ic discipline , but
a power that challenges us to think differently and to create new
problems. Deleuze's theory of tin1e is a particularly telling exarnple
of this method of thought as heterogenesis: it exhibits a great variety
and heterogeneity of elements demanding much openness and effort
on the part of the reader. However , if one is willing to go on this
journey , one is rewarded with stunning new thoughts or at least
problellls that can incite new investigations.
Nietzsche's Eternal Retum

As it has been shown , Deleuze's third synthesis of time s profoundly
linked to the notion of a f1'actu 1' ed 1, and thus takes as its point of
departure Kant's empirico-transcendental subject , split by the en1pty
forrn of tin1e. However the Kantian subject of the first Critique
is after all a subject of knowing. It strives fo 1' the reproduction of
representations that can count as knowledge. lt is itself the source
of representations: th transcendental 1 affects the passive empirical
242

self through the interior fonn of time. he Kantian transcendental
conditions are conditions of knowledge qua representation. Deleuze ,
on the contrary , searches for conditions under which something new
is produced: a new thought , a new WOI a political act, even a new
subject. It is in the direction of Nietzsche that we must look in order
to specify the transcendental conditions of creative production. The
answer lies in Nietzsche's thought of the eternal return.
It should be noted that Nietzsche never fully laid out his thought
of the eternal return in his writings and that the existing interpretations in secondary literature vary to a great extent. What interests us
here is solely Deleuze's reconstruction of the eternal return , which
is considerably influenced , as we will see , by Pierre Klossowski's
reading of Nietzsche .1O
According to Deleuze , the thought of the eternal return is not to
be understood as a return of the Same or the Similar. Rather , what
passes the test of the eternal return is that which differs internally ,
simulacra or the dissolved sel f. The eternal return has to be seen as a
test of selection , which banishes identity , that is the identity of God ,
the identity of the world or the represented object, and the coherence
of the self. However , it is Inportant to note that the eternal return
does not only have a destructive or lethal impact , rather it manifests
a positive and productive powe r. Thus the subject , which succumbs
to the power of eternal return , is carried to a point of metan1orphosis , when all its possibilities of becoming are set free. o illustrate
this fact , Deleuze quotes fr Oln Henry 1iller's book on Rimbaud: I
realised that 1 was free , that the death 1 had gone through had liberated me' (DR 19/30) .1 1 Hence the landscape of the eternal return is
that of differencentself the disparate , the irreducibly unequal , and
also that of metamorphosis , chance , multiplicity and becoming. In
Deleuze's words:
Essentially , the unequal , the different is the true rationale for the eternal
return. It is because nothing is equal , or the same , that it' comes back.
In other words , the eternal return is predicated only of becoming and the
multiple. It is the law of a world without being , without unity , without
identity.72
Therefore , in a first move , it is important to distinguish Nietzsche's
eternal return from the conception of a return of the Same and the
SInilar' which is the essence of a cyclical conception of time. Thus
the ancient Greeks presupposed an identity or resemblance in general
of all the instances that are supposed to recu r. They regarded the
243
recurrence of planetary motion , the uniform change of seasons and

qualitative changes in things as laws of nature. According to Deleuze,
Nietzsche's thought of the eternal return cannot be presented as
a natural law and identied with the ancient Greek hypothesis of
cyclical time .7 3 First , he argues that as a connoisseur of the Greeks ,
Nietzsche could not have been ignorant of the Greek hypothesis of
time as a cycle. Thus when Nietzsche insists that his thought of the
eternal return is something effectively new , we have to take him
seriously. Furthermore , Nietzsche is a thinker who is very much
opposed to the notion of law. I-Ie would not have submitted to the
simple notion of a law of nature. As textual evidence , one can adduce
two passages in Zarathustra , where Nietzsche explicitly rejects
the interpretation of the eternal return as cyclical time: (1) During
the encounter between Zarathustra and the dwarf , the dwarf says
'All truth is crooked; tirne itself is a circle' , whereupon Zarathustra
replies 'Thou spirit of gravity! [. . .] do not take it too lightly' .74 (2)
On another occasion , Zarathustra rebukes his animals that they have
already made a refrain' out of his doctrine of the eternal return. In
a reain the same always returns , but apparently Zarathustra does
not want the eternal return be understood as a refrain. 75
In what way then is the eternal return different from a natural
cycle , a cycle of time? The crucial issue is at the eternal return is
selective and creative. It is selective with regard to desires or thought
and with regard to being. Let us frst consider its 111echanism of selection with regard to desires. lf the doctrine of the eternal return is
stated in terms of an ethical rule , it becomes a sort of Kantian imperative: 'Whatever you will , you have to will it in such a way that you
will its eternal return.' That which is expelled by the selection test of
the eternal return are all instances of willing that want a thing only
this one time\In an unpublished note conten1poraneous with The
Gay Science , Nietzsche says that it does not n1atter whether the act 1
am about to perforrrl is inforrrled by ambition , or laziness , or obedience , if only 1 re-will my present action again , innumerable times .7 6
The eternal return excludes any half-hearted wi l1 ing and a 1111S the
extreme forms. lt separates an active , superior will which wants to
enact its force to its highest power , fron1 a reactive , gregarious wil
244

he
eternal return creates the superior forrns. It is in this sense that the
eternal return is the instrurnent and the expression of the will to power: it
raises each thing to its superior form , that is , its nth power .7 7
he
eternal return is thus the accornplice of the will to power. Both

the will to power and the eternal return are in effect the transcendental conditions for the creation of something new , that is of superior
forms that pass the test of eternal return. These superior forms
are what Nietzsche refers to with his concept of overman': The
Overrnan is defined as the superior form of everything that "is" , (DR
41160). But how can we understand this very vague concept of the
overn1an? As Deleuze explains:
The overrnan very rnuch resembles the poet as Rimbaud defines it: one
who is loaded with hurnanity , even with anirnals ,' and who in every case
has retained only the superior form , and the extrerne power. n
Elements of the overman can thus be found in the artist , the poet - or
as we might say , someone who is willing to llndergo metamorphoses
and to become-other in favour of an act or a work yet to come. What
role exactly does the eternal return play here? According to Delellze ,
the eternal retllrn is a power of repetition , which is a truly genetic ,
creative force.
We produce something new only on condition that we repeat - once in
the mode which constitutes the past , and once rnore in the present of
rnetarnorphosis. Moreover , what is produced , the absolutely new itself, is
in turn nothing but repetition , the third repetition , this time by excess , the
repetition of the future as eternal return. (D R 901121)
For Deleuze , repetition is a condition of action and of creation.
He cites as an example the French revolutionaries who identified themselves with Romans , wearing tllnic skirts and venerating
Roman values. However , these resuscitated Romans' were not
sin1ply representations' of the original Romans , but actors themselves that have becorne capable of the reTolutionary act under the
condition of repetition. At this point it might be helpful to briefly
explain Deleuze's distinction between a 111aterial and bare repetition
(rptition materielle et nue) and a spiritual and clothed repetition
(rptition spirituelle et vtue).
Bare repetition is defined as a repetition of the same , that is the
recurrence of identical elements. It is based on the model of representation according to which Being is distributed in detern1inable
forms (genera of Being or categories) , f1 xed determinations (speci f1 c
245
or individual differences) and deterrnined objects (individuals). Thus

at the sight of several red objects , we can speak of the recurrence
of instances of red and subsurne these particular instances under
a comrnon genus , the colour red. A rnaterial or bare repetition is
based on generality or universal concepts that subsllrne particlllars.
hese particlllars are in principle exchangeable: one particular lTlay
be sllbstituted for anothe r. Hence the bare repetition lTlay involve
rnany instances , bllt all of these instances point beyond thernselves
to a cornrnon genlls (the One or the Sarne). Material or bare repetition belongs to the order of laws , as we can see frorn the important
role that bare repetition plays in scientific experin1ents. There the
repetition of particulars is explained as instances of a naturallaw.
By contrast , spiritllal or clothed repetition is against the la w:
against the sirnilar forrn and the eqllivalent content of law' (DR
2/9). Repetition in this second sense never has to do with generality.
It does not deal with particlllars that are equivalent in one way or
another , but operates on non-exchangeable and non-sllbstitutable
singlllarities. We have already corne across this type of repetition ,
when we discussed the Platonic world of sirnlllacra , of false pretenders and rnasks. Deleuze sustains a reversed PlatonislTI' that overtllrns
the Platonic hierarchy of the original or rnodel followed by the copy
followed by the sirnlllacrllrn (cf. DR 126-8/165-8). There is no
original: The rnasks do not hide anything except other rnasks. There
IS no rst terrn which is repeated' (DR 17/28). The clothed repetition
denollnces any principle or law , in favour , as Delellze says , 'of a lTIOre
profound and lTIOre artistic reality' (DR 3/9). It is in art , for example
in theatre , that the deeper reality cornes to the surface. In this sense ,
Delellze says , history is theatre' (DR 91 /1 23) and the French revollltionares are n1rnes or actors. his idea of theatre resonates with a
sirnilar thOllght that Nietzsche presents in The Gay Science , when he
reflects on the problern of the actor':
Falseness with a good conscience; the delight in pretence erupting as a
power that pushes aside , f1 oods , and at times extinguishes one's cocalled
character'; the inner longing for a role and mask , for an appearance
(Schein); an excess of capacities for all kinds of adaptation that can no
longer be satisfied in the service of the nearest , most narrowly construed
utility - perhaps all of this is distinctive not only of the actor?79
Nietzsche sllggests that the actor rnanifests a power , the power of
the false , which is rnore fundamental than being sirnply an expression of his profession. This power carries with it a desire for masks
246
and disguise that can even be found among animals in the disposition of mimicry'. This power is so strong that its eruption threatens
and destroys the so-called character of a living being. It challenges
all the categories and values which are so dear to our scientific
and rational world: the identity of a thing , its determinability , its
lawfulness and truth. lnstead of a will to truth , this obscure power
shows a alseness with a good conscience' , a delight in pretence' ,
an inner longing for a role and rnask , for an appearance [Schein]'.
This parodic' power , which sets free meta1TIorphoses and masks , is
what Deleuze identifies with Nietzschean repetition , that is to say
the thought of the eternal return. his is why he says that Nietzsche
lS deeply theatrica l': Nietzsche brought theatre into philosophy
itself'.80 The philosophical doctrine of the eternal return (i.e. the
return of that which differs , of difference-in-itself) is precisely what
undermines the privilege of identity and the model of representation.
'The eternal return affirms difference , it affirms dissemblance and
disparateness , chance , 1TIultiplicity and becoming' (DR 300 /3 83).
Repelling any identity , the wheel of the eternal return rneans the
death of the one and only God. And as God is dead , this means that
the judge supporting the identity of the subject disappears and so the
subject dissolves. As Deleuze says , the eternal return concerns only
simulacra , it causes only such phantasms to return' (DR 1261165). It
thus becomes clear that the eternal return is n10re than a 'theoretical
representation' (DR 4 1/60) or ethical rule to be made a selchosen
principle of life. Rather it is a positive principle , 'the royal repetition'
(DR 94/125) , that actively creates the superior forms that pass the
test of eternal return.
In his book Nietzsche and the Vicious Circle , Pierre Klossowski
shows how the thought of the eternal return itself enacts this selective and creative power. According to Klossowski's analysis , the
thought of the eternal return jeopardises the subject's identity; it is
an aggression against the apparently limited and closed whole of the
subject. The reason for this is that the thought of the eternal return
demands that 1 re-will myself again innumerable times , but this
demand makes me at the same tin1e fall into incoherence. In relation
to the codes of everyday societ) 1 am a particular identifiable indi.
247
All that remains , then , is for me to re-will myself, no longer as the

outcome of these prior possibilities , no longer as one realization among
thousands , but as a fortuitous moment whose very fortuity implies the
necessity of the integral return of the whole series. 81
1 deactualize my present self in order to will myself in all the other selves
whose entire seres must be passed through. 82
hus
the eternal return does not derrland that 1 return the same as
1 am , once and for all' (this would alnount to a 'bare repetition' in
Deleuzian terms) , but as a variation , a simulacrmn , for an infinite
number of times (this would be a repetition y excess the repetition of the future as eternal return' , DR 901122). The coherence of
the subject is thus jeopardised. Nietzsche h lnself suffered the consequences of the thought of the eternal return: '1 am every name
in history' ,83 'Dionysus and the Crucified'. 84 In Deleuze's reading ,
which coincides with Klossowski's interpretation in this regard , the
thinker , undoubtedly the thinker of the eternal return , is [. . .] the
universal individu (DR 254/327). For Deleuze , the universal indi
vidual or the 'lan without a name' (DR 911122) designates son1eone
who has relinquished the well-defined identity of the subject with
xed boundaries , and afrrned the system of a dissolved self with all
its processes of becoming.
What the self has become equal to is the unequal in itself. ln this manner ,
the 1 which is fractured according to the order of time and the Self which
is divided according to the temporal series correspond and find a common
descendant in the man without name , without famil) without qualities ,
without self or 1, the plebeian' guardian of a secret , the already-Overman
whose scattered members gravitate around the sublime image. (DR
90/1 21)
The universal individual or man without name s thus to be understood as Nietzsche's overman. he ovennan is not another higher
species of man , but a non-identical a dissolved self, which is liberated
from the judgement of God and open to intensive processes of
becoming. Interpretations that regard the overman as an eTolution
ary product , rising higher , as n1an does relative to the worn1 , to SOlne
indeterminate evolutionary height fron1 which he can look back ,
amused , at that from which he carrle' ,85 treat the overman as someone
beyond n1an , a higher species that
248
stone , becoming plant, becorning animal , becoming sta r. 86 Deleuze

takes up this thought and states that the thinker of the eternal return
s laden with stones and dian1onds , plants "and even anirnals" , (DR
254/327).87 We have seen that examples of the overman are the poet
or the artist another suitable example might be the political subject ,
someone who exposes himself to the public and engages in processes
that not only demand a becoming-other , that is an annihilation of
the past self that he or she was , but that also put the existence of a
future self at risk and thus leave the process of becoming open to
success or failure. he ancient rhetorical practice of parrhesia can
serve as an exarnple here: parrhesia can be translated as the telling of
the unvarnished truth' and specifies a type of discourse in which the
speaker commits himself to a free and unbound speech and in doing
this puts hin1self at considerable risk , including the risk of death. 88
The parrhesiast or truth-teller' cannot be defined in terms of a selfauthoring subject , but must be understood as a split subject: through
his words he constitutes himself as the one who speaks freely and
who is willing to pay for it with his life. He forsakes the identity and
securities of his past self, and projects an ideal future self that would
find the approval of his listeners , but his project might just as well
end in failure.
All these examples , the poet or artist and the political subject (e.g.
the parrhesiast) can be seen as instantiations of a dissolved self, or the
overman in Nietzschean ternlS. Thus the overman is a real possibility or even a present reality , if one thinks the thought of the eternal
return and wills oneself through the entire series of all the other
selves , that is affirrns all possibilities of becoming. That is expelled
by the wheel of eternal return and its centrifugal force is only that
which desperately clings to its identity.
In his recent book Deleuze's Philosophy of Time , James Williams
expresses worries about the human- and subject-centred properties
of Deleuze's account of the third synthesis of time' , which seem to be
present in Deleuze's formula that the image of the excessive act , i.e.
the caesura constituting the third synthesis of tin1e , appears too great
for n1e'. He continues that we can avoid is existen t alism 'once
we realise that the image applies to any novel process , for instance ,
when a virus mutates and achieves sonlething "too great for it' or
when pressures on rocks transfonn an organic layr caught between
them into something new'. 89 One reason that he gives for linking the
third syn esis of tin1e with any process of creation and not only with
249
human creativity (of the artist , the thinker or the political subject) is
his fear of a psychological explanation of time , according to which
time is constituted by human imagination. hus Williams argues:
First, the focus on human imagination and particular instances of drama
very uneasily with Deleuze's points about an a priori definition. If
the past or before was determined each time in the imagination , then
we would be dealing with an empirical psychological test; [. . .] Second ,
psychological exanation is avoided at every turn in Difference and
Repetition because it fails to grasp the importance of habit and extramental processes [. . .] Third , Deleuze uses the term image in a technical
manner partly indebted to Bergson. For Deleuze , the notion of image is
not one of a mental image , but rather one of a reductive yet necessary
process of assembly.90
ts
le
agree that Deleuze abandons psychological explanations on

the whole; time , for him is certainly not a synthetic product of the
human faculty of imagination (as Heidegger conceives it in his book
Kant and the Problem of Metaphysics). Neither can the third synthesis of time be exclusively derived frorn elements of Greek n1yth
or drama theory , such as H lderlin's notion of the caesura. For this
reason we took recourse to a n1athematical explanation: t llle as
a static synthesis of discrete elements (past and future n10ments) ,
distributed by a cut (the pure present , or tirne of the event) in a
before and after. It is difcult to see how Willian1s' exalllples of the
mutation of a virus or the transformation of organic layers through
the pressure of rocks help to attain an a priori de nition of time ,
since they describe empirical (though extra-rnental) processes. Let us
consider Williams' last point of critique , narnely that the symbolic
image of the excessive act (the murder of God , for example) is not
sirnply a n1ental lllage. Certainl) the symbolic falls outside the
psychoanalytical , binary model of the real and the 111aginary and
instead constitutes a new orde r. 91 This new order Deleuze will later
call - no longer an order of the syrnbolic (which is still too much
Lacanian) - but the order of the virtual or virtual events. hus we
understand the caesura' or 'cut constituting the third synthesis of
time , as the moment when the virtual event breaks into the chronological and ernpirical order of tme. Nietzsche's death of God and
the dissolution of the self is j ust one such virtual event that beC Ollles
lTlanifest in philosophy , in Nietzsche's life itself and possibly in the
life of any other person capable of becoming equal to the even t.Te
certainly do bestow a Ii existentialist dirnension to the third synthesis
250

of time and we see no reason why we should not , since Deleuze
insists on its disruptive irnpact on the identity of the subject and its
liberating power with regard to all prior possibilities of life. As we
see it , Deleuze's three syntheses of tirne offer an explanation of the
formation of the subject, that is its various habits , practices , agency ,
rnemory and creativity. It is important to emphasise , however , that
the sllbject or rather sllbjective strllctllres are something that emerge ,
transforrn , dissolve and are always renewed throllgh the temporal
syntheses. hllS we shollld not presllppose a prior identity of the
subjec which then ha ppens to become destroyed in the third synthesis of time. Rather , the discollrse on the identity of the sllbject is part
of the dogrnatic Image of thought , a cllltllral machine that prodllces
the illllsion of the identical sllbject and thereby constrains ollr thinking , feeling and acting. If we follow Delellze , we always have to do
so with fractllred sllbjects , which rnore or less cling to the traditional
notions of identity , agency , free will and so on. With Nietzsche's
words , we can distingllish between noble and base ways of being ,
that is modes of being that affirm difference and becoming and those
that are negative , reactive and conservative.
Williams acknowledges that Delellze's third synthesis of time is
deeply indebted to Nietzsche's eternal return and to Klossowski's
reading of it. However , he sees a sharp difference between Delellze
and Klossowski:
Klossowski introduces an element much harder to reconcile with
Deleuze's philosophy of time: reversibility (as opposed to Deleuze's insistence on asymmetry). 50 for Klossowski it is the creative will itself that
frees its acts from the past by willing reversibility - a willing incompatible
with Deleuze's account of time - whereas for Deleuze it is the passing
of the same and eternal return of difference that accomplish [sic] this
freedom without a necessary appeal to will or to affrmation (perhaps in
contrast to Deleuze's work on Nietzsche in Nietzsche and Philosophy).92
In Ollr explanation of Delellze's reading of Nietzsche's eternal retllrn ,
the selective and creative power of the eternal retllrn has been partly
referred to the wiU and the power of affirn1ation. We said that the
eternal retllrn exclllded any instances of half-hearted willing (willing
son1ething only this one tIne) and allowed only that to retllrn which
affirms becoming or becomir-othe r. lt is true that for Klossowski
this becon1ing means becoming all prior possibilities of being , in his
words 'to will myself in all the other selves whose en 'e series must
he passed through'. 93 But this 'return of the whole series' certainly
251
RANSCENDENTALIDEAS
does not mean a reversal of the order of time , a going back in tirne ,
but time as an eternally decentred circle (a vicious circle') , which as
Deleuze has shown is cornpatible with a purely formal , linear tirne
constituted by a cut. 'The form of time is there only for the revelation of the formless in the eternal return' (DR 911122). For us , the
rnain difference between Klossowski and Deleuze lies not so much
in the difference between the principle of reversibility or asymmetry
with regard to their conception of t lne. We rather see one major
difference in their respective notion of sign' and its function as a
transcendental condition of thought. Let us begin with Klossowski's
notion of 'sign'.
ln his essay , in which Klossowski spells out the in1port that the
thought of the eternal return had for Nietzsche and in fact that it
would have for anyone taking this thought seriously he also reconstructs the mental state in which Nietzsche conceived the thought of
the eternal return. 94 According to this reconstruction , thought occurs
like an event , it bursts into the mind as something from outside , a
force or intensity which leaves the thinker no choice. As Nietzsche
himself says: Everything is in the highest degree involuntary' a
ren1ark which resonates in lnany of Deleuze's texts. 95 he thinker is
exposed to sornething that exercises an almost unbearable pressure
and tension. According to Nietzsche's description in Ecce H mo
a thought flashes up like lightning , with necessity , unfalteringly formed - 1
have never had any choice. An ecstasy whose tremendous tension some
times discharges itself in a flood of tears , while one's steps now involuntarily rush along, now involuntarily lag; a complete being outside of oneself
with the distinct consciousness of a multitude of subtle shudders and trick
les down to one's toes; a depth of happiness in which the most painful and
gloomy things appear , not as an antithesis , but as conditioned , demanded ,
as a necessary colour within such a superfluity of light. 96
According to Klossowski's analysis , it seems that Nietzsche himself
experienced the thought of the eternal return as such an event. In a
letter that Nietzsche sent to Peter Gast from Sils-Maria in August
1881 , he gives a brief account of the moment when he conceived this
thought , and this account n1atches exactly the description in Ecce
Homo. 97 Klossowski takes this letter as a cause to reflect upon the
rnysterious role of intensity as the materal condition of thinking.
First of all , Klossowski describes Nietzsch s experience as a f1 uctuation of intensity that occurred all of a sudden in the midst of a hohe
Stimmung , i.e. an elevated tonality of the sou l. In order for it to be
252
cOlnmlul cable and to acquire sense , the flow of intensity must turn
back on itself and must take itself as an object. A sort of intentionality' , an aiming of intensity at itself, is needed for the creation of
sense: For this , the intensity rnust divide , separate from itself, and
come back togethe r. '98 The flow of intensity , the interruption of flow ,
and a new afflux of intensity make up what Klossowski calls a sign'.
According to Klossowski's analysis , Nietzsche's experience when he
first conceived the thought of the eternal return is to be characterised
as such a rise and fall of intensity , i.e. the encounter with a slgn\
Klossowski attaches a great in1portance to the prior experience of the
slgn treating it as a transcendental but material condition that first
and foremost generates and constitutes thought. As Klossowski says:
It is thanks to this sign, which nonetheless is nothing but an alwaysvariable trace of a f1 uctuation , that we constitute ourselves as thin ing
that a thought as such occurs to us even though we are never quite sure
if it is not others who are thinking and continue to think in us. But what
is this other that forms the outside in relation to this inside we believe
ourselves to be? Everything is led back to a single discourse , namely , to
f1 uctuations of intensity that correspond to the thought of everyone and
no one. 99
Klossowski relates the 'sign' back to a kind of universal f1 0w of intensity , a universal f1 0w of thought that is prior to any segrrlentation into
particular thinking subjects. Although we do not literally find this
idea of a universal f1 0w of intensity or thought in Nietzsche he does
write in one of his notebooks:
A thought . . . comes up in me - where from? How? 1 simply don't know.
It comes , independently of my will , usually surrounded and obscured by a
mass of feelings , desires , aversions , and also other thoughts . . . One pulls
it [the thought] out of this mass , cleans it off, sets it on its feet , and then
sees how it stands and how it walks - all of this in an astonishing resto
and yet without any sense of hurry. Just who does all this - 1 have no idea,
and 1 am surely more a spectator than originator of this process. IOO
In this quotation Nietzsche confirms that there is no '1 think' who
is the author of thought , but that thought happens to the thinker
independently from his wiU or wish.his is not to say that a par
ticular thought or concept is given to him ready-made. As Nietzsche
is careful to emphasise: one has to extract the concept or thought
from a rnass of feelings and other thoughts; one has to 'clean' it and
ensure that it can stand upright. In this sense , thinking is both an
involuntary and in1personal adventure and an act of creation.
253
At first glance it seems that Deleuze con1pletely aligns with

Klossowski (andNietzsche). In Diflerence and Repetition Deleuze
describes how the encounter with son1ething violent and formless ,
sOlnething beyond the grasp of cOlnmon sense , creates an intensive
impression which triggers a )rced movement' of thollght. This
thought is not 'n1y thought' in the sense of a well-detern1ined posses
sion; it is rather a violent and involuntary adventure which challenges
the categories of inner and outer , '1' and the Other. He points to the
dangers involved in thinking , in particular philosophical thought
and the shattering effect it has on the identity of the sllbject. He tells
us that philosophical thought ollght not to be
related to a substantial , completed and well-constituted subject, such as
the Cartesian Cogito; thought is , rather , one of those terrible movements
which can be sustained only under the conditions of a larval subject.
These systems admit only such subjects as these , since they alone can
undertake the forced movement by becoming the patient of the dynamisms which express it. Even the philosopher is a larval subject of his
own system. (DR 118 /1 56)101
Thollght is thus a orced movement' that carries the subject
to the borders of the liveable , to the point which only larval subjects can support. The 'larval' or en1bryonic subject' in Delellze
designates those living beings able to tolerate extreme forces and
spatio-temporal dynamisms under which any skeletal systen1 Tould
break. 102 It should be noted , however , that not only philosophical
thOllght involves sllch terrible movement'. On the contrary , Deleuze
rnakes it clear that this might also be true of science and art. 103
To conclude , there are certainly similarities between Klossowski's
and Deleuze's account of the encollnter with a sign' and the
'forced movement of thought but we believe that Deleuze is
rnuch rnore specific than Klossowski in describing the nature of the
'sign Although he does use sign' in tern1S of a sensuous , material
impression or complex of intensive forces that sets off the forced
n10vement of thought, he rather prefers the notion of differential
Ideas-problems. It is problerns or questions that force us to think ,
to liquefy rigid concepts , fixed representations or opinions and
also affct our mode of existence. Thus , while Klossowski does not
further exarnine the sign' and seems to be satis ed with relatir it
back to a kind of llniversal flow of intensity , Deleuze takes greater
care to analyse this mysterioussign'. As we have seen in Chapter 3 ,
Deleuze elaborates a whole dialectic of Ideas.
also goes beyond
254

the dimension of conditions of thought and considers extra-rnental ,
differential processes of Ideas. Exarnples of differential Ideas that we
already discussed concerned atomism as a physical Idea , the organism as a biological Idea and the socio-economic Idea (with recourse
to Marx). In this sense , James Williarns is certainly right if he points
out that a kind of existentialist , human or subject-centred reading of
Deleuze's syntheses of time and processes of differential production
is reductive , and we need to keep this in mind when dealing with
Deleuze. However , the aim of this book has been to narrow down
a Deleuzian notion of transcendental conditions that act as a sufficient reason for the production of the new , that is a new thought , a
new work , a new act , even a new subject or universal people yet to
come.Nietzsche's eternal return , at least in Deleuze's understanding
as a positive , selective and creative power of repetition , allows the
generation of complete novelties. As we have seen , the eternal return
is allied with the will to power understood as a differential structure
of forces , or with Deleuze's words as differential Ideas-problems
forcing themselves upon those who are ready to welcome the event
of thought. Both the will to power and the eternal return or agaln ln
Deleuze's words the powers of difference and repetition , signify an
absence of ground or a 'universal ungrounding' from which emerges
a world of siInulacra or dissolved selves that make up the Deleuzian
'subject' of thought and creation.
Notes
1. The French word moi may be translated into English either as ego' or
as self'. Here, the passive , empirical nature of the Kantian subject (/e
moi passive) is rendered as the passive self' in contrast to the subject's
active , transcendental nature , which is referred to as '1' (j e). The trans
lator Paul Patton maintains the translation self' for moi in all contexts
except those where it is explicitly a question of psychoanalysis, in
which case he has used ego' (cf. Translator's Preface to DR , p. xiii).
2. Cf. Deleuze's Lecture Course on Kant , 14 March 1978. See also
Deleuze, On four poetic formulas which might summarize the
Kantian philosophy' , KCP vii-xiii.
3. Act 1, scene v: 'The time is out of joint: 0 cursed spite, / That ever
1 was born to set it right' (in Shakespeare , Hamlet , p. 74). It should
be noted that Hamlet explicitly says the time is out of jOiI referring to a particular time , i.e. the time through which he is living. The
French translation 'le temps est hors de ses gonds' [l iterally:time is
off its hinges is somewhat ambiguous because of the different way
255
the definite article is used in French and English. The French phrase
could be translated back into English as either time is out of join t'
or the time is out of joint'. The English translation of Differe l1 ce al1 d
Repetitio l1 renders the French phrase into ime is out of joint' (D R
88/119) , thereby indicating the metaphysical import of time in general
that Deleuze reads into this formula. According to Deleuze , time in
general has become demented time [temps affol]' (DR 88/119). This
means that it has lost its balance , its groundedness , its stability: time
has gone crazy. This meaning is preserved in the French phrase le
temps sort de ses gonds' that Deleuze uses in the Lecture Course on
Kant from 14 March 1978. (We are indebted to Nick Midgley for this
clarication.)
4. See Wright, Cosmology il1 A l1 tiquity , Chapters Three and Eigh t.

5. Aristotle , Aristotle's Physics: Boo s III al1 d IV , and Aristotle , On the
heave l1 s 1 & I1; hereinafter the Latin De Caelo will stand for 011 the
heave l1 s.
6. Deleuze refers to Borges' short story Death and the Compass' (1 942).
Having walked into the trap of the murderer Scharlach , detective
L nnrot tells him: " ln your labyrinth there are three lines too many
[. . .]. 1 know of one Greek labyrinth which is a single straight lin
Along that line so many philosophers have lost themselves that a mere
detective might well do so , too." [...] "The next time 1 kill you ,"
replied Scharlach , "1 promise you that labyrinth , consisting of a single
line which is invisible [sic] and unceasing'" (Borges , Labyril1 ths , pp.
867). The English translation wrongly renders the Spanish ndivis
ible' in the phrase una sola lnea recta y que es indivisible , incesante'
into English as invisible\
7. KCP viii-ix. See Rimbaud , Lettres du Voya l1 t, pp. 113 and 135.
8. Cited in KCP ix. ln fact , Deleuze contracts here two quotations by
Rimbaud from different letters. See Rimbaud's letter to Georges
lzambard from 13 May 1871 , p. 113 , and his letter to Paul Demeny
from 15 May 1871 , p. 135.
9. See also CPR B 156: Through inner sense we intuit ourselves only as
we are internally affected by our selues, i.e. , as far as inner intuition is
concerned we cognize our own subject only as appearance but not in
accordance with what it is in itself.'
10. Leibniz says in his letter to Clarke of 25 February 1716: 1 hold space
to be something merely relative , as time is; [. . .] 1 hold it to be an
order of coe
256

12. Tilliams Gilles Deleuze's Difference and Repetition , p. 15.
13. See , for instance , Hume , Enquiries concerning Human Understanding ,
p. 75: This connexion , therefore , which we feel in the mind , this
customary transition of the imagination from one object to its usual
attendant , is the sentiment or impression from which we form the idea
of power or necessary connexion.'
14. Hume , Enquiries concerning Human Understanding , p. 75.
Corresponding statements can also be found in Hume , A Treatise of
Human Natu1 pp. 93 and 156.
15. Hume rather speaks of a kind of af nity ( some bond of union' , some
associating quality') that prevails among simple ideas in the mind and
exer Cl ses a gentle force' on the faculty of imagination; see Hume ,
A Treatise of Human Natu1 p. 10. Deleuze's interpretation of a
contractile power' of the mind reminds one rather of Bergson , for
whom the movements of contraction and expansion constitute different levels of our mental life. Bergson , in fact , criticises the theory of
associationism for not being able to explain the mysterious attractions
between individu independent ideas; see Matter and Memory , pp.
21217.
16. 'For even the corn in the elds grows upon a superstitious basis as to
its own existence , and only turns the earth and moisture into wheat
through the conceit of its own ability to do so , without which faith
it were powerless' (in Butler , Life and Habit , p. 82). Cited in DR
75 /1 02.
17. C f. Bryant: Transcendental empiricism reconciles the two halves of
aesthetics (the theory of the sensible and the theory of beauty) insofar
as it is able to explain how the being of the sensible allows for a genesis
of experience (thus treating experience as an aesthetic production in
the sense of artistic production) , which in turns creates a domain of
experience of the given (the aesthetic in the sense of sensible receptivity).' See Bryant, Difference and Givenness , p. 64.
18. Proust , A la recherche du temps perdu , vo l. III , p. 872. Deleuze takes
up this formula of Proust in his second book on cinema - see CIT
79 /1 10.
19. For Bergson's two schemata of the cone , see Matter and Memory , pp.
127-8 and 21 1.
20. Bryant tries to steer a course between critical and speculative phil
osophy , emphasising that Deleuze cannot be easily described as
either a speculative r a critical philosopher' (Bryant , Difference and
Giuenness , p.176). Yet , his
257
22. Ibid. , p. 175.

23. Ibid. , pp. 173 ar 179.
24. Ibid. , p. 180: Sensation is , in its essence , extended and localized; it is
a source of movement; pure memory , being inextensive and power
less , does not in any degree share the nature of sensation.' See also
p.176.
25. See also B 59/54 and CIT 76-7/105-6. In reconstructing Bergson's
three paradoxes of time , Deleuze refers to Bergson , Matter and
Memory , ch. II I.
26. Bergson , Matter and Memory , p. 217.
27. Ibid. , pp. 128-9.
28. Ibid. , p. 171: 1henever we are trying to recover a recollection , to
call up some period of our history , we become conscious of an act
sui generis by which we detach ourselves from the present in order to
replace ourselves , frst in the past in general , then in a certain region
of the past - a work of adjustment , something like the focussing of a
ca lera.'
29. It should be noted that Deleuze extends this idea that art is able to
create pure intensive impressions (affects and percepts) or blocs of sensation (sensation in itself) to other arts besides literature. As Deleuze
makes clear in his book on Francis Bacon and his Cinema books ,
painting and flm can equally capture intensive forces or the pure past
ma crystal image [image-cristal]'. See CIT 79/109.
30. In the same way , Sauvagnargues argues in her book Deleuze:
L'Empirme transcendantal , p. 98: Chronos , the present that passes ,
recaptures the actuality of the frst synthesis , while Aion in a disjunctive manner joins together the pure past and the achronological
becoming of the second and of the third synthesis. Aion comprises
the virtual dimensions of the past and of the future that insist in the
present and elude actuality' (my translation , D. V.).
3 1.he presentation of the schema can be found in Deleuze , CIT 285/1 09,
note 23/22.
32. Sauvagnargues , Deleuze: L'Empirisme transcendantal, pp. 97-8 (my
translation , D. V.).
33. H lderliAnmerkungen zum Oedipus' , pp. 729-36. For an English
translation see pfau (ed.) , Friedrich Hlderlin: Essays and Letters on
Theory , pp.101-8.
34. For Deleuze's reference to derlin and Beaufret , see DR 315/1 18 ,
footnote 10/1.
35. H lderlin Anmerkungen zum Oedipus pp. 735-6 (my translation ,
D.V.).
36. Ibid. , p. 736 (my translation , D. V.).
37. Beaufret , Hlderlin et Sophocle , p. 2 1.
38. Ibid. , pp. 19 2 1.
258
Tnle and the Splt Subject

39. It should be mentioned that the three parts of limitation , transgression
and atonement , which Delellze analyses , match neither with the acts
of Aeschyllls' Agamemnon , nor with the sllrviving triology Oresteia
(the parts of which are Agamemnon' , The Libation Bearers' and 'The
Euminides'). We have to llnderstand Deleuze's strllctural analysis as
independent of any actual strllcture of the play. What Deleuze tries to
illllstrate is a certain conception of tragic time which is cyclical , just
like the Greeks' conception of astronomical time.
40. Beallfret, Hlderlin et Sophocle , p. 32 llnd Deleuze , LK II , pp. 3-4.
4 1. H lderlin Anmerkllngen Zllm Oedipus' , p. 736 (cf. Pfau , Friedrich
Hlderli p. 108; translation modified , D. V.).
42. Hlderlin's Oedipus raises the following problem: if Sophoclean
tragedy already involves a thinking of the pure and empty form of
time , of a time of abandonment by the gods , then the Kantian revolutio i.e. the breaking with ancient cyclical time and the unrolling of
time as a straight line , had argllably already been performed in Greece
two thousand years before Kant. his view fits well with Deleuze's
thought that Ideas are virtual and that they become actualised in different places at different times. We should therefore read Deleuze's
claim that Kant revollltionised the ancient cyclical model of time not
simply in chronological terrr i.e. in terms of a historical development, but as the actualisation of a virtual Idea , which is to say that the
movement is not going from one actual term to another , but from the
virtual to the actua l. (We owe this suggestion to Nick Midgley.)
43. Beaufr Hlderlin et Sophocle , p. 28.
44. Ibid. , pp. 18-19.
45. Nietzsche , The G Science pp.11920.
46. Ibid. , p. 120.
47. Ibid.
48. For instance , Deleuze refers to the different possibilities of ecom
ing in 'Literature and Li fe' , CC 1-6. See also ATP , ch. 10: '1730:
Becoming-Intense , Becoming-Animal , Becoming-imperceptible' , and
WP 169/1 61: Everything is vision , becoming. We become llniverses.
Becoming animal , plant , moleclllar , becoming zero.'
49. See also CPR A 144/B 183 and A 182/B 224-5.
50. C f. Schallb: Deleuze therefore sticks to a concept of form that still
appears somewhat like a seclllarised Platonic Idea , thinking of form
as a nelltral , "vanishing" and timeless mediator that neither adds anything to its content n<
259
52.
53.
54.
55.
56.
57.
58.
59.
60.
6 1.
62.
in a second sense is said to be characteristic for the linear conception

of time , time as pure and empty form. According to Deleuze , limit
designates an internallimit , which he describes as that towards which
something tends'. This denition matches with the mathematical
definition of limit during the early geometrical phase of infinitesimal
analysis , which depended upon theorems stating the continuous and
infinite approximation of magnitudes toward a limiting value (LK 1I,
pp.8-9).
Cf. Bunn, Developments in the Foundations of 1athematics pp.
220-6.
1hile in his Lecture Course on Kant from 21 March 1978 , Deleuze
refers to the mathematical notion of limit' as that towards which
something tends' , in Difference and Repetiton Deleuze defines limit'
as a genuine cut [coupure]' in the sense of Dedekind' (DR 172/223).
Dedekind , Continuity and Irrational N umbers' , p. 2.
Ibid. , pp. 10-1 1.
Bunn , Developments in the Foundations . Mathematics p.223.
Dedekind , 'Continuity and Irrational N umbers' , p. 9.
Boyer, The History of the Calculus , p. 292.
Summarising Dedekind's idea of cuts' , Deleuze c1 aims that the cut
designates the irrational numbers which differ in kind from the terms
of the series of rational numbers' (DR 172/224). This is not completely correct as the cut defines a real number that can correspond to
either an irrational quantity or a rationalnumber. But the important
point is that numbers defined in this way do indeed constitute a new
kind of numbe r.
It should be noted that both the rational numbers and the real numbers
are infinite , ordered and dense number systems. The property of denseness means that between any two numbers there is at least one other
numbe r. Denseness is not continuity, as Bolzano mistakenly believed.
he property of continuity , which is attributed to the system of real
numbers (but not to that of rational numbers) , is precisely defined by
Dedekind's method of cuts. Cf. Kline , Mathematical Thought, p. 985.
As Bunn puts it , a densely ordered system is complete (co l1 tinuous)
in Dedekind's sense if every cut in the system is produced by exactly
one element of the system , that is , if there is an element of the system
which is either the maximum of the lower section or the I11 lfllmUm
of the upper section' (Bunn , Developments in the Foundations of
Mathematics' , p. 222).
Ibi
260
65.
66.
67.
68.
69.
really succeeded in giving a purely ideal , arithmetic de nition of continuity. In fact , he is criticised by Russell and Wittgenstein for failing
to get away from the geometrical image of the number line. For more
on Russell's and Wittgenstein's criticism , see Widder , Reflections on
Time and Politics , ch. 2: Point , Line , Curve' , pp. 22-33.
Williams rightly insists that Deleuze's model for the third synthesis of
time cannot be the ordered line of time , and that the division produced
by the caesura does not equal a thin logical point. In his own words:
he caesura is an event and has a depth to it. It is not instantaneous
but rather must be considered with its effect on the points before and
after it. This is why the caesura implies a drama: it divides time such
that a drama is required to encompass this division. This event-like
and dramatic division is in contrast with the thin logical point and
set account of the line of time where an arbitrary point is taken on
a line and every point before it is defined as before in time and every
point after as after in time' (Williams , Gilles Deleuze's Philosophy ol
'ime p. 91). It is true that Deleuze's third synthesis of time cannot
be reduced to the number line and the cut to a thin logical poin t. But
it should come as no surprise that Deleuze makes use of Dedekind's
idea of a cut' and static synthesis' without following him in everything that he says. Deleuze certainly over-interprets the notion of cut ,
insofar as he will equate it with the irruption of the virtual event (the
unthought , the inexplicable , the incommensurable) and the fracture in
the subject.
Williams , Gilles Deleuze's Philosophy ol Time , p. 89.
1 am indebted to Anne Sauvagnargues for pointing out this specific
Deleuzian technique of cutting theories together'.
The exdusive right of concept creation secures a function for philosophy , but it does not give it any pre-eminence or privilege since
there are other ways of thinking and creating , other modes of ideation
that , like scientific thought , do not have to pass through concepts'
(WP 8/13-14). See also WP 127/121 and 66/64.
C f. Deleuze: 1 belong to a generation , one of the last generations , that
was more or less bludgeoned to death with the history of philosophy.
The history of philosophy plays a patently repressive role in philosophy , it's philosophy's own version of the Oedipus complex:
261
70.
7 1.
72.
73.
74.
75.
76.
77.
78.
79.
80.
8 1.
82.
83.
84.
wrote a n l1 mber of books on some canonical philosophical a l1 thors ,

b l1 t he always l1 nderstood this commentary as creative: he bro l1 ght to
light new aspects and extracted new concepts , not least by importing
heterogeneo l1 s elements or points of views into the original tex t.
Sl1 ffce here the following references to Klossowski by Dele l1 ze: D R
66-7/92 , 901/ 122 951127 , 312/81-2 note 19/2 , 33 1/313 note 16/1
and in LS Appendix III 28030 1/325-50.
Dele l1 ze refers to Henry Miller's book The Time of the Assassns.
Dele l1 ze , Concl l1 sions on the Will to Power and the Eternal Ret l1 rn' ,
DI 124/173. Dele l1 ze gave this paper at the end of a colloq l1 il1 m on
Nietzsche which he organised and which took place in the Royaumont
Abbey , 4-8 J111y 1964. Cf. also DR 126/164-5 and 24 1/311.
See DR 6114 , 241 3/311-13 and 299/382.
Nietzsche , Thus spa e Zarathustra , p. 167.
Ibid. , p. 234.
Nietzsche: 'My doctrine teaches: live in s l1 ch a way that you m l1 st
desire to live again , this is your d l1 ty - you willlive again in any case!
He for whom striving proc l1 res the highest feeling , let him strive; he for
whom repose prOC l1 res the highest feeling , let him rest; he for whom
belonging , following , and obeying prOC l1 res the highest feeling , let him
obey. Provided that he becomes aware of iVhat procures the highest
feeling , and that he shrinks back from nothing. Eternity depends llpon
it' (in KSA , vo l. 9 , p. 505 , 11[163] , Spring-.Fall1881).
Delellze Conclllsions on the Will to Power' , DI 1251174. See also DR
41160.
Dele l1 ze , 'Conclllsions on the Will to Power' , DI 125/175 (translation modifed , D. V.). In order to avoid the misleading connotations
of sllperman' , we prefer to translate Jbermensch' as overman'.
We thus agree with the Nietzsche scholar and translator Walter
Kaufmann who justifed the use of overman' both in terms of
breaking with the connotations of superman' and of preserving the
etymological connections with other words in Nietzsche that contain
the prefx ber-' (s l1 ch as berwinden' which translates best as
overcoming') .
Nietzsche , The Gay Science , pp. 225-6.
Deleuze , 'Concl l1 sions on the Will to Power' , DI 127/177.
Klossowski , Nietzsche and the Vicious Circle , p. 58.
Ibid. , p. 57.
See Nietzsche's letter to Jacob B l1rckhar 6 January 1889 , in
Schlechta (ed.) , Werke , p. 135 1.
See Nietzsche's letters to Peter Gast , Georg Brande
262
Tine and the Split Subject
86.
Klossows
ki
dsir p.223.
87. Deleuze makes this remark in the context of discussing Leibniz and
the notion of implication , that is enveloping intensities , fields of indi
viduation and individual differences. However , he links these thoughts
explicitly to the concept of eternal return , the universal individual and
the system of the dissolved self.
88. 1ichel Foucault analysed the practice of parrhesia in six lectures
given at the University of California at Berkeley in 1983 as part of
his seminar entitled Discourse and Truth\The complete text compiled from tape-recordings is published under the title Fearless Speech
(2001).
89. Willia Deleuze's Philosophy of Time , p. 92.
90. Ibid. , pp. 92-3.
9 1. Deleuze's rejection of the binary model of the real and the imaginary is clearly expressed in his early essay 'How Do re Recognize
Structuralism?' (Dl1711240).
92. Williams , Deleuze's Philosophy ofTime , pp. 184-5.
93. Klossowski , Nietzsche and the Vicious Circle , p. 57.
94. Klossowski , 'The Experience of the Eternal Return' , in Nietzsche and
the Vicious Circle , pp. 55-73.
95. Nietzsche , Ecce Homo , p. 73. In Proust and Signs , Deleuze calls the
movement of thought an 'adventure of the involuntaq (PS 9511878) and in Difference and Repetition Deleuze states: There is only
mroluntary thought , aroused but constrained within thought , and all
the more absolutely necessary for being born , illegitimately , of fortuitousness in the world' (DR 139/181).
96. Nietzsche , Ecce Homo , p. 73.
97. Letter to Peter Gast , 14 August 1881 , in Schlechta (ed.) , Wer e pp.
1172-4. An English translation can be found in Klossowski , Nietzsche
and the Vicious Circle , pp. 55 6.
98. Klossowski , Nietzsche and the Vicious Circle , p. 60.
99. Ibid. , pp. 62-3.
100. Nietzsche , Writings from the Late Noteboo p.34.
10 1. See D198/136: And thought itself, considered as a dynamism proper
to the philosophical system , is perhaps in its turn one of these terrifying movements that are irreconcilable with a form q ualified , and
composed subject , such as the subject of the cogito in representation.'
102. Cf. DR 1181155-6 and 250/322. See also D1971136.
103. C f. WP 8113-14 , 66/64 and 127/121. Although this view that philosophical thought is just one specific mode of thinking which is distinct
but nt superior to science and art is explicitly stated in De
263
seen no contradiction in extending this view to his early tex t. In the

Preface to the English edition of Differe l1 Ce al1 d Repetitio l1, Deleuze
says that 'philosophy obviously cannot claim the least superiority [to
arts and sciences ], but also creates and expounds its own concepts
only in relation to what it can grasp of scientific functions and artistic
constructions' (DR xvi).
264
Conclusio
In this study we have examined Deleuze's critique of the implicit

and tacit Image of thought that emerged in the history of philosophy
and that subjects the act of thinking to the postulates of good sense ,
common sense and recognition , thereby separating thought frorrl its
vital and genetic conditions. Calling this classic Image of thought
into question , Deleuze sets out to determine the nature of thought
anew and to relate it back to those elements which account for the
genesis of the act of thinking in thought. He specifies the relation
between thought and its conditions as 'transcendental' , thereby
making use of a concept which enjoys a long established tradition
and which was first introduced by Kant to bring philosophy onto the
secure course of a science' (CPR B vii) in search of truth. However
the original Kantian concept of the transcendental clearly adheres
to the classic Image of thought that takes recognition as its mode l.
Given that Deleuze always maintained that 'philosophy does not
consist in knowing and is not inspired by truth' (WP 82/80) , how
can we Inake sense of Deleuze's appropriation of the notion of the
transcendental? How can we understand his apparently paradoxical
relation to Kant , whorn he describes both as an enemy'l and as the
analogue of a great explorer' who discovered 'the prodigious domain
of the transcendenta l' (DR 135/176)? Deleuze hirrlself indicates
the solution to this paradox in the following description of how he
believes the history of philosophy should be approached:
Criticism implies new concepts (of the thing criticized) just as much as
the most positive creation. [...] Nothing positive is done , nothing at
all , in the domains of either criticism or history , when we are content to
brandish ready-made old concepts like skeletons intended to intimidate
any creation , without seeing that the ancient philosophers from whom we
borrow them were already doing what we would like to prevent modern
philosophers from doing: they were creating their concepts , and they
were not happy just to clean and scrape bones like the critic and historian
of our time. Even the history of philosophy is completely without interest if it does not undertake to awaken a dormant concept and to play it
265
again on a new stage , even if this comes at the price of turning it against
itsel f. (WP 83/80-1)
Deleuze's encounter with Kant nlust be understood as a resumption of critical philosophy and a radicalisation of critique that bears
on the presuppositions of Kant's original project itself. he aim of
Deleuze's 'deconstructive technique' is to construct a concept of the
transcendental that paves the way for a new model of thought which
in his later uvre co-authored with Guattari will carry the narne of
the rhizome.
In this book we have analysed the characteristics of the Deleuzian
concept of the transcendental , determined its sources of inspiration
and brought to light its novelty in comparison to the traditional
lneaning of the term. We have shown that apart from his retention
of certain aspects of the transcendental developed by Kant , Deleuze
adds new features which can be traced back to the in f1 uence of
both post-Kantian philosophers and other major thinkers (Leibniz ,
Maimon , Nietzsche , Bergson , Proust , Lautman , etc.). We have found
that the following general features are distinctive for Deleuze's
notion of the transcendental:
1. The emergence of thought s necessitated through the encounter with something exterior (a transcendental sign or Idea) that
exercises a force upon thought in order to constrain it to think.
Thought is an involuntary adventure and not a naturally given
process furthered by good will and guided through lnethod.
2. The transcendental designates a genetic and differential principle
by which thought arises.
3. It is essentially a plastic principle: this rneans that the transcendental condition is not larger than what it conditions. In other
words , it is deterrrl ned at the sanle time that it detennines the
conditioned.
4. The transcendental condition does ot resemble what it conditions , which is to say that it is not produced retroactively in the
lnage of the ernpirical and elevated to a transcendental status.
5. hought's encounter with the transcendental happens in the
confrontation with genetic and differential Ideas which neither
belong to reason (the Kantian faculty of Ideas) nor to the understanding (as Maimon's differential Ideas of the understanding)
but to a virtual intersubjecve uncoJzscious. 2
6. Ideas are problematic objective structures. That is to say , they
do not indicate a telnporary subjective lack of knowledge and
266
Co clusion
cannot be reduced to simple questions whose answers already

pre-exist and which disappear in the solution.
7. he theory of Ideas is correlated with a theory of the faculties.
More concretel) the encollnter with Ideas-problerns forces their
corresponding faculties to transgress their limits and to merge
into a disharmonious transcendent exercise (freed from the
constraints of empirical recognition).
8. Ideas-problems partake in a dialectic ol Jdeas , whence follows
their statlls as simultaneollsly transcendent (i.e. irredllcible to
the cases of sollltion) and immanent (that is Ideas-problems are
incarnated in sensible natllI history social life , arts , sciences ,
philosophy and so on).
9. Ideas are defined as virtual n1ultiplicities consisting of differen
tial relations shaped by the distriblltion of singlllar points. They
are governed by processes of dillerent /c iation (that is internal
processes of differentiation and processes of actllalisation and
specification).
10. Differential Ideas-problems relate to a split subject, in Delellze's
words the system of a dissolved self' which is constitllted by
thousands of passive syntheses and fractllred by the empty form
of time.
In order to olltline this new Deleuzian conception of transcendental philosophy , we have lTIainly drawn frOlTI Delellze's early
texts , in particular Dirence and Repetition. However , a seminal
discllssion of the transcendental viewpoint retllrns in his last published essa y Immanence: A Life' (1995). here Deleuze makes it
plain that his transcendental philosophy has to be understood as a
philosophy of immanence: the transcendental field which has been
defined as a genetic or potential n1iliell swarming with differential
Ideas - or as Deleuze prefers to say in the Logic ol5ense: poplllated
with impersonal and pre-individllal singlllarities (c f. L5 103 /1 25)
- becomes a 'plane of imnlanence\Deleuze emphasises the modification in the n1eaning of the term 'transcendent that his philosophy
of immanence requires , stating that Kant failed in two ways: 'The
transcendental is entirely denatured , for it [. . .] simply redollbles the
empirical [. . .], and immanence is distorted , for it [. . .] finds itself
enclosed in the transcendent. '3 In other words , he accuses Kant of
tracing the transcendental fron1 the errlpirical - a criticism that we
already know from Dilference and Repetition and of enclosing
all possible experience in the confines of a transendental subjec t.
267
TRANSCENDENALIDEAS
Deleuze argues that Kant rnade the field of possible experience

imrnanent to a transcendental subject , which itself is excluded from
the empirical spatio-temporal world and cons 1'itutes its transcendent
ground.
Deleuze's demand for imrnanence poses a great challenge to
Kant but also to post-Kan 1'ianism that builds on the transcendental
subject and turns it into the absolute subject of Gennan Idealism. For
Deleuze , a philosophy of immanence must not distort lnlnanence by
making it immanent to something. 'Absolute imn1anence is in itself:
it is not in son1ething , to something; it does not depend on an object
or belong to a sul :t. The plane of lnrnanence is a11 there is and ,
as Deleuze and Guattari suggest in What Is Philosophy? , we must
understand it as containing both being and thought: it 'presents two
sides to us , extension and thought , or rather its two powers , power
of being and power of thinking' (WP 48/50). he individuation of
subjects and objects is only a secondary , derived phenon1enon , a
product of immanence. Consciousness becomes a fact only when a
subj:t is produced at the same time as its object , both being outside
the field and appearing as "transcendents".
Deleuze rejects any rst term - such as a transcendental consciousness , a universal subject 0 1' an object which is able to contain
lmmanence and along with it the idea of foundation. Deleuze's
plane of imrnanence , although he calls it something Absolute and
refers to it as THE plane of immanence (WP 59/S9) , does not ful i
a grounding function. THE plane of immanence is not a universal
or abstract totality , but a multiplicity: it is interleaved' (WP 50/S1) ,
that is it has different layers which are sOlnetimes separated and
sometimes joined togethe r. Furthermore HE plane of iIIllnanence is
itself virtual , that is pure variation' (WP 39/41) or pure bec Olning
ln the essay 'Imn1anence: A Life' he describes it as a n10verrlent that
neither begins nor ends' ,6 as an in1personal , indefinite , and imrrlanent life carrying with it the events or singularities that are rrlerely
actualized in subjects and objects 7
Although Deleuze in his lateuvre surely introduces new concepts
(such as 'pure bec Oll1 ing' , 'even t' 8 or 'in1personal life') , we can
1' etrieve some IIlotives that he had already elaborated in Differe C
and Repetitio lZ. 'e have seen 1' hat his transcendental ernplnClsrn
abandons the ultn1a 1'e ground of the transcendental subjec t.e
Deleuzan transcendental designates a vi 1' tual realrn of differential Ideas in the depth .of the ernpirical world that di erent/ciate
268
Conclusion
themselves according to different speeds and rhythms and that force

themselves through lninds generating the act of thinking in thought.
We can see here the overriding importance to Deleuze of the two
notions of difference and repetition. heyare rst principles' in the
sense of providing not an ultimate ground , but rather a groundlessness' or 'universal ungrounding' (DR 9 1/ 123). hey function as
principles of genesis , that is as real conditions of experience , thought ,
consciousness , subjects and objects alike.
Even though the aspect of a philosophy of in1manence is relatively unelaborated in Ditference and Repetition , we have seen
that this early work already harbours the seeds of Deleuze's future
philosophical development. Thus it can be argued that the plane of
immanence is a revival of the transcendental field , which is conceived
as a sub-representative , unconscious , virtual and immanent depth. 9
Therefore a thorough understanding of the meaning and use of the
notion of the transcendental is all the more important as the notion
underpins his later work , either explicitly or in disguise. This is not
to say that there is a strict continuity in Deleuze's uvre; rather , it
rnust be understood as a discontinuous and interrupted line of solutions or solution curves that revolve around problems and subsets of
problems iVhich share the proper nan1e Deleuze.
Notes
1. Deleuze describes his book on Kant as a book about an enemy that
tries to show how his system works , its various cogs the tribunal of
Reason , the legitimate exercise of the faculties (our subjection to these
made all the more hypocritical by our being characterized as legislators)'
(N 6/1 4 15).
2. It should be noted that the term intersubjective' by no means signies
a collective unconscious common to a number of integral subjects. The
point of view of the integral subject must be replaced by the split subject
which is dissolved in a series of selves (cf. DR 124/1 62).
3. Deleuze , Immanence: A Life' , p. 27.
4. Ibid. , p. 26.
5. Ibid.
6. Ibid.
7. Ibid. , p. 29.
8. It might have been noticed that throughout this book we have not discussed the Deleuzian concept of the event. his is so because it plays a
minor role in Difference and Repetition and is fully developed only in
The Logic of SensNeverth ss already in Difference alld Repetiti n
269
the concept of events is mentioned and it appears that we have to under

stand Deleuzian Ideas-problems in terms of events rather than essences
(cf. DR 187/242-3). According to Deleuze ,problems are of the order
of events' (DR 188/244) and he distinguishes real events on the level of
the engendered solutions , and ideal events embedded in the conditions of
the problem' (DR 189/244).
9. 'The transcendental field is defined by a plane of immanence , and the
plane of immanence by a life' (in Deleuze , Immanence: A Life' , p. 28).
270
Bibliography
Allison , Henry E. , Kant's Transcendental Idealism: An Interpretation

and Defense (New Have CT and London: Yale University Press ,
1983).
Allison , Henry E. , '1s the Critique of Judgment "Post-Critical"?' , in
Sally S. Sedgwick (ed.) , The Reception of Kant's Critical Philosophy
(Cambridge: Cambridge University Press , 2000) , pp. 7892.
Alunni , Charles, Continental Genealogies: Mathematical Confrontations
in Albert Lautman and Gaston Bachelard' , in Simon Duffy (ed.) , Virtual
Mathematics: The Logic of Difference (Manchester: Clinamen Press ,
2006) , pp. 65 99.
Aristotle , Topics , in ]onathan Barnes (ed.) , The Complete Works of
Aristotle, vo l. 1 (Princeton: Princeton University Press , 1984) , pp.
167-277.
Aristotle , Aristotle's Physics: Boo s III and IV, trans. Edward Hussey
(Oxford: Clarendon Press , 1983).
Aristotle , On the Heavens 1 & II , trans. and ed. Stuart Leggatt (Warminster:
Aris & Phillips , 1995).
Atlas , Samuel , From Critical to Speculative Idealism: The Philosophy of
Solomon Maimon (The Hague: Martinus Nijhoff, 1964).
Badiou , Alain , 'Mathematics and Philosophy' , in Simon Duffy (ed.) , Virtual
Mathematics: The Logic of Difference (Manchester: Clinamen Press ,
2006) , pp. 12-30.
Beaufret, ]ean , Hlderlin et Sophocle (Brionne: Grard Monfort, 1983).
Beiser , Frederick C. he Fate of Reason: German Philosophy from Kant to
Fichte (Cambridge 1A and London: Harvard University Press , 1987).
Bergman , Samuel Hugo , The Philosophy of Solomon Maimon , trans. from
the Hebrew by Noah]. Jacobs (J erusalem: Magnes Press , 1967).
Bergson , Henri , Matter and Memory , trans. Nancy Margaret Paul and
W. Scott Palmer (Mineola , NY: Dover Publications , [1896] 2004).
Bergson , Henri , 'The Possible and the Real' , in The Creative Mind: An
Introductio l1 to Metaphysics by Henri Bergson , trans. Mabelle L. Andison
(New York: Citadel Press , 1992) pp. 91-106.
Bogu Ronald Deleuze and Guattari (London and New York: Routledge ,
1989).
Borges , Jorge Luis Death and the Compass' ('La muerte y la brujula'
271
TRANSCENDENAL
IDEAS
1942) , in Donald A. Yates and ]ames E. Irby (eds) , Labyrinths: Selected

Stories and Other Writings (New York: New Directions Publishing ,
1962) , pp. 76-87.
Bos , H.]. M , Differentials, Higher-Order Differentials and the Derivative
in the Leibnizian Calculus' , Archive for History ol Exact Sciences , vo l.
14, no. 1 (March 1974) , pp. 1-90.
Bowden , Sean , The Pri rity ol Events: Deleuze's Logic of Sense (Edinburgh:
Edinburgh University Press , 2011).
Boyer, Carl The History ol the Calculus and its Conceptual Development
(New York: Dover , [1949] 1959).
Bryant, Levi Differe ce and Givenness: Deleuze's Transcendental
Empiricism and the Ontology oflmmanence (Evanston , IL: Northwestern
University Press , 2008).
Bunn , Robert , Developments in the Foundations of Mathematics , 1870
1910' , in Ivor Grattan-Guinness (ed.) , From the Calculus to Set Theory~
1630-1919: An Introductory History (London: Duckworth , 1980) ,
pp.220-55.
Butler , Samuel , Lile and Habit (London: ]onathan Cape , [1878] 1910).
Cassirer , Ernst , Das Erkenntnisproblem in der Philosophie und
Wissenscha der neueren Zeit , vo l. 3 (Berlin: Bruno Cassirer , 1920).
Cauchy , Augustin L. uvres Se r. 2 , vols 3 and 4 (Paris: Gauthier-Villars ,
1897-9).
Choat , Simon , Deleuze, Marx and the Politicisation of Philosophy' ,
Deleuze Studies , vo l. 3 , supplement (2009) , pp. 8-28.
Couturat , Louis , On Leibniz's Metaphysics' , in Harry G. Frankfurt (ed.) ,
Leibniz: A Collectio l1 ol Critical Essays , trans. R. Allison Ryan (London:
Anchor Books , [1902] 1972) , pp. 19-4 5.
Dedekind , Richard Continuity and Ir rational Numbers' , in Essays 0 the
Theory ol Numbers , trans. T ooster W oodruff Beman (Chicago: Open
Court , 1901) , pp. 1-30.
Deleuze , Gilles , Empiricism and Subjectivity: A l1 Essay on Hume's Theory
l Human Nature , trans. Constantin V. Boundas (New York: Columbia
University Press , [1953] 1991).
Deleuze , Gilles , Bergson's Conception of Difference' ( La conception de la
diffrence chez Bergson' , 1956) , in David Lapoujade (ed.) , Desert Islands
and Other Texts (1953-1974 trans. Michael Taormina (New York:
Semiotext(e) , 2004) , pp. 32-5 1.
Deleuze , Gilles , Lecture Course on Chapter Three of Bergson's Creative
Euolution\Presentation at the Ecole Suprieure de Saint-Cloud , 14
March 1960 , trans. Bryn Loban , SubStance , issue 114 (vo l. 36 , no. 3)
(2007) , pp. 72 90.
Deleuze , G
272
Bblography
trans. Hugh Tomlinson and Barbara Habberjam (London: Continuum,

[1963] 2008).
Deleuze , Gilles ,The Idea of Genesis in Kant's Esthetics' ('L'ide de gense
dans l'esthtique de Kant' , 1963) , in David Lapoujade (ed.) , Desert
Islands and Other Texts (1 953-74) , trans. Michael Taormina (New
York: Semiotext(e) , 2004) , pp. 56-7 1.
Deleuze , Gilles , Conclusions on the Will to Power and the Eternal Return'
( Conclusions: Sur la volont de puissance et l'trnel retour' , 1964) , in
David Lapoujade (ed.) , Desert Islands and Other Texts (1 953-74 trans.
1ichael Taormina (New York: Semiotext(e) , 2004) , pp. 117-27.
Deleuze , Gilles , Proust and Signs: 'h e Complete Text , trans. Richard
Howard (Minneapolis: University of Minnesota Press , 2000 [Marcel
Proust et les signes , 1964, revised edn 1970 and 1976 retitled Proust et
les sgnes]).
Deleuze , Gilles , Nietzsche' [1965] , in Pure Immanence: Essays on A Life ,
trans. Anne Boyman (New York: Zone Books , 2001) , pp. 53-102.
Deleuze , Gilles , Bergsonism , trans. Hugh Tomlinson and Barbara
Habberjam (New York: Zone Books , [1966] 1988).
Deleuze , Gilles he Method of Dramatization' ( La mthode de dramatisation' , 1967) , in David Lapoujade (ed.) , Desert Islands and Other Texts
(1953-74 trans. Michael Taormina (New York: Semiotext( 2004)
pp.94-103.
Deleuze , Gilles HowDo le Recognize Structuralism?' ( A quoi reconnaton le structuralisme?' , 1967, first published 1972) , in David Lapoujade
(ed.) , Desert Islands and Other Texts (1 953-74 trans. Michael Taormina
(New York: Semiotext( 2004) , pp. 170-93.
Deleuze , Gilles , 'On Nietzsche and the Image of Thought' [1968 ], in David
Lapoujade (ed.) , Desert Islands and Other Texts (1 953-74 trans.
1ichael Taormina (New York: Semiotext( 2004) , pp. 135-42.
Deleuze , Gilles , Diflerence and Repettio trans. Paul Patton (New York:
Columbia University Press , [1968] 1994).
Deleuze , Gilles , The Logic of Sense , ed. Constantin V. Boundas and trans.
Mark Lester with Charles Stivale (New York: Columbia University Press ,
[1969] 1990).
Deleuze , Gilles , Anti-Oelipus: Capitalism and Schizophrenia , co-authored
with Flix Guattari , trans. Robert Hurley , Mark Seem and Helen R. Lane
(London: Athlone Press , [1972] 2000).
Deleuze , Gilles , Dalogues , co-authored with Claire Parnet , trans. Hugh
Tomlinson and Barbara Habberjam (London: Athl
273
Deleuze , Gilles , Lecture Courses on Leibniz at the University of Vincennes ,

15 April1980 , 22 April1980 , 29 April1980 , 6 May 1980 , 20 May 1980 ,
trans. Charles J.
S ti
va
le dltt
p://www rebdeleuze.com
1/
php/somr
htmb (l ast accessed 25 August 2012) ..
Deleuze , Gilles , A Thousand Plateaus: Capitalism and Schizophrenia ,
co-authored with Flix Guattari , trans. Brian Massumi (Minneapolis:
University of Minnesota Press , [1980] 1987).
Deleuze , Gilles , Lecture Course on Spinoza. Presentation at the University of
Vincennes , 17 February 1981 , trans. Timothy S. Murphy , <http://www.
webdeleuze.com/php/sommaire.htmb (l ast accessed 25 August 2012).
Deleuze , Gilles , Cinema 2: The Time-1mage , trans. Hugh omlinson and
Robert Galeta (London: Continuum , [1985] 1989).
Deleuze , Gilles , Foucault , trans. San Hand (Minneapolis: University of
1innesota Press , [1986] 1988).
Deleuze , Gilles 'he Fold: Leibniz and the Baroque , trans. Tom Conley
(Minneapolis: University of Minnesota Press , [1988] 1993).
Deleuze , Gilles , Bartleby; or The Formula' ( Bartleby ou la formule' ,
1989) , in Essays Critical and Clinical , trans. Daniel W. Smith and
Michael A. Greco (Minneapolis: University of Minnesota Press , 1997) ,
pp. 68-90.
Deleuze , Gilles , Negotiations 1972-1990, trans. Martin Joughin (New
York: Columbia University Press , [1990] 1995).
Deleuze , Gilles , What 1s Philosophy? , co-authored with F ix Guattari ,
trans. Hugh Tomlinson and Graham Burchell (New York: Columbia
University Press , [1991] 1994).
Deleuze , Gilles , Literature and Life' ('La littrature et la vie' , 1993) , in
Essays Critical and Clinical , trans. Daniel W. Smith and Michael A.
Greco (Minneapolis: University of linnesota Press , 1997) , pp. 1-6.
Deleuze , Gilles , 'To have done with judgment' ('Pour en finir avec le jugement' , 1993) , in Essays Critical and Clinical , trans. Daniel W. Smith and
1ichael A. Greco (Minneapolis: University of Minnesota Press , 1997) ,
pp. 126 35.
Deleuze , Gilles , He stuttered' ( Bgaya-t-il . . .', 1993) , in Essays Critical
and Clinical , trans. Daniel W. Smith and Michael A. Greco (Minneapolis:
University of 1innesota Press , 1997) , pp. 107-14.
Deleuze , Gilles , Immanence: A Life' ('L'Immanence: une vie' , 1995) , in
Pure 1mmanence: Essays on A Lile , trans. Anne Boyman (New York
274
Bibliography
Descartes , Ren , Discourse on the Method , in The Philosophical Writings

of Descartes , trans. John Cottingham, Robert Stoothoff and Dugald
Murdoch , vo l. 1 (Cambridge: Cambridge University Press , [1637] 1985) ,
pp. 111-5 1.
Dosse , Franois , Gilles Deleuze et Flix Guattari: Biographie croise (Paris:
La Dcouverte , 2007).
Duffy , Simon, Albert Lautman' , in Graham Jones and Jon Roffe (eds) ,
Deleuze's Philosophical Lineage (Edinburgh: Edinburgh University Press ,
2009) , pp. 356-79.
Duffy , Simon , The Logic of Expression: Quality, Quantity and Intensity in
Spinoza , Hegel and Deleuze (Aldershot: Ashgate , 2006).
Duffy , Simon , 'The Mathematics of Deleuze's Differential Logic and
Metaphysics' , in Simon Duffy (ed.) , Virtual Mathematics: The Logic of
Diflerence (Manchester: Clinamen Press , 2006) , pp. 118-4 4.
Dur Robin , Problems in the Relation between 1aths and Philosophy' ,
in Simon Duffy (ed.) , Virtual Mathematics: The Logic of Difference
(1anchester: Clinamen Press , 2006) , pp. 169-86.
Engstler , Achim , Untersuchungen zum Idealismus Salomon Maimons
(Stuttgart: Frommann-Holzboog , 1990).
Evens , Aden , Math Anxiety' , Angela i vo l. 5 , no. 3 (2000) , pp. 105 15.
Fichte , Johann Gottlieb , Gesamtausgabe. Briefwechsel 1793-95, vo l. 111 ,
2 , ed. Reinhard Lth and Hans Jacob (Stuttgart: Frommann-Holzboog
Verlag , 1970).
Foucault , Michel , Fearless Speech , ed. Joseph Pearson (Los Angeles:
Semiotext( 2001).
Frege , Gottlob, On Sense and Reference' , in A. T. Moore (ed.) ,

Meaning and Reference (Oxford: Oxford University Press , [1892] 1993) ,
pp.23 -4 2.
Franks , Paul All or Nothing: Systematicity , Transcendental Arguments
andS epticism in German Idealism (Cambridge , MA: Harvard University
Press , 2005).
Freudenthal , Gideon, A Philosopher Between Two Cultures' , in
G. Freudenthal (ed.) , Salomon Maimon: Rational Dogmatist, Empirical
s eptic (Dordrecht: Kluwer , 2003) , pp. 1-17.
Grosz , Elizabeth , The Nick of Time: Politics , Evolutio and the Untimely
(Durham , NC and London: Duke University Press , 2004).
Guroult , Martial , La Philosophie transcendantale de Salomon Mai
(Paris: Librairie Flix Alcan , 1929).
Guroult 1artial L'Evolution et la structure de la doctrine de la science
chez Fichte , vo l. 1 (Paris: Les Belles Lettres , 1930).
Heidegger , Martin , Kant and the Problem of Metap cs (Blo
275
Friedric:h , Remarks on "0fipus"' in Friedrich Hlderlin: Essays

and Letters on Theory , trans. and ed. Thomas pfau (Albany , NY: SUNY,
1988) , pp. 101-8 [ Anmerkungen zum Oedipus' , in Friedric:h Beiner und
Joc:hen Sc:hmidt (eds) , Hlderli Wer und Briefe (Frankfurt am Main:
Insel Verlag , 1969) , pp. 729-36].
Hughes , Joe , Deleuze's 'Difference and Repetition': A Reader's Guide
(London and New York: Continuum, 2009).
Hume , David , A Treatise of Human Nature , ed. L. A. Selby-Bigge (Oxford:
Clarendon Press , [1739 -4 0] 1958).
Hume , David , Enquiries concerning Human Understanding and concerning
the Principles of Morals , ed. L. A. Selby-Bigge (Oxford: Clarendon Press ,
[1748 , 1751] 1975).
Jones , Graham, Solomon Maimon' , in Graham Jones and Jon Roffe (eds) ,
Deleuze's Philosophical Lineage (Edinburgh: Edinburgh University Press
2009) , pp. 10429.
Kant , lmmanuel , Critique of Pure Reason , trans. and ed. Paul Guyer and
Allen W. Wood (Cambridge: Cambridge University Press , 1998).
Kant , Immanuel , The Critique of Practical Reason and Other Writings in
Moral Philosophy , trans. and ed. Lewis White Bec:k (Chic:ago: University
of Ch ic: ago Press , 1949).
Kant , Immanuel Critique of the Power of}udgment , ed. Paul Guyer , trans.
Paul Guyer and Eric: Matthews (Cambridge: Cambridge University Press ,
2000).
Kant , Immanuel Kan t's Briefwechsel (1744-1788 ed. Kniglic:he
Preuisc:he Akademie der Tissensc:haften vo l. 10 (Berlin: Reimer ,
1922); selec:tions translated in Arnulf Zweig (ed.) , Kan t's Philosophical
Correspondence , 1759-1799 (Chicago: University of Chicago Press ,
1967).
Kant , Immanuel , Briefiuechsel, ed. Rudolf Malter (Hamburg: Felix Meiner ,
1986).
Kant , Immanuel , lmmanuel Kant's Werke , vo l. X (l etters from and to Kant) ,
ed. Ernst Cassirer (Berlin: Bruno Cassirer , 1923).
Kerslake , Christian , Immanence and the Vertigo of Philosophy: From Kant
to Deleuze (Edinburgh: Edinburgh University Press , 2009).
Kline , Morris , Mathematical Thought om Ancient to Modern Times (New
York and Oxford: Oxford University Press , 1972).
Klossowski , Pierre ,
dsir (Par.iS: Gallir
mard 1963) pp. 185-228.
Klossowski , Pierre he Experienc:e of the Eternal Return' ('Oubli et
anamnse dans l' exprience vcue de l'ternel retour du mme' , 1964) ,
in Nietzs he and the Vicious Circle , trans. Da
H lderlin
276
Bibliography
Ideas and the Physical Real , trans. Simon Duffy (London and New York:
Continuum , 2010) , pp. 27 30.
Lautman , Albe Essay on the Notions of Structure and Existence in
Mathematics' ( Essai sur les notions de structure et d'existence en math
matiques' , 1938) , in Mathematics , Ideas and the Physical Rea! , trans.
Simon Duffy (London and New York: Continuum , 2010) , pp. 45-83.
Lautman , Albert, New Research on the Dialectical Structure of
Mathematics' ('Nouvelles recherches sur la structure dialectique des
mathmatilues' 1939) , in Mathematics , Ideas and the Physical Real ,
trans. Simon Duffy (London and New York: Continuum, 2010) , pp.
197-219.
Lebrun , Grard , 'Le transcendantal et son image' , in Gilles Deleuze: Une vie
philosophique: Rencontres Internationales Rio de Janeiro So Paulo ,
June 10 14, 1996 (Paris: Institut Synthlabo , 1998).
Leibniz , Gottfried Wilhelm , Discourse on Metaphysics , in The Rationalists ,
trans. George Montgomery (New York: Dolphin Books , 1960) , pp.
405-53.
Leibniz , Gottfried Wilhelm , The Leibniz-Arnauld Correspondence , ed.
H. T. Mason (Manchester: Manchester University Press , 1967).
Leibniz , Gottfried 1ilhelm On Freedom' , in G. H. R. Parkinson (ed.) ,
Philosophical Writings , trans. Mary Morris and G. H. R. Parkinson
(London: Dent & Sons , [c.1689] 1973).
Leibniz , Gottfried Wilhelm , The Leibniz Clarke Correspondence , ed. H. G.
Alexander (Manchester: Manchester University Press , 1956).
Leigh , James A. , Deleuze , Nietzsche and the Eternal Return' , Philosophy
Today , Fall (1978) , pp. 20623.
Lord , Beth , Kant and Spinozism: Transcendental Idealism and Immanence
fl'om Jacobi to Deleuze (New York: Palgrave 1acmillan 2011).
Maimon , Salomon , Essay on Transcendental Philosophy , trans. Nick
Midgley , Henry Somers-Hall , Alistair Welchman and Merten Reglitz
(New York and London: Continuum , [1790] 2010).
Maimon , Salomon , Philosophisches Wrterbuch , in Valerio Verra (ed.) ,
Gesammelte Werke , vo l. 3 (Hildesheim: Georg Olms , [1791] 1970).
1vlaimon , Salomon , Solomon Maimon: An Autobiography , trans. Clark
Murray (Chicago: University of Ill inois Press , [1792] 2001).
4limon Salomon , Streifereien im Gebiete der Philosophe , in Valerio Verra
(ed.) , Gesammelte Wer e vo l. 4 (Hildesheim: Georg Olms , [1793] 1970).
Maimon , Salomon , Versuch einer neuen Logi oder Theore des Den el
277
Mller-Lauter , Wolfgang , "Der Wille zur Macht" als Buch der "Krisis" philosophischer Nietzsche-Interpretation' , in E. Behler et a l. (eds) , Nietzsche
Studien: lnternationales Jahrbuch fr die Nietzsche-Forschung , vo l. 24
(Berlin: Walter de Gruyter , 1995) , pp. 223-60.
Nietzsche , Friedrich , Schopenhauer as Educator' , in Untimely Meditations ,
trans. R. J. Hollingdale (Cambridge: Cambridge University Press , [1874]
1983) , pp. 125-95.
Nietzsche , Friedrich, On the Uses and Disadvantages of History for Life' ,
in Untimely Meditations , trans. R. J. Hollingdale (Cambridge: Cambridge
University Press , [1874] 1983) , pp. 57-123.
Nietzsche , Friedrich , Human , All Too Human , in The Nietzsche Reader, eds
Keith Ansell Pearson and Duncan Large (Malden , MA: Wiley-Blackwell ,
[1878-80] 2006) , pp. 161 9 1.
Nietzsche , Friedrich , The Gay Science , ed. Bernard Williams , trans. Josefine
Nauckhoff (Cambridge: Cambridge University Press , [1882] 2001).
Nietzsche , Friedrich , Thus sa e Zarathustra , ed. Manuel Komroff and
trans. Thomas Common (New York: Tudor , [18835] 1936).
Nietzsche , Friedrich , Beyond Good and Evil , in Basic Writings of Nietzsche ,
trans. and ed. Walter Kaufmann (New York: Modern Library , [1886]
1968) , pp. 181 -4 35.
Nietzsche , Friedrich , On the Gel1 logy of Morals , in Basic Writings of
Nietzsche , trans. and ed. Walter Kaufmann (New York: Modern Library ,
[1887] 1968) , pp. 439-599.
Nietzsche , Friedrich , Ecce Homo , trans. R. J. Hollingdale (London: Penguin
Books , [1888] 1979).
Nietzsc Friedrich , Twilight of the Idols , in Twilight of the Idols/The
Anti-Christ, trans. R. J. Hollingdale (London: Penguin Books [1889]
1968) , pp. 29-122.
Nietzsche , Friedrich , Nachgelasse l1 e Fragme l1 te: 1880-1882 (unpublished
notes , Spring-Fall 1881) , in Giorgio Colli and Mazzino 10ntinari (ed
Kritische Studie l1 ausgabe (KSA) , vo l. 9 (Munich: dtv; Berlin and New
York: De Gruyter , 1980).
Nietzsche , Friedrich , Writings from the Late Notebooks , ed. Rdiger
Bittner , trans. Kate Sturge (Cambridge: Cambridge University Press ,
2003).
Nietzsche , Friedrich , Brie (1861 1889) in Karl Schlechta (ed.) , Werke ,
vo l. 3 (Munich: Hanser , 1969) , pp. 927-1352.
Nietzsche , Friedrich , The Will to Power , trans. Walter Kaufmann and R. J.
Hollingdale (New York: Vintage Books , 1968).
Patton , Paul , Anti-Platonism and Art' , in Const
278
Bibliography
Patton , Paul , 'The World Seen From Within: Deleuze and the Philosophy of
Events' , Theory & Event , vo l. 1 , no. 1 (1997).
Patton , Paul , 'Mobile Concepts , Metaphor , and the Problem of Referentiality
in Deleuze and Guattari' , in 1aria Magraroni and E e Yiannopoulpou
(eds) , Metaphoricity and the Politics of Mobility (Amsterdam and New
York: Rodopi Press , 2006) , pp. 27-4 6.
Patton , Paul , Deleuzian Concepts: Philosophy, Colonization , Politics
(Stanford: Stanford University Press , 2010).
Plato 'h e Sophist & The Statesma eds R. Klibansky and E. Anscombe ,
trans. A. E. aylor (London: Dawson of Pall Ma Il, 1971).
Plato , Republic , trans. Robin Waterfield (Oxford: Oxford University Press ,
1993).
Plato , The Timais ofPlato , ed. R. D. Archer-Hind (New York: Arno Press ,
1973 [reprint of 1888 edition]).
Plotnitsky, Arkady , 'Manifolds: On the Concept of Space in Riemann
and Deleuze' , in Simon Duffy (ed.) , Virtual Mathematics: The Logic o{
Dirence (Manchester: Clinamen Press , 2006) , pp. 187-208.
Plotnitsky , Arkady, Bernhard Riemann's Conceptual 1athematics and
the Idea of Space' , Congurations vo l. 17, no. 1 (Winter 2009) , pp.
105-30.
Proust , Marcel , A la Recherche du Temps Perdu , eds Pierre Clarac and
Andr Ferr (Paris: Gallimard Biblioth que de la Pliade' , 1954).
Read , J ason , Fetish Is Always Actual , Revolution is Always Virtual: From
Noology to Noopolitics' , Deleuze Studies , vo l. 3 , supplement (2009) , pp.
78-10 1.
Riemann , Bernhard, On the Hypotheses which lie at the Foundation of
Geometry' , in William B. Ewald (ed.) , From Kant to Hilbert: A Source
B in the F ndations of Mathenics 2 vols (Oxford: Oxford
University Press , [1854] 1996) , pp. 652 6 1.
Rimbaud , Arthur , Lettres du Voyant , ed. Grard Schaeffer (Geneva: Droz
and Paris: 1inard 1975).
R li Marc , Gilles Deleuze: Philosophie des transzendentalen Empirismus
(Vienna: uria & Kant , 2003).
Roubach 1ichael 'Salomon Maimon's Philosophy and It s Place in the
Enlightenment. Wandering in the Land of Difference' , in G. Freudenthal
(ed.) , Salomon Maimon: Rational Dgmatist Empirical Skeptic
(Dordrecht: Kluwer , 2003) , pp. 80-8.
Rsell Bertrand , A Critical Exposition o{ the Philosophy of Leibniz
(London: Routledge , 1992).
Salanskis , Jean-Michel Mathematics Metaphysics , Philosophy' , in Simon
Duffy (ed.) , Virtual Mathematics: The Logic of DiJ
279

Schaub 1irjam
Gilles Deleuze im
underland:
Zeit- als
Ereignisphilosophie (1unich: lilhelm Fink , 2003).
Shakespeare , William , Hamlet (German/English) , ed. Manfred Pfister md
trans. Frank Gnther (Cadolzburg: Ars vivendi , 2008).
Smith, Daniel W. , Deleuze's Theory of Sensation: Overcoming the Kantian
Duality' , in Paul Patton (ed.) , Deleuze: A Critical Reader (Oxford:
Blackwell 1996) , pp. 29 56.
Smith , Daniel W. ,Deleuze , Hegel , and the Post- Kantian radition'
Philosophy Today (2001) , pp. 126-38.
Smith , Daniel W. , Deleuze , Kant , and the Theory of Immanent Ideas' ,
in Constantin V. Boundas (ed.) , Deleuze and Philosophy (Edinburgh:
Edinburgh University Press , 2006) , pp. 43-61.
Smith , Daniel W. , Axiomatics and Problematics as Two Modes of
Formalisation: Deleuze's Epistemology of Mathematics' , in Simon
Duffy (ed.) , Virtual Mathel tics: 'h e Logic of Diff'erence (Manchester:
Clinamen Press , 2006) , pp. 145 68.
Smith , Daniel W. , 'Genesis and Difference: Deleuze , Maimon , and the
Post-Kantian Reading of Leibniz' , in Niamh McDonnell and Sjoerd van
Tuinen (eds) , Deleuze and the Fold: A Critical Reader (London: Palgrave
Macmillan , 2009) , pp. 132 54.
Somers. Hall Henry , Hegel and Deleuze on the Metaphysical Interpretation
of the Calculus' , Contin tal Phlosophy Review , vo l. 42 , no. 4 (March
2010) , pp. 555-72.
Spinoza , Benedictus de , The Letters , in Michael L. Morgan (ed.) , Complete
1orks trans. Samuel Shirley (Indianapolis: Hackett , 2002) , pp. 755-959.
Weierstrass , Karl , Mathematische Werke , 7 vols (Berlin: Mayer & Mller ,
1894-1927).
Widder , Nathan , Reflections on Time and Politics (University Park:
Pennsylvania State University Press , 2008).
Williams , James , Gilles Deleuze's Difference and Repetition: A Critical
lntroduction and Guide (Edinburgh: Edinburgh University Press , 2003).
Williams , James , Gilles Deleuze's Philosophy of Time: A Critical
lntroduction and Guide (Edinburgh: Edinburgh University Press , 2011).
Wright , Maureen R. , Cosmology in A tiquity (London: Routledge , 1995).
Zammito , John H. , The Genesis of Kant's Critique of Judgme l1 t (Chicago:
University of Chicago Press , 1992).
Zourabichvili , Franois Deleuze: une philosophie de l'vnemen t', in
F. Zourabichvili , A. Sauvagnargues and P. Marrati (eds) , La Philosophie
de Deleuze (Paris: Presses Univer
280
Index
Actllalisarion , 8, 14, 128 , 191 ,

195 , 196 , 197, 198 , 207n ,
208n, 259n , 267; see also
Differenciation
Allison , Henry , 133n, 148 , 202n ,
203n
Alllnni , Charles , 205n
Appearance , 1, 2, 3, 17n, 74 , 78 , 79 ,
83 , 216 , 217, 256n
Apprenticeship , 27 , 28 , 61 , 63 , 64 ,
65 , 149
Aristotle , 367 38-4 1 57
6970 87 , 146 , 152 , 179 , 180 ,
187, 213-14 , 215
Atlas , Samllel, 92 , 138n
Badioll, Alain , 201
Beallfret, Jean , 231 , 232 , 233
Becoming, 68n , 86-7, 89 , 128 , 149,
209n , 228 , 231 , 236 , 242 , 243 ,
Caesllra ,
Cllt
Calcullls , 57, 60 , 84 , 85 , 90 , 93 ,
102 , 103-7, 110-11 , 118 , 121 ,
122-5 , 13511, 140n, 145 , 156 ,
180, 186 , 187, 191 , 199200 ,
202 , 237
Cassirer, Ernsr, 139n
Common sense , 10, 21 , 22 , 25 ,
32 4, 35 , 41 , 43 , 62 , 129
160-1 , 164, 169 , 178 , 180, 198 ,
254 , 265
Aesrhetic common sense , 33 , 160 ,
161 , 164 5, 174, 178
Conceprllal persona(e) , 46 , 47, 70n ,
81 , 130n
Condition(s) (transcendenra l), 1-4,
9, 12-14, 15r~21 24 25 26
28 , 29 , 44 , 47 , 48 , 49-54 , 56 ,
247 248-9 251 258 259n
5~59 60 61 64 65 67~68n
268
Bergman, Samllel HllgO , 111 , 112,
137n
Bergson, Henri , 5, 8, 1511, 57, 70n ,
75 , 91 , 126 , 127, 185 , 190,
201 , 20711 , 223-6 , 229-30 , 250 ,
257 266
Borges , Jorge Lllis , 214 , 241 ,

256n
Boscovich , Roger Joseph , 82-3
Boye~Carl 104, 239
Bryant, Levi , 78, 17n, 93 ,
257n
Bllnn Rober~238 239 260n
Blltler Samllel
219 , 221 , 257n
122 212 231 233-~240
241 249 250 261n;seea~o
71 , 74-5 , 79 , 87, 88 , 89 , 90 , 92 ,
95 , 98 , 113-14, 116, 11~ 147,
149 , 150, 152-4, 156 , 157, 187,
189 , 193 , 207n , 210 , 216 , 217
222-3 , 224 , 225 , 228 , 233 , 243 ,
245 252-5 265 266 26~
270
Consciollsness of represenration ,
111 , 112 , 116 , 117, 118
COlltllrat, LOllis , 59 , 60
Cririqlle , 1-2, 3, 19 , 22-3 , 36 ,
47 , 50 , 53 , 66n , 67n , 74-5 ,
76 , 79 , 84 , 87, 118-19 , 129n ,
159 , 210 , 212 , 261n , 265 ,
266
281
Cut, 121 , 212 , 229 , 231 , 234 ,

237-4 1, 250 , 252 , 260n , 261n;
see also Caesura
Cuvier, George, 196-7
Dedekind , Richard , 231 , 237-4 1,
260n , 261n
Deduction, 1, 24 , 97, 134n, 148,
150-4, 165 , 169 , 170 , 173 , 174,
202n
Descartes, Ren , 3, 31 , 33 , 36 , 45-7 ,
120, 216-17
Dialectic(s) , 14, 57, 76 , 81 , 142,
145 , 179-82, 184, 185 , 186 ,
191 , 196 , 199-200, 254 , 267;
see also Ideas: dialectical Ideas
Difference, 2, 7, 13 , 20-1 , 38 , 41 ,
42 , 43 , 53 , 64 , 65 , 66 , 76 , 82 ,
83 , 84 , 85 , 86 , 93 , 99 , 125 ,
127, 128 , 132n, 135n, 179,
181 , 182 , 191 , 193 , 240 , 243 ,
247 , 251 , 255 , 269
Concept of difference , 21 , 37, 38 ,
41 , 42 , 93
Conceptual difference , 21 , 37, 94
Difference in degree , 127, 225
Difrence in kind , 119 , 124 , 127,
128 , 193 , 225
External difference , 3, 4
Generic and specifc difference , 21 ,
36 , 39 -4 2, 69n , 245
Individual difference , 39 , 42 , 246
Infnitely small difference , 105-6
InternaVintrinsic difference , 4, 7,
15 16n 3~93 216 234 236
Differenciation, 194-6, 267; see also

Actualisa tion
Differential(s) , 7, 14, 56 , 61 , 66 , 75 ,
84-5 , 93 , 102 , 103-12 , 113 ,
114 , 115 , 116 , 117, 118-20 ,
121 , 122 , 124, 128 , 129 ,
135-6n , 138n, 139n, 140n ,
179 , 187, 192 , 193 , 194 , 199 ,
219 , 223; see also Relation:
differential relation(s)
Differentiation, 83 , 106, 115 , 122 ,

128 , 140~ 193 , 194, 208n , 267
Discordant accord , 157, 178-9; see
also Para -sense
Duffy , Simon, 12 , 140n, 183 , 184,
185 , 186 , 188 , 206n
Encounter, 21 , 25 , 27, 61 , 63 , 65 ,
143 -4 178-9 199 20 242
254, 266, 267

Ens reale , 108, 115
Eternal return, 4, 12, 14-15, 86.7,
210 , 231 , 2425, 247-9 , 251-3 ,
255
Event, 16n, 82 , 175 , 181 , 205n ,
229 , 234 , 235 , 241 , 242 , 250 ,
252 , 261n , 268 , 269 70n
Formal intuition , 3, 51 , 98-101; see

also Pure intuition
Fractured 1 see split subject
Frege, Gottl. ol 489 50
Freud , Sigmund, 120, 172
Freudenthal, Gideon , 92
Genesis , 2, 5, 6, 7, 8, 9, 12 , 14, 20 ,
25 , 27 , 35 , 48 , 54 , 745, 84 , 87,
88 , 90 , 92 , 93 , 102 , 113 , 117
120 , 125 , 127, 142 , 150, 154,
156 , 158 , 165 , 168 , 169 , 173 ,
177, 178 , 184-5 , 192 , 194, 195 ,
207r~208n 210 257n 265
269
Genius , 156 , 158 , 165 , 173 ,
174-8
Good sense , 32 , 334, 37, 41 , 43 ,
129 , 178 , 265
Guattari , Flix, 5, 6, 29 , 30 , 66n ,
190, 191 , 198 , 199 , 208n , 241 ,
242 , 266 , 268
Guroult, Ivlartial , 116-17, 135n,
138n , 139n
Hegel , G. W. 10, 23 , 24 , 36 , 42 ,
76 , 92 , 110 , 179 , 181-2
282
lndex
Heidegger , Martin , 6, 24 , 47 , 182 ,
205n , 211 , 250
Heterogenesis , 241-2
H lderlin Friedrich , 9, 212 , 231-3 ,
241 , 250
Hughes , Joe , 207 8n
Hume , David , 7, 15n, 44 , 76 , 96 ,
97, 219-20 , 257n
Ideas , 1, 2, 3, 8, 9, 12 , 14, 15 , 22 ,
25 , 26 , 27 , 33 , 37, 38 , 53 , 56 ,
61-3 , 64 , 65 , 91 , 93 , 107, 108 ,
113 , 114, 117, 118 , 125 , 126,
128 , 129, 137n, 142-5~ 159,
161 , 164, 165 76 , 178-99 ,
203n , 205n , 207n , 208n , 211 ,
224 , 254-5 , 259n , 266-8 ,
270
Aesthetic Ideas , 156 , 173 , 175-6
Dialectical Ideas , 65 , 182-5 , 199
Ideas of (theoretical or practica l)
reason , 22 , 26 , 108 , 119, 138n,
139n, 145-8 , 151 7, 16574 ,
175 , 176 , 178 , 179 , 202n , 204
239
Ideas of the understanding , 93 ,
108 , 111 , 114, 118 , 119 , 139n,
142 , 266
Problematic Ideas (or Ideasproblems) , 12 , 14, 25 , 56 , 61-3 ,
64 , 65 , 125 , 143 , 145 , 150,
185 , 186 , 187, 191 , 192 , 194,
195 , 196 , 197 199 202 206r~
254 , 255 , 266 , 267 , 270n
5ee al50 Problems
Identity , 4, 13 , 20-1 , 24 , 33 , 34 , 37,
38 , 39 , 40 , 41 , 42 , 43 , 50 , 86 ,
125 , 128 , 137n, 192, 210 , 211 ,
212 , 215 , 217 , 218 , 219 , 230
235 , 236 , 243 , 247 , 248 , 249 ,
251 , 254
Ill usion , 2, 43 , 55 , 57, 146 , 157,
180 , 187
Ill usions of reason , 1, 22 , 155
Ill usions of representation , 1, 20
Image of thought, 1, 11 , 13 , 18-23 ,

26 30 , 34 , 35 , 37, 43 , 45 , 47 ,
55 , 61 , 64 , 65 , 68n , 148 , 187,
251 , 265
Immanence , 8, 10, 116 , 117, 122 ,
181 , 184, 185 , 208n , 267-9 ,
270n
Infinire understanding, 111 , 112-13 ,
11517, 125 , 128 , 137n, 138n,
203n
Intensive magnitude , 102 , 109-10;
5ee al50 Qualiry
Intuitive intellect, 94 , 132n
Judgement
Aesthetic judgement of taste ,
157-8 , 159, 161-3 , 165 ,
169-71 , 173-4, 177, 178 ,
204
Re f1 ecring judgement, 156, 157,

158 , 159, 170
Kant , lmmanuel , 1-3 , 4, 5, 7,
10 14 1 21 -4 33 , 34,
36 43-4 48 51 3 56
66n , 67n , 72n , 74 , 75 , 76 ,

87, 88 , 90 , 92 , 93-102 ,
107 8, 111 , 112, 113 , 114,
116 129 130n, 131n ,
132n, 133n, 137n, 145-8,
149 , 150-76 , 179 , 180 ,
181 , 187, 201 , 202n , 203
204n , 210 , 211 , 212 , 213 ,
214 , 215 18 , 222 , 223 ,
229 , 230 , 231 , 233 , 234 ,
236 237 241 242 259
265 , 266 , 267, 268

Kerslake , Chrisrian , 9-11 , 93 , 148 ,
153 , 202n
Klossowski , Pierre , 236 , 243 , 247-8 ,
251 -4
Lautman , Albert, 12 , 65 , 145 ,
182-6 , 188 , 190, 199 , 205n ,
206n , 208n , 266
283
Leibniz , G. W. , 7, 36 , 42 , 57-61 ,
62 , 72n , 93 -4, 100, 102, 103-7,
109 , 112 , 114, 117-21 , 123 ,
124, 125 , 128 , 132 139n
140n, 211 , 218 256 266
Life , 10-11 , 16n, 28 , 63 , 77-8 , 83 ,
86 , 127, 219 , 221 , 226 , 230 ,
268
Limit, 63 , 65 , 90 , 104, 105 , 111 ,
112 , 136n, 143 , 146 , 167 23~
240 , 25960n
Lord , Beth, 93 , 133n, 137 8n
Maimon , Salomon , 5, 7, 14, 67n

75 , 92-102 , 107-19 , 121 , 125 ,
128 , 131n, 132n, 133n, 134n,
135n, 137n, 138n, 139n, 142 ,
185 , 203n , 266
Marx , Kad , 198 , 255
Metaphor, 173 , 192, 200 , 208n ,
242
Method of Dramatisation , 13 , 79 ,
84
lethod of Exhaustion , 105 , 196
Midgley , Nick , 93 , 97-8 , 132n,
256n , 259n
Minute perceptions , 62 , 114, 118 ,
119-21 , 124, 125 , 128 , 143
Multiplicity , 12 , 13 , 15n, 68n , 127,
145 , 156 , 181 , 185 , 186 , 187,
189-91 , 192, 195 , 196 , 197,
198 , 201 , 202 , 206n , 207n ,
239 , 243 , 247 , 267 , 268
Objective Reality , 52 , 95 , 96-8 , 113 ,

133n, 150, 152, 153 , 174
Objective Validity , 52 , 94 , 95-7,
133 152 , 153 , 154
Overman , 245 , 248-9 , 262n
Para-sense, 129 , 157, 178; see also
Discordant accord
Parrhesia , 249, 263n
Passive Synthesis , 6, 195, 210 , 218 ,
219-24 , 229 , 267
Patton , PauI , 5, 130n, 255n
Plasticity/plastic principle, 12, 28 ,
29 , 6~68n 76 88 , 90 , 91 ,
149 , 266
Plato , 1-2, 5, 23 , 26 , 35 , 36-8 , 43 ,
53 , 64 , 69n , 74 , 78 , 146-8 , 179,
180-1 , 182, 202n , 213 , 227
Plotnitsky, Arkady , 201 , 207n ,
209n
Poincar , Henri , 184 , 188 9
Present (l iving or lived) , 218 , 219 ,
220, 222~ 223, 230; see also
Time
Presentations , 112 , 116, 117, 119,
121 , 219
Problem(s) , 9, 12, 14, 15n, 25 , 27,
28 , 29 , 44 , 45 , 48 , 54 , 55-61 ,
63 , 65 , 70n , 71n , 128 , 142 ,
143 , 145 , 146 , 148-50, 165,
180 , 181 , 182 , 183 , 184, 185,
186 , 187, 189 , 193 , 194, 197,
199 , 202 , 207-8n , 242 , 254 ,
269 , 270n; see also Ideas:
problematic Ideas
Proust, Marcel , 16 27 63 64
126 , 141n, 144, 223 , 224 , 227,
266
Pure intuition , 88 , 95 , 97 8, 100;
see also Formal intuition
Pure past, 8, 144, 218 , 223 30 ,
258n; see also Time
Purposiveness/purposive
relationship , 154, 163 , 167,
170 , 171 , 17~202n 204n
Necessity, 2-3 , 23-5 , 46 , 52 , 53 , 58 ,

59 , 60 , 67n , 74 , 101 , 140n, 152,
162 , 170, 219 , 228 , 248 , 252
Newton , Isaac , 103-4, 106 , 107,
109 , 136n, 214
Nietzsche , Friedrich , 4, 5 13 ,
14 , 35 , 46 , 67r~68n 75-91
129n, 131n , 234 , 235 , 236 ,
243-8 , 250 , 251-3 , 254 , 255 ,
262n , 266
Nou1e11011 107-8
284
lndex
Quality , 82 , 84 , 85 , 86 , 88 , 89 ,
102 , 109-10, 119, 124 , 127,
128 , 193; see also Intensive
magnitude
Quid juris? , 95 , 101 , 102 , 111 , 113 ,
114 , 165
Sense , 3, 12 , 13 , 14, 44 , 47-55 , 56 ,

57 58 67n 70r~71n 74 87 ,
89-90 , 91 , 92 , 153 , 155 , 201 ,
253
Sign , 12, 13 , 17, 25 , 27 , 63 5, 89 ,
143-5 179 224 252 266
Real thought, 101 , 111 , 133n

Recognition , 14, 22 , 24 , 32 , 33 ,
34-5 , 43 , 44 , 47, 62 , 64 , 65 , 87,
143 , 265 , 267
Relation
Differemial relation(s) , 7, 8,
12 , 56 , 62 , 65 , 86 , 93 , 102
106 , 110, 111 , 118 , 121 , 122,
123 , 124, 186 , 187, 189, 191 ,
193 , 194 , 196 , 267; see also
Differential(s)
Pure relations , 109 , 110
Reciprocal (or reciprocally
determined) relations , 85 , 108 ,
110 , 124, 193 , 198
Reperirion , 9, 13 , 64 , 131n, 144 ,
210 , 220 , 221 , 245-8 , 255 ,
269
Representation (orderllogiclmodel
of) , 1, 2 , 13 , 19 , 20 , 23 , 26 , 28 ,
31-2 , 35 , 36-9 , 41-3 , 46 , 65 ,
69n , 80 , 111-12, 115 , 117, 124,
125 , 143 , 218 , 242-3 , 245 ,
247
Riemann , Bernhard, 190-1 , 207n
Rimbaud , Arthur , 63 , 72n , 215 , 245
R lli Marc , 6-7
Saint-Hilairetienne Geoffroy ,
196-7
Salanskis , Jean-1ichel 194, 201 ,
209n
Sauragnargues Anne , 11 , 93 , 230 ,
258n
Schaub , Mirjam , 204-5n , 259n
Schema , 94 , 95 , 113 , 134n , 152 ,
153 -4, 155 , 162 , 168 , 172 , 173 ,
183 , 184, 185
Simulacra , 37-8 , 46 , 74 , 210 , 227,

243 , 246 , 247, 248 , 255
Singularity, 62 , 65 , 121 , 122 , 124,
128 , 139n, 140n, 149, 186 , 187,
188-9, 191 , 192 , 193 , 194 , 195 ,
199 206n 246 26~268
Smith, Daniel , 7, 11 , 93 , 156 , 178 ,

179 , 189
Sophocles , 231-3
Spinoza , Benedictus de, 43 , 112-13 ,
137n, 211
Spinozism , 112-13 , 133n, 137n
Split subject, 4 , 14 , 142, 210 , 212 ,
230 , 249 , 267, 269n
Static synthesis (of time) , 212 , 218 ,
236~239 240 250 261n;see
also Time
Sublime , 14, 156 , 158 , 165-9 , 170 ,

178 , 204n
Symbolisation , 169 , 172-3 , 205n
Time , 3-4, 8, 12, 14, 26 , 28 , 51 , 64 ,
66n , 68n , 86 , 87, 94 , 95 , 96 ,
98 , 99 , 100 , 101 , 134n, 195-6 ,
210 , 21219 , 220 , 222 , 223-5 ,
226 , 228 , 229-30 , 231 , 233 -4,
236 , 237 , 239-4 2 , 243 -4, 248 ,
255-6n , 259n; see also Present
(l iving or lived); Pure Past; Static
Synthesis (of time)
hird synthesis/empty form of
time , 3-4, 12 , 14, 210 , 212 ,
214 , 215 , 216 , 218 , 222 , 228 ,
229 , 230 , 231 , 233 , 236 , 240-2 ,
243 249-51 252 259 260n
261n , 267
Transcendental empiricism , 1, 6-7 ,
9, 11 , 23 , 25 , 85 , 91 , 93 , 257n ,
268
285
subject, 1, 2, 52 ,
181 , 210 , 230 , 242 , 267,
268
ruth 2 3, 21 , 22 , 23 , 24 , 26 ,
27, 31 , 32 , 35 , 43 , 44-54 , 57,
59-61 , 63 -4, 6611 , 71n , 74 , 76 ,
77-9 , 89-90 , 97, 116 , 129n,
140n, 144, 154 , 180, 187, 202 ,
20311 , 211 , 247, 249 , 265
ranscendental
Unco l1 ditioned , 10, 54 , 77, 147

152, 156
UI1CO I1 Scious (differential) , 114 , 117,
119-21 , 125 , 128 , 142
Ul1 thought in thought, 114, 116,
117, 169 , 261n
Untimely , 28 , 68n
RANSCENDENTALIDEAS
Vrtualitylthe virtu 8, 12 , 13 ,
56 , 91 , 126-8 , 184 , 185 , 189 ,
195-6 , 197 8, 205n , 207-8n ,
223 226 230 241 250 258
259n , 26111 , 267 , 268 , 269

Vuillemin , J ules , 116
Weierstrass , Karl , 107, 122-3 , 124,
13611, 14011 , 187-8 , 206n
Will to power , 75 , 76 , 77 , 79 , 80 ,
83~ 88-91 , 131n , 245 , 255
Williams , James , 241 , 249-51 , 255 ,
26111
Zammito, JOh l1, 159-60, 20411
Zourabichvili , Fra l1'ois , 5, 67n , 68n ,
70n , 71n
286

Daniela Voss - Conditions of Thought - Deleuze and Transcendental Ideas

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Daniela Voss - Conditions of Thought - Deleuze and Transcendental Ideas

Hochgeladen von

Copyright:

Verfügbare Formate

Conditions of Thought:

Deleuze and Transcendental Ideas

Plateaus - New Directions in Deleuze Studies

Christian Kerslake , Immanence and the Vertigo of Philosophy: From Kant to

Jean-Clet Martin , Variations: The Philosophy of Gilles Deleuze , translated by

<D Dani eI a Voss , 2013

The bibliography contains details of the editions and translations

Anti-Oedipus: Capitalism and Schizophrenia (vo l. 1)

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

Critique of the Power of Jume11t

This project on a Deleuzian transcendental philosophy is born out of

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

by means of an intellectual comrrlU nion with the origin of all things'. 3

be impossible , if not for some condition'.4 The necessity involved is

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

two aspects of human nature in the forrn of an internal Difference

It is perhaps not surprising that Deleuze's transcendental philosophy

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

particular Difference and Repetition (1968). We will also refer to

sense in which Deleuze's philosophy is transcendental , R lli focuses

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

thought of as a hyper-rationalism rather than an errlpiricisn1\14

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

a bizarre reading of Deleuze , as if Deleuze's philosophy only served

Throughout our st l1 dy , Daniel Srrl th's essays on Deleuze in relation

DELEUZE AND TRANSCENDENTAL IDEAS

Roe and published by Edinburgh University Press in 2009 , also

it to a dissolved subject or fractured 'I' which is shown to be the

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

combination of forces that determines the sense and val l1 e of what

opening up the subject and referring it to the realm of Ideas. We will

C f. NP 1 8990/102 9 1/ 104 and 106112 1.

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

The Dogmatic Image of l'hought

The image of thought is a recurrent thenle in Deleuze's philosophy:

The Dogmatic Image of Thought

ThllS for Deleuze , a critiqlle of the In1age of thOllght cannot simply

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

need not operate visibly; rnore often they rernain llnconsciollS

The Dogn1atic Image o{ Thought

Deleuze's Critique of the Dogmatic lmage of Thought

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

According to Kant , however , the greatest threat to the legitirrlate

The Dogmatic Image of Thought

(DR 1441187 , 56/79-80).

Like Kan Deleuze is interested in the necessity of thought , that is

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

Kant seeks to ground our knowledge of the exterior world on

TheD tic 111ge fThought

If Deleuze calls his philosophy transcendental empiricism' , then

A sign is material and intensive in nature. In Proust and Signs ,

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

theory of thought depends on a typology of forces. And once again

The Dogmatic In1age of Thought

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

Finally , the philosopher has an essential relation to tirne: he is a

Dogmatic Image o{ Thought

demand for a thought without image is due to the fact that in

CONDITIONS OF THOUGHT: DELEUZE AND TRANSCENDENTAL IDEAS

the rhizome grows frorn the rniddle and proliferates in an endless

Postulates of the Doglnatic Jmage of Thought

The Dogmatic Image ol Thought