[In the following essay, Neil traces Deleuze's philosophical development by discussing important influences on his work.]

Even 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 itself.¹

HISTORY OF PHILOSOPHY: RECOUNTING IMAGINARY BOOKS

Gilles Deleuze observes a distinction between writing history of philosophy and “doing” philosophy. We are interested here in how this distinction is set up. How does Deleuze conceive of the relation between history of philosophy and doing philosophy? What is the use of historiography and how can it be brought to bear on the present? In the preface to the English edition of Difference and Repetition Deleuze writes this:

There is a great difference between writing history of philosophy and writing philosophy. In the one case we study the arrows or the tools of a great thinker, the trophies and the prey, the continents discovered. In the other case we trim our own arrows, or gather those which seem to us the finest in order to try and send them in other directions, even if the distance covered is not astronomical but relatively small. We try to speak in our own name only to learn that a proper name designates no more than the outcome of a body of work—in other words the concepts discovered, on condition that we were able to express and imbue them with life using all the possibilities of language.²

This metaphor of the arrow is instructive. Deleuze says that “After I had studied Hume, Spinoza, Nietzsche, and Proust, all of whom fired me with enthusiasm, Difference and Repetition was the first book in which I tried to ‘do philosophy.’” Deleuze distinguishes here between his earlier texts on the history of philosophy and his first text which does philosophy—on the one hand, the selection of some arrows and tools, the tensioning of a bow, and, on the other, their deployment in a new direction. Following Nietzsche, Deleuze wants to be an untimely philosopher—he desires a philosophy that is not a philosophy of history, nor a philosophy of the universal, but untimely. That is to say—and these words are Nietzsche's “acting counter to our time and thereby acting on our time and, let us hope, for the benefit of a time to come.” Deleuze anticipates a time when it will no longer be possible to write philosophy books as they have been traditionally produced. The search for new means of philosophical expression was begun by Nietzsche, and Deleuze hints here that his own work on cinema, theatre, and the visual arts is animated by this need for new modes of philosophical expression. At any rate it is in this context that Deleuze raises the question of the history of philosophy.

He says that the history of philosophy should play a role roughly analogous to collage in painting. This means that elements, or more precisely concepts, are lifted from the textual context in which they were created and juxtaposed with concepts from other places in ways that make familiar concepts strange and imbue them with new powers. A commentary reproduces a concept but it is reproduced in a new context—or rather, on a different plane—and for Deleuze commentary must be conscious of the mutations that occur when historical concepts are reproduced in the present. In fact it is the field of possible mutations which constitute commentary's possibilities.

The history of philosophy is the reproduction of philosophy itself. In the history of philosophy a commentary should act as a veritable double and bear the maximum modification appropriate to a double.³

Even more provocatively, Deleuze suggests that it should be possible to recount a real book of past philosophy as if it were an imaginary and feigned book. And here, by way of explanation, he mentions a story by Borges. The story is “Pierre Menard, Author of the Quixote.”⁴ As a guide to what Deleuze might mean here it is quite illuminating.

Borges had an extraordinary talent for the invention of imaginary texts. In “Pierre Menard, Author of the Quixote” he tells the story of a fantastical repetition of a real text—Don Quixote de la Mancha—which is reproduced by an imaginary author, Pierre Menard, without reference to the original. He invents Menard as an obscure literary figure of the early part of this century, and supplies him with a list of publications that suggests some knowledge of Descartes, Leibniz, Boolean logic, and an interest in the metric laws of French prose. Inspired by the ideal of a total identification with a given author, Menard sets himself an apparently impossible project. He sets out to write the Quixote; not a contemporary version but the actual Quixote. But Menard is not interested in simply transcribing Don Quixote. Such a purely mechanical repetition would be of no interest. His intention was to write himself, without reference to the book, a few pages that would correspond word for word and line for line with Miguel de Cervantes' early seventeenth century novel. The first method he tried was relatively simple:

Know Spanish well, recover the Catholic faith, fight against the Moors or the Turk, forget the history of Europe between the years 1602 and 1918—in short, be Miguel de Cervantes.⁵

He discards this approach because it is too easy. We will reply that, on the contrary, this is an impossible task but “the undertaking was impossible from the very beginning and of all the impossible ways of carrying it out, this was the least interesting.” Also Menard thinks that, in the twentieth century, to be a popular novelist of the seventeenth century is something of a diminution. He decides that it is more interesting to try and write the book as Pierre Menard, to reach the Quixote through Menard's experiences, and so this is his undertaking. (Borges interest in infinities emerges here because he has Menard say that his task is not essentially difficult, but he would need to be immortal to carry it out.) We are told that Menard actually managed to produce a couple of chapters which were identical with the original, although, tragically, they have been lost, depriving us of perhaps the most significant literary achievement of this century.

For our purposes the key to the story is found in the following sentence, a possibility that we can imagine carrying out: “Shall I confess that I often imagine he did finish it and that I read the Quixote—all of it—as if Menard had conceived it.” What is achieved by such a reading? “Cervantes' text and Menard's are verbally identical, but the second is almost infinitely richer. (More ambiguous his detractors will say, but ambiguity is richness.)” Borges lists some of the many important differences he finds between the two Quixotes, of which I think this passage sheds the most light on Deleuze's reference to the story:

It is a revelation to compare Menard's Don Quixote with Cervantes'. The latter for example wrote (part one, chapter nine):

truth whose mother is history, rival of time, depository of deeds, witness of the past, exemplar and adviser to the present, and the future's counselor.

Written in the seventeenth century, written by the ‘lay genius’ Cervantes, this enumeration is mere rhetorical praise of history. Menard, on the other, writes:

truth whose mother is history, rival of time, depository of deeds, witness of the past, exemplar and adviser to the present and the future's counselor.

History, the mother of truth: the idea is astounding. Menard, a contemporary of William James, does not define history as an inquiry into reality but as its origin. Historical truth, for him, is not what has happened; it is what we judge to have happened. The final phrases—exemplar and adviser to the present, and the future's counsellor—are brazenly pragmatic.⁶

What if the Imitatio Christi were attributed to Celine or to James Joyce? This technique, Borges claims, fills the most placid works with adventure. The interesting question then is, how is it possible that two texts which are word for word identical could be different? How could they be any more different than two copies of a book? What difference does it make to suppose two authors and two separate acts of production? Put that way, the answer is fairly obvious. Nothing is different about the texts, but something changes in the reader. Our knowledge of the circumstances of the text's production alters our reading strategies. We can attribute certain philosophical sensitivities to Menard that could never be attributed to Cervantes. We have to suspect that Menard is up to something in places where an identical sentence by Cervantes is surely an innocent choice of words with no such metaphysical implications—just as the style and mode of expression call for some explanation in Menard's writing, whereas in Cervantes' book the style is paradigmatic; a canonical work of the period.

INSANE CREATION: DELEUZE ON HUME

Deleuze's interest in Borges' story then is that Borges imagines the most exact, the most strict repetition, which nevertheless produces a maximal difference. The difference resides not in the text but in the subject who contemplates. One of the most striking features of Deleuze's history of philosophy is the transformation he effects in the figure of the contemplative subject. Indeed for Deleuze the subject is produced as an effect of repetition. The subject is produced as a virtual repetition of an actual passive synthesis that bears the same relation to the given as commentary to primary text. By way of example I wish to look at one of Deleuze's historical books, his book on Hume, to illustrate more concretely how this strategy works.

Deleuze says that commentaries on the history of philosophy should represent a kind of slow motion; an immobilization of the text; not only the text they relate to but also the text in which they are inserted. Commentaries “have a double existence and a corresponding ideal; the pure repetition of the former text and the present text in one another.”⁷Empiricism and Subjectivity was Deleuze's first book, published in 1953 when he was twenty-eight years old.⁸ This commentary is quite extraordinary because it is a radical reading which is also an extremely rigorous reading. Most of the discussion concerns the Treatise of Human Nature, and Deleuze gives nearly four hundred citations to that book—a book which is not much longer than six hundred pages. He quotes extensively and devotes considerable and meticulous attention to Hume's exact wording. Yet Deleuze's Hume is nearly unrecognizable by the industry standards in Hume scholarship. Deleuze's Hume is not primarily concerned with establishing the foundation of knowledge in experience, nor with skepticism about metaphysical concepts that cannot be derived from experience, nor with the attempt to create a strictly naturalistic ethics. Rather than being a reaction against concepts or an appeal to lived experience, Deleuze says that “empiricism undertakes the most insane creation of concepts ever seen or heard.”⁹ Deleuze does not force the text. Hume does indeed say everything that Deleuze attributes to him. This commentary observes the highest scholarly standards of textual evidence. A rigorous yet unfamiliar commentary; a careful reproduction which produces a significant difference. How is this effect achieved? Obviously it is not by positing some fictional contemporary author, but Deleuze does exploit the surprising effects of anachronism that are described in the Borges story.

How can we criticize a philosophy? How should we distinguish between good, helpful, progressive, and constructive criticism and the pedantic and reactive scribblings of underpaid academic scriveners? Firstly we can say what good criticism is not. It is not simply writing off philosophers as examples and victims of the peculiar delusions characterizing their period. That is, we do not present a psychology of an author as if we can denounce his system by showing its basis in his personal tastes or the spirit of the times. Some fictitious psychology of an author's intentions cannot be a sufficient criticism of his theory. In this sense, Deleuze's approach to the history of philosophy is curiously ahistorical. We do not read philosophers against some official chronology of concepts and problems as if that (perhaps with some biographical supplements) would license us to say that a thinker must have meant X or could not have meant X because of the publication date. That kind of criticism cannot be useful for doing philosophy. Nor is it interesting to point out that a certain system has a premise or a result that we are not inclined to believe. For instance, we are told that, on account of his atomism, Hume has pulverized the given and we know better. Perhaps we do know better, but that is not critique.

Rather we have to understand what a philosophical theory is and Deleuze's offers a remarkable definition here.¹⁰ A philosophical theory is an elaborately developed question and nothing else. It is not the resolution of a problem (as if problems stand outside of philosophy demanding answers). Rather it is the elaboration, to the very end, of the necessary implications of a formulated question. The theorist questions the world, but that is to say that the theorist has a question—and this question is like a lens through which he examines things; by subjecting things to this question, by this forced subsumption, things will reveal an essence or a nature.

On this view to engage with a philosophy is to feel the force of a question, to take a question seriously. To criticize a philosophy is to criticize the question that motivates it.

In truth only one kind of objection is worthwhile: the objection which shows that the question raised by a philosopher is not a good question, that it does not force the nature of things enough, that it should be raised in another way, that we should raise it in a better way, or that we should raise a different question. It is exactly in this way that a great philosopher criticizes another: for example … this is how Kant criticizes Hume.¹¹

I want to take these remarks and bring them to bear on my question about Empiricism and Subjectivity. How does Deleuze achieve this effect of a repetition which produces a remarkable difference? The answer, I suggest, is that Deleuze insinuates a question into Hume. Deleuze insists that Hume has a central question, a question that recurs in a progressively more refined form—and that the whole of the Treatise bears a relation to this question. The book opens with a claim about Hume's fundamental project. Hume wants to create a science of humanity and he sees that this requires the substitution of a psychology of the mind by a psychology of the mind's affections. In the following pages the question is repeated in two different forms—how does the mind become a subject? or—how is the subject constituted inside the given?

For Hume the mind is not the subject. The mind precedes the subject and the development of subjectivity is a process that happens to the mind. Hume constantly affirms the identity of the mind, the imagination, and ideas. The mind is not nature and it does not have a nature. It is nothing other than the ideas in the mind. Ideas are given, or rather, ideas are the given; they are the elements of experience. The mind is given as the collection of ideas. In the beginning the mind is not a system but a purely contingent collection—an assemblage. When we say that ideas are “in” the mind this is just a requirement of grammar. The word “in” here does not designate inherence in a subject or containment in a place. The mind is nothing other than the collection of ideas. The place is not different from what takes place in it. It is a collection without an album, a theatre without a stage, a flux of perceptions. Hume insists that the mind is not a thing, although the fact that we have a noun, “mind,” can mislead us into thinking that way. He offers an analogy of the mind as a kind of theater—an analogy that clearly appeals to Deleuze. Hume says:

The mind is a kind of theatre where several perceptions successively make their appearance; pass, re-pass, glide away and mingle in an infinite variety of postures and situations. There is properly no simplicity in it at any one time, nor identity in difference; whatever natural propension we may have to imagine that simplicity and identity. The comparison of the theatre must not mislead us. They are the successive perceptions only that constitute the mind; nor have we the most distant notion of the place where these scenes are represented, or of the materials, of which it is composed.¹²

To say then that the mind becomes a subject is to say that the collection of ideas is organised into a system. In becoming a system the mind acquires a nature; it develops tendencies and processes, and ultimately, through the formation of beliefs, the constituted subject is able to transcend the given.

Towards the end of this essay I will discuss why Deleuze wants to develop an empiricist conception of subjectivity. Next, however, I wish to draw out the implications of Deleuze's reading strategy here. We have a question then: How is the subject constituted inside the given? Deleuze tells us that this is Hume's central question and this question, with its variants and the further questions it demands, defines the interpretive locus of Empiricism and Subjectivity. For a reader like myself, whose image of Hume has been mediated by the significance of Hume within the analytic tradition, this is a surprising and unfamiliar formulation of Hume's philosophical motivation. I do not know of any other commentators who attribute this question to Hume; and certainly not as his fundamental and overriding concern. (More familiar, for instance, would be to locate Hume as a key moment in the history of attempts to make philosophy Newtonian.) What should we say about this? It is very significant, I think, that throughout Empiricism and Subjectivity Deleuze writes of “the subject,” “the given,” and Hume's “idea of subjectivity.” We notice that the question Deleuze attributes to Hume, which is certainly not a quotation, is posed in anachronistic terms. Hume never uses the term “subject” as the philosophical name for the epistemically self-conscious self. The only “subjects” Hume knows are those persons owing obedience to a government. Similarly the philosophical notion of “the given” was not yet in circulation in 1739. More specifically, we note that “the subject” and “the given” are Kantian terms whose meaning, in the first instance, is determined within the project of transcendental idealism.

Why speak of the subject here? Where Hume speaks of the development of tendency, custom, expectation, disposition, and habit, why say that Hume is speaking of the subject? Why make Hume untimely in this way? I think it is in order to effect a positive repetition that enables Humean tools to act on the present. Hume is typically seen to have been instrumental in raising the philosophical problem of subjectivity; but Hume is thought to have foundered on his inability to have grasped the real implications of the subject. Hume wants to account for the subject in terms of the given, and so the forces that create the subject, the principles of human nature, must be prior to and independent of the subject. These principles would transcend and account for the subject. Kant reproaches Hume for a presupposition required by his dualism between the principles of human nature (the principles of association) and the subject constituted by the action of those principles. Hume must assume a harmony or agreement between the subject and the given; that is, between human nature and nature. Kant objects that, unless the given is initially subject to the very same principles that regulate experience in an empirical subject, then the subject would never encounter this agreement except by accident. In Hume's principles the image of human nature has already been projected onto a pre-subjective world in order to account for the subject who experiences a world. The great Kantian correction here is to see that the problem must be reversed. Rather we see that the given is always posited by a subject, and so it is the given that has to be related to the subject. The given is not a thing in itself but rather a set of phenomena—and the set of phenomena can be presented as a nature only by means of an a priori synthesis.¹³

For Deleuze, the core tenet of empiricism, and indeed the defining doctrine of his own commitment to empiricism, is associationism. Put simply, associationism is the thesis that relations are external to their terms. Nothing intrinsic to things determines the relations that we establish between them.

Deleuze finds in Hume the elements of a response to Kant's critique of empiricism. We are interested here in why Deleuze phrases Hume's fundamental question in Kantian terms. The striking effect of this strategy is that it enables a reading of Hume as if he were actually sensitive to the Kantian critique. This is exemplified in the following passage which concludes Deleuze's discussion of Kant:

Let us return, then, to the question that Hume raised in the way that he raised it, which we can now better understand: how can it be developed? According to Hume, and also Kant, the principles of knowledge are not derived from experience. But in the case of Hume, nothing within thought surpasses the imagination, nothing is transcendental, because these principles are simply principles of our nature, and because they render possible an experience without at the same time rendering necessary the objects of this experience. Only one device will permit Hume to present the agreement between human nature and nature as something more than an accidental, indeterminate, and contingent agreement; this device will be purposiveness.¹⁴

Deleuze reads Hume as if (like Deleuze) he wants to affirm the primacy of immanence rather than referring immanence back to some transcendental condition.¹⁵ Purposiveness is “the agreement of the subject with the given, with the powers of the given, and with nature.”¹⁶ Again we note that “purposiveness” is not Hume's term but Deleuze's. The idea of the world is, for Hume, a fiction of the imagination. Yet this fiction actually becomes one of the principles of human nature. This illegitimate belief in the world appears as the horizon of all possible legitimate beliefs and so it must co-exist with those principles of the understanding whose role is to correct fictions and eliminate illegitimate beliefs. Nothing in the given justifies the relations we establish and the totalities that the mind assembles. The totalities we form depend on the principles of the passions and the principles of association—and as such they are purely functional. For the empiricist, the idea of disinterested belief is incoherent. Yet, although we cannot know the hidden powers on which the given depends, those powers must agree with the relations and functional totalities that the subject establishes within the given. The agreement between intentional finality and nature can only be thought, “and it is undoubtedly the weakest and emptiest of thoughts.”

Philosophy must constitute itself as a theory of what we are doing, not as a theory of what there is. What we do has its principles; and being can only be grasped as the object of a synthetic relation with the very principles of what we do.¹⁷

A thorough account of the meaning of “purposiveness” here would take us too far away from our theme. The point to note is that, under the heading of “purposiveness,” Deleuze develops a Humean response to Kant's critique, while drawing only on the resources of Hume's writings. I think though, that we can now begin to see how Deleuze's “Borgesian” approach to history of philosophy produces such remarkable surprises. Consider the following passage which we find in the first chapter of Empiricism and Subjectivity.

When Hume speaks of an act of the mind—of a disposition—he does not mean that the mind is active but that it is activated and that it has become a subject. The coherent paradox of Hume's philosophy is that it offers a subjectivity which transcends itself, without being any less passive. Subjectivity is determined as an effect; it is an impression of reflection. The mind, having been affected by the principles, turns now into a subject.¹⁸

The use of the term “passive” here is not innocent. When Deleuze “does” philosophy in Difference and Repetition and later works, this theme of purposiveness, and more specifically Hume's principle of habit, is explicitly aligned with Bergsonian differentiation and with the phenomenological concept of “passive synthesis.”¹⁹ Hume's (and Bergson's) examples of the effects of repetition on the subject “leave us at the level of sensible and perceptual syntheses.”²⁰ We arrive at the concept of passive synthesis when we extend the Humean concept of habit to the sub-representative level; that is, to the organic level. I would argue that Deleuze also finds significant resonance between Humean purposiveness and his interpretation of the “will to power” in Nietzsche. But again, the analysis of these connections cannot be undertaken here.

Should we say then that the apparent anachronism of Deleuze's reading is a mistake; that it is improper to find within the empiricist tradition an inquiry into the constitution of “the subject” inside “the given”? As we have seen Deleuze discards that kind of criticism as unhelpful. He deliberately disregards the strictures on textual interpretation that are imposed by extra-textual information about an author's life and times. The difference between Cervantes' Quixote and Menard's is a product, not of any difference in the words but of a difference in the concepts we are willing to attribute to these two authors separated by some three hundred years. Deleuze's reading of Hume makes Hume philosophically sensitive to his own legacy—so the powers of fiction are put into play here. This is perhaps what Deleuze means by suggesting that it should be possible to recount a real book of past philosophy as if it were an imaginary and feigned book. If Deleuze's Hume is, in some way, an imaginary and feigned author, we must note that any criticism of Deleuze on this point can only appeal to equally fictional claims about Hume's psychology.

Just before he alludes to Borges, Deleuze parenthetically imagines a philosophically bearded Hegel and a clean shaven Marx. As I understand this, the moral to draw from the Borges story is that the reading of historical texts is often governed by a desire to protect the identity of the proper name. By contrast Deleuze's commentary is a double which seeks the maximum modification appropriate to a double. The question becomes not “what could Hume have been saying?” but rather “what can Hume be made to say?” A better description then is to say that Deleuze's reading is untimely. For Deleuze, Hume, like Nietzsche, belongs to a select group of untimely philosophers who insist on the concrete richness of the sensible, who do not subordinate difference to identity, who reject the logic of negation, who affirm the powers of chance and contingency and who express a hatred of interiority. For Deleuze a philosophy is primarily characterized by its motivating question. We criticize a question if “it does not force the nature of things enough.” The idea that a good question is forceful, a bad question is not forceful enough, directs us to look at what a question does. Hume, like Nietzsche, conceives of the subject (despite grammatical appearances) as an effect, not a ground; as an event, not a first cause. Thus in Nietzschean terms, Hume's question is active and affirmative, because it casts the subject as an effect of pre-personal, external forces acting on an arbitrary (chance) collection of raw materials.

DOING PHILOSOPHY: THE NECESSITY OF STYLE

In Deleuze and Felix Guattari's last work, What Is Philosophy, we are told that philosophy “is the art of forming, inventing, and fabricating concepts.”²¹ The philosopher's vocation is the creation of concepts, and it is stressed (in Nietzschean tones) that concepts are not discovered “waiting for us ready-made, like heavenly bodies”²²-rather they are created. Concepts are responses, or solutions, to problems and they can only be assessed “as a function of their problems and their plane.”²³ This characterization of the relation of concepts to problems repeats and amplifies the view that we found in Deleuze's first book, Empiricism and Subjectivity—that a philosophical theory is an elaborately developed question. Concepts can only be usefully criticized from the perspective of the problem to which they respond and from which they take their value and their meaning.

A concept always has the truth that falls to it as a function of the conditions of its creation. … If one can still be a Platonist, Cartesian, or Kantian today, it is because one is justified in thinking that their concepts can be reactivated in our problems and inspire those concepts that need to be created. What is the best way to follow the great philosophers? Is it to repeat what they said or to do what they did, that is, create concepts for problems that necessarily change?²⁴

Theories are defined by their problems (formulated rigorously) and good critique is, at bottom, critique of problems. As we have seen, this idea is a constant throughout Deleuze's work, present in the first and the last books. But in What Is Philosophy? we find a much more complex and nuanced picture of the relation of concepts to problems. Concepts inhabit a “plane of immanence” that must also be created, and they are brought to life by “conceptual personae.” Philosophy, Deleuze insists, is a constructivism, and there are three interdependent dimensions to the construction of a philosophy: the laying out of a plane of immanence, the invention of the personae who see things in terms of this plane (who “carry out the movements that describe the author's plane of immanence”),²⁵ and the creation of concepts. We are interested then in how an understanding of Deleuze's distinction between history of philosophy and doing philosophy helps to illuminate this picture of the philosopher's enterprise.

Two attributes of concepts are of interest here. Firstly concepts are always complex. All concepts have several components. They are a multiplicity of components which, although distinct, are rendered inseparable within the concept.²⁶ Because of this composite nature concepts may overlap, having a “threshold of indiscernability” with another concept. Secondly concepts are signed (Aristotle's substance, Descartes' cogito, Kant's condition, etc.). The name of a concept may be an ordinary word, an archaism, or a neologism “shot through with almost crazy etymological exercises.”²⁷ In every case however we find that the language employed or invented for the creation of concepts is determined, in part, by necessities of style. In the creation of a concept we find the exercise of a specific philosophical taste. The idea that there are ineliminable elements of style and taste in the fashioning of concepts is not just the claim that there are distinct styles of philosophy. Rather we see here the Nietzschean idea that style itself is a kind of rigor. A style is a form of consistency that imposes strict criteria on the selection and arrangement of elements. Any unified and coherent point of view, if it is identifiable as a point of view, is also, inevitably, a style. Deleuze follows Nietzsche in insisting that philosophy is a constructivism—that concepts are not discovered but enter the world as new-born, created entities which must then be “baptized.” It follows then that style is not a superficial dressing over the essential and eternal content of concepts. On the contrary, style is a defining and determining dimension of a philosophical system.

Style is a form of discipline. The philosopher (like the artist) gives the law to himself. The Nietzschean legislator however is not like the Kantian legislator who, through the movement of pure reason, derives the categorical imperative which applies equally to all rational creatures. The Nietzschean legislator submits to a style of his own invention, which is only ever one possible style. To make a law of style is to create a mode of obedience, a discipline, which remains mindful that the law is invented. An affirmative nature embraces the necessity of style and signs its creations. Conversely the rejection of style—the depiction of style as that which merely obscures the truth of a concept—is a symptom of ressentiment.²⁸ It is for this reason that concepts remain signed.²⁹ The rigorous thinker must understand the imperatives of style just as the philosophical creation irreducibly involves the creation of a style. Nevertheless, the author of a concept cannot determine its subsequent history:

Although concepts are dated, signed and baptized, they have their own way of not dying while remaining subject to constraints of renewal, replacement and mutation that give philosophy a history as well as a turbulent geography, each moment and place of which is preserved (but in time) and passes (but outside time).³⁰

Philosophy's “turbulent geography” here refers to the planes or territories which philosophers establish—planes which their concepts inhabit and presuppose.

THE PLANE OF IMMANENCE AS THE IMAGE OF THOUGHT

Deleuze and Guattari offer a number of somewhat elliptical descriptions of “the plane of immanence.”³¹ The plane is a fluid milieu which “moves infinitely in itself.” Concepts are the “archipelago or skeletal frame,” whereas the plane is “the breath that suffuses the separate parts.” Concepts are “concrete assemblages, like the configurations of a machine,” but the plane is “the abstract machine of which these assemblages are the working parts.” “Concepts are events, but the plane is the horizon of events.” Concepts occupy a plane, but “the plane is like a desert that concepts populate without dividing up.” All of these statements suggest different places in Deleuze and Guattari's work where we might look for assistance. But for the purposes of this essay the statement that gives us our starting point is this one:

The plane of immanence is not a concept that is or can be thought but rather the image of thought, the image thought gives itself ofwhat it means to think, to make use of thought, to find one's bearings in thought.³²

The plane of immanence is “the image of thought.” The “image of thought” refers us to chapter three of Difference and Repetition and to the problem of presuppositions in philosophy. In philosophy, unlike science, presuppositions are as much subjective as objective, and cannot be eliminated by axiomatic rigor. The image of thought is that representation of the nature of thought that is presupposed in a philosophical system: the character of thought taken as implicit, transparent and indisputably present to anyone who reflects on thought itself. Descartes, for instance, in order to avoid any presuppositions about rationality or the validity of sense experience, refers all presuppositions back to the absolute subjective certainty of self presence—“a subjective understanding implicitly presupposed by the ‘I think’ as first concept.”³³

The image of thought retains only what thought can claim by right: what pertains to thought, what must belong to thought by its very nature. Deleuze says that features of the plane are diagrammatic whereas concepts are intensive features. The plane is constructed of intuitions whereas concepts are intensions. If philosophy is the creation of concepts then the plane of immanence is prephilosophical, in that concepts must refer to some prephilosophical understanding. The plane of immanence is prephilosophical, not in the temporal sense of pre-existent but “rather something that does not exist outside of philosophy although philosophy presupposes it. These are its internal conditions.”³⁴ The plane of immanence is the substrate on which concepts are created and philosophy requires both the institution of a plane and the creation of concepts.

For the purpose of this essay what needs to be emphasized is that the plane of immanence is closely connected with the idea that a philosophical system has a fundamental problem or question which determines its sense and structure. Concepts belong to a plane and they cannot be assessed except as a function of their plane and their problems. Deleuze sometimes refers to the plane as a “plane of consistency.” Concepts are defined on a plane, on an image of thought that they presuppose. The concepts that inhabit a given plane are consistent in virtue of their belonging to the same plane. The exoconsistency between concepts depends on their being able to inhabit the same plane.

With this (overly compressed) discussion of “concepts” and “planes of immanence” in mind we can return to our original question about the relation between history and doing philosophy. We can now give a more precise sense to what I have so far called the “anachronism” in Deleuze's historical texts. What is happening is that particular concepts are being raised on another plane: a plane other that the one on which they were originally constructed.

Which concepts can function on other planes? What mutations does the concept undergo—which components must be added or removed—when a concept is reconstituted on another plane? What, for instance, happens to the subject when it is constructed on the terrain of a radical empiricism? Perhaps the best way to address these questions is to take up our example again and follow it through a little further. Deleuze reads Hume in such a way as to find there a theory of the formation of the subject inside the given. We now ask—why this concept and why Hume? What is at stake in the articulation of an eighteenth century empiricist concept of the subject?

We can glean something of Deleuze's intentions here from his short essay “A Philosophical Concept.”³⁵ The essay begins with his now familiar critique of scholarly critique. Philosophical concepts fulfill functions within fields of thought (planes) that are themselves at least partly functions of social and historical circumstance. Concepts do not die, in the sense that they are proven “wrong.” Rather they are made redundant when their function is discharged by new concepts in new fields. And “it is never very interesting to criticize a concept.” It is far more useful to “build the new functions and discover the new fields which make it useless or inadequate.”

Historically Deleuze thinks that the concept of the subject has fulfilled two functions. The first is universalization. With the Enlightenment and the rise of secular philosophy the universal “was no longer represented by objective essentials but by deeds, noetic or linguistic.” Hume is identified as a key moment in a philosophy of the subject because he raises the question of how it is that we form beliefs which transcend the data given in experience (particularly when we judge some state of affairs to be necessary). Hume transforms epistemology by reconfiguring the field of knowledge in terms of “belief.” He is seminal in advancing a distinctively modern inquiry into the principles that determine the formation of beliefs and the conditions under which belief is legitimate. Hume's significance here concerns his contribution in laying out the rights and conditions under which the particular and limited subject makes universal assertions.

The second function of the subject is the function of “individualization”—“in a field where the individual can no longer be a thing or a soul, but a person alive and felt, speaking and spoken to (I-you).” Deleuze asks if these two aspects of the subject, the universal “r” and the individual “me,” are necessarily linked. “Even if they are, isn't there a conflict between them and how could it be solved?” These questions actuate the philosophy of the subject and Deleuze attaches these questions to the names Hume, Kant, and Husserl.

What functions will displace those of universalization and individualization? Functions of “singularization” (singularities in the mathematical sense) replace universalization. Deleuze anticipates a new field in which there is no use for the universal. This is a transcendental field without subject. In the place of “knowledge” and “belief” we have notions of “arrangement” or “contrivance,” “which indicate an emission and a distribution of singularities.” This transcendental field is composed of dice rolls, events and their arrangements, and configurations a kind of philosophical craps table perhaps. Importantly, in this field, multiplicities are substantive in themselves. Multiplicities or assemblages do not refer to a subject as a preliminary unity. As for the function of individuation we find, on this field, new types of impersonal individuation. Individuations that no longer constitute persons or “egos” Deleuze calls “ecceities” or “hecceities”; and indeed we are ecceities and not egos. Addressing a French audience, Deleuze suggests that Anglo-American philosophy and literature are particularly interesting on this point “because they are conspicuous for their inability to find a sense to give to the word ‘me’ but that of a grammatical fiction.”³⁶

In short we believe that the notion of the subject has lost much of its interest on behalf of pre-individual singularities and non-personal individuations.³⁷

We have already noted Deleuze's belief that philosophy must find new modes of expression. It is in regard to the development of non-personal individuations that, for Deleuze, Anglo-American literature (Joyce, Beckett, et al.) is philosophically revolutionary (and so a revolutionary philosophy must become more literary). The subject becomes redundant as a philosophical concept when we possess the new functions and the new fields which make it redundant. For Deleuze, it becomes possible to speak of a plane of immanence when immanence is no longer immanent to something other than itself. The power of the plane of empiricism is that it presents the given as pre-subjective. A radical empiricism “does not present a flux of the lived that is immanent to a subject and individualized in that which belongs to a self.”³⁸ To appreciate the effect of dispensing with the subject; to understand what a radical empiricist plane does, we need to examine why the concept of the subject is, at root, a moral concept.

RECOGNITION AND THE MORAL IMAGE OF THOUGHT

The “plane of immanence” is the “image of thought”; the ground or foundation (deterritorialisation) on which philosophy creates its concepts. In Difference and Repetition the discussion of “the image of thought” develops the Nietzschean idea that, historically, the image of thought is a moral image. The essential element of the dogmatic or moral image of thought consists “of the presupposition that there is a natural capacity for thought endowed with a talent for the truth or an affinity with the true, under the double aspect of a good will on the part of the thinker and an upright nature on the part of thought.”³⁹ Canonical philosophy from Plato to Kant has been subject to this Image; and insofar as thought remains subject to this Image it does not ultimately matter whether philosophy begins with the subject or the object, with Being or with beings. The distribution ofthe object and the subject, as well as that of Being and beings, is prejudged in the Image. If the most general presuppositions of philosophy are essentially moral then a philosophy which would be without presuppositions begins with a radical critique of this Image. The question we need to take up then is why Deleuze thinks that the overthrow of the orthodox Image of thought requires the instituting of a new plane; a transcendental field without subject.

Descartes begins the Discourse on the Method with the claim that “Good sense is of all things in the world the most equally distributed.”⁴⁰ It is Descartes who explicitly constructs an image of thought as it is in principle. “Natural good sense or common sense are thus taken to be determinations of pure thought.” To the extent that this Image holds in principle it is not a matter of assessing the evidence that supports this presupposition. Rather the Image presupposes a “certain distribution of the empirical and the transcendental” and so it is the transcendental model implied by the image that must be judged.

According to Deleuze the transcendental model implied in the Image of thought is that of recognition. Recognition is defined as the “harmonious exercise of all the faculties upon a supposed same object.” The same object may be seen, touched, remembered, imagined or conceived. Although the nature ofthe given is different for each faculty (perception, memory, imagination, understanding) an object is recognized when one faculty locates it as identical to that of another. The “recognition” of an object means that the relationship of the faculties to their given and to each other is determined by a form of identity in the object. The idea of “common sense” is the presumption that this “subjective principle of collaboration of the faculties” holds for everybody. Importantly, Deleuze points out that the form of identity in objects is grounded in the unity of the thinking subject.

Suppose that the activities of the faculties were taken to be substantially independent of each other. That is, suppose that the agreements that we find between the faculties were not taken to be necessarily determined by the unity of the object on which they are collectively exercised. Then it would appear equally sensible to assume that the alignments that are established between the faculties and their respective givens are just that—established alignments between distinct faculties—and it is by no means obvious that the faculties have to be coordinated according to the model of a universal common sense. The model of recognition (as the transcendental model contained in the Image of thought) has to ground the identity of the object on the identity of the thinking subject. If, on the contrary, the faculties are a genuine multiplicity then relations between the faculties could not be structured according to the model of recognition. The form of identity in objects has to be grounded in the unity of the thinking subject, such that the faculties are simply modalities of that unified subject.

This is the meaning of the Cogito as a beginning: it expresses the unity of all the faculties in the subject; it thereby expresses the possibility that all the faculties will relate to a form of object which reflects the subjective identity; it provides a philosophical concept for the presupposition of a common sense; it is the common sense become philosophical.”⁴¹

We can now get at the meaning of the claim that the concept of the subject has served the function of universalization. Deleuze identifies the “I think” as the most general principle of representation; necessarily the same for everyone it grounds the commonality of common sense.

We have noted that in What Is Philosophy? the “plane of immanence” is equated with “the image of thought.” In this later text however, a key concern of the discussion is to reveal the ways in which transcendence is always reintroduced in philosophy. The transcendent is always revived whenever immanence is interpreted as immanent “to” something. This results in a confusion of plane and concept “so that the concept becomes a transcendent universal and the plane becomes an attribute in the concept.” The introduction of a transcendent universal reduces the immanent to a “simple field of phenomena that now only possesses in a secondary way that which is first of all attributed to the transcendental unity.”⁴² With this claim in mind we can see how Deleuze interprets the respective contributions of Hume and Kant to a philosophy of the subject.

The classical image of thought presents thought as willing the truth, as animated by a natural desire to turn towards the light. Knowledge is a matter of grasping the truth: of perceiving and apprehending what is visible by this divine light. The worst that can happen to thought on this plane is error: losing one's way, losing the way. Hume is the key persona in the construction of another plane characteristic of the Enlightenment. Hume substitutes belief for knowledge and in so doing institutes another plane, another image of thought. Once what is given to thought by right is not knowledge but belief then thought is no longer a matter of turning towards the Truth, but rather of “following tracks,” of inference rather than grasping. The empiricist concepts of association, relation, habit, probability, and convention determine the criteria for the legitimacy of beliefs. Thought becomes secular and religious belief is transformed into a particular, and for the most part illegitimate, case of inference. On this plane, thought's primary danger is not error but illusion.⁴³

Of Hume's principles of human nature, the principles that govern the formation of the subject within the given, habit is the principle which makes of the past a rule for the future. Consider the classic Humean problem of causation. Causation, for Hume, is nothing other than the experience of constant conjunction and the determination of my expectation for the future on the basis of that experience. Every time A has appeared B has followed it, so we say that A causes B. Experience directs us to observe particular conjunctions. The essence of experience is the repetition of similar cases and causality is the effect of experience. But experience alone cannot explain the expectation. All repetitions are logically independent of each other. No repetition, nor any number of repetitions, can ever, of itself, guarantee any future repetitions. This disposition to form an expectation on the basis of experience, depends on experience but cannot be a product of experience alone. The movement to transcend the given of experience and form a belief about some future event must be determined by another principle—and that principle is habit.⁴⁴ Because in habit we transcend past experience towards an anticipated repetition there is a sense in which habit necessarily falsifies experience. It is not a matter of deciding what inferences are true so much as what inferences are legitimate, reliable, probable, useful. On the Humean plane thought's permanent danger is the contraction of bad habits.

Kant inherits many of these features from Hume, but he transforms them “on a new plane or according to another image”⁴⁵. “The distribution of thought's rights and dangers has changed. Already in Hume thought's primary danger is not error so much as illusion. But illusions are still like traps laid on thought's path; tempting but dangerous inferences which may mislead superstitious, prejudiced or untrained minds but which a temperate and scientific man may hope to avoid. With Kant the delirium which penetrates thought becomes far more profound. Inevitable illusions come from within reason “as if from an internal arctic zone where the compass of every needle goes mad.”⁴⁶

In fact, Kant possessed the resources to overturn the Image of thought. Instead of errors that are attributed to the body and its imperfect senses, Kant finds that there are irremediable illusions that are interior to reason itself. But the Kantian Critique remains ultimately respectful of the Image of thought. In Kant the Image confronts its dangers and is restored with new certainty. In the unity of the “I think” the respective contributions of the thinking faculties are harmonized according to the model of recognition.⁴⁷ It is with Kant that the subject achieves its modern role as the transcendent unity which endows the phenomenal with its universal properties (a movement that is completed with Husserl).⁴⁸

We can now see how Deleuze assesses the consequences of Kant's transformation of the plane of immanence. The extraordinary resources of empiricism for the creation of concepts, which Hume discovers, are overthrown by Kant who restores the orthodox Image of thought with the transcendental subject. The Humean subject is a concept that inhabits the plane of immanence. It is not the subject that explains the given and thought but rather the given and the (transcendent) principles of association that explain the subject. The Humean subject is merely an effect of repetition, a habit, “nothing but the habit of saying ‘I’.”⁴⁹ The Humean mind is a theater comprised of nothing more than the perceptions that appear there. It is not a unity nor a substance nor an ego but an organized collection. For Hume, the mind's faculties are substantially independent and enjoy no predetermined harmony. Rather there is a fundamental conflict (at the level of principle) between the imagination and the understanding. In the Humean system fiction is not simply the unreal ideas of the imagination. Fiction becomes a principle of human nature and a condition of the constituted subject.⁵⁰Deleuze finds an intriguing affinity between a Humean conception of fiction and a Nietzschean appreciation of style. And, very importantly for Deleuze, Hume creates the first great logic of relations, showing in it that all relations are external to their terms. Deleuze calls this thesis “associationism” and makes it the core doctrine of empiricism. (Associationism is very important for Deleuze and Guattari's “rhizomatic” model of thought.) With the Kantian transcendental subject all of these resources are lost.

We began this essay with a question about how Deleuze does history of philosophy; or better, how he uses his history of philosophy. Deleuze identifies the subject as the stronghold of transcendence in modern philosophy. His attack on this stronghold involves the creation of a Nietzschean genealogy of the subject. Deleuze finds that the transcendental subject was erected over the top of a more heretical empiricist subject. Latent in the Humean subject are dangerous and anti-Christian implications that threaten the moral Image of thought. Deleuze gives us a history of philosophy that unsettles the transcendental subject at its origins.

Notes

1. Gilles Deleuze and Felix Guattari, What Is Philosophy?, trans. Hugh Tomlinson and Graham Burchill (New York: Verso, 1994), p. 83.

2. Deleuze, Gilles, Difference and Repetition, trans. Paul Patton, (New York: Columbia University Press, 1994), p. xv.

3. Ibid., p. xxi.

4. Jorge Luis Borges, “Pierre Menard, Author of the Quixote,” in Labyrinths, edited by Donald A. Yates and James E. Irby (Penguin, 1970), pp. 62-71.

   There is a fascinating anecdote about the writing of this story which adds a certain irony to Deleuze's reference to it. In The Western Canon Harold Bloom discusses how Borges' characteristic style evolved:

   The crossing point was a terrible accident he suffered near the close of 1938. Always afflicted by poor eyesight, he slipped on a badly lit staircase and fell, sustaining a severe head injury. Seriously ill for two weeks in the hospital, he had fearful nightmares and then a painfully slow comvalescence, in which he began to doubt his mental condition and his ability to write. And so, at thirty-nine, he tried to compose a story to reassure himself. The hilarious consequence was Pierre Menard, Author of the Quixote,” the forerunner of “Tlon, Uqbar, Orbis Tertius” and all the other brilliant short fictions with which we associate his name.

   Harold Bloom, The Western Canon (New York: Harcourt, Brace and Co., 1994), p. 464.

5. Borges, “Pierre Menard,” p. 66.

6. Ibid., pp. 69-70.

7. Difference and Repetition, p. xxii.

8. Gilles Deleuze, Empiricism and Subjectivity: An Essay on Hume's Theory of Human Nature, trans. Constantin Boundas (New York: Columbia University Press, 1991).

9. Difference and Repetition, p. xx.

10. Empiricism and Subjectivity, pp. 105-07.

11. Ibid., p. 107. Atp. 105 Deleuze writes: we must distrust the objections often raised against Hume's empiricism. We shouldn't, of course, present Hume as an exceptional victim, who more than others has felt the unfairness of constant criticisms. The case is similar for all great philosophers. We are surprised by the objections constantly raised against Descartes, Kant, Hegel, etc. Let us say that philosophical objections are of two kinds. Most are philosophical in name only, to the extent that they are criticisms of the theory without any consideration of the nature of the problem to which the theory is responding, or the problem which provides the theory with its foundation or structure.

12. David Hume, A Treatise of Human Nature, edited by L. A. Selby-Bigge (Oxford, Clarendon Press, 1978), p. 253. Discussed in Deleuze, Empiricism and Subjectivity, p. 23.

13. “For Kant, relations depend on the nature of things in the sense that, as phenomena, things presuppose a synthesis whose source is the same as the source of relations. This is why the critical philosophy is not an empiricism” (Empiricism and Subjectivity, p. 111).

14. Ibid., pp. 111-12.

15. The following passage, which occurs in What Is Philosophy? (pp. 47-48), is particularly instructive in this context: When immanence is no longer immanent to something other than itself it is possible to speak of a plane of immanence. Such a plane is, perhaps, a radical empiricism: it does not present a flux of the lived that is immanent to a subject and individualized in that which belongs to a self. It presents only events, that is, possible worlds as concepts and other people as expressions of possible worlds or conceptual personae. The event does not relate the lived to a transcendent subject = Self but, on the contrary, is related to the immanent survey of a field without subject; the Other Person does not restore transcendence to an other self but returns every other self to the immanence of the field surveyed. Empiricism knows only events and other people and is therefore a great creator of concepts. Its force begins from the moment it defines the subject: a habitus, a habit, nothing but the habit of saying I.

16. Ibid.

17. Ibid., p. 133.

18. Ibid., p. 26.

19. See Empiricism and Subjectivity, pp. 66-69 and Difference and Repetition, pp. 70-85. In Empiricism and Subjectivity we find the following: It is not necessary to force the texts to find in the habit-anticipation most of the characteristics of the Bergsonian duree or memory. (p. 92)

20. Difference and Repetition, p. 72.

21. What Is Philosophy?, p. 2.

22. Ibid., p. 5.

23. Ibid., p. 27.

24. Ibid., pp. 28-29

25. Ibid., p. 63.

26. Ibid., p. 24. Deleuze's example here is the Cartesian cogito.

27. Ibid., p. 8.

28. The following well known aphorism from Nietzsche gives good support to this reading: One thing is neto “give style” to character-great and rare art! It is practiced by those who survey the strengths and weaknesses of their nature and then fit them into an artistic plan until every one of them appears as art and reason and even weaknesses delight the eye. Here a large mass of second nature has been added; there a piece of original nature has been removed-both times through long practice and daily work at it. Here the ugly that could not be removed is concealed; there it has been reinterpreted and made sublime. Much that is vague and resisting shaping has been saved and exploited for distant views; it is meant to beckon towards the far and immeasurable. In the end, when the work is finished, it becomes evident how the constraint of a single taste governed and formed everything large and small. Whether this taste was good or bad is less important than one might suppose, if only it was a single taste! The Gay Science, no. 290.

29. On the significance of philosophical taste and instinct, see What Is Philosophy?, pp. 77-79. The analysis of “conceptual personae” in part 1, chapter 3 of What Is Philosophy? is quite significant for the interpretation of Deleuze-s historiographical methods advanced here. The conceptual personae is the subject of a philosophy “on a par with the philosopher”.

30. Ibid., p. 8.

31. The following quoted phrases are found in What Is Philosophy? pp. 36-37

32. Ibid., p. 37.

33. Ibid,. p. 40.

34. Ibid., p. 41.

35. Gilles Deleuze, “A Philosophical Concept …”, Topoi 7 (September 1988): 111-12. This was the edition of Topoi in which a number of thinkers were invited by Jean-Luc Nancy to answer the question “Who comes after the subject?”

36. Ibid., p. 11 I.

37. Ibid,. p. 112.

38. What Is Philosophy?, p. 47.

39. Difference and Repetition, p. 131.

40. Ibid., p. 132.

41. Ibid., p. 133.

42. What Is Philosophy?, pp. 4445.

43. Empiricism and Subjectivity, p. ix.

44. Deleuze seizes on the Humean principle of habit as the principle which explains the formation of the subject within the given. Habit is a non-mechanical repetition: a repetition which does something. Habit is the principle which explains how something new is extracted from repetition-and ultimately how the subject is formed as an effect of repetition. The relevant discussion here is found at the beginning of chapter 2 of Difference and Repetition. There Deleuze analyses the function of repetition in the synthesis of Time. That is, of subjective time lived as a perpetual present. As a synthesis of the succession of past instants, a synthesis of experience, the present contains the past in what Deleuze calls a “contraction”. The future is given in this same contraction as expectation.

45. What Is Philosophy?, p. 54.

46. Ibid., p. 52.

47. Ibid., p. 135.

48. “Beginning with Descartes, and then with Kant and Husserl, the cogito makes it possible to treat the plane of immanence as a field of consciousness. Immanence is supposed to be immanent to a pure consciousness, to a thinking subject. Kant will call this subject transcendental, rather than transcendent, precisely because it is the subject of the field of immanence of all possible experience from which nothing, the internal as well as the external, escapes. Kant objects to any transcendent use of the synthesis, but he ascribes immanence to the subject of the synthesis as new, subjective unity. He may even allow himself the luxury of denouncing transcendent Ideas, so as to make them the “horizon” of the field immanent to the subject. But, in so doing, Kant discovers the modern way of saving transcendence: this is no longer the transcendence of a Something, or of a One higher than everything (contemplation), but that of a Subject to which the field of immanence is only attributed by belonging to a self that necessarily represents such a subject to itself (reflection)” (What Is Philosophy?, p. 46).

49. The idea that the subject is a cluster of habits is very important for Deleuze. In the “Preface to the English-Language Edition” of Empiricism and Subjectivity he writes that “perhaps there is no more striking answer to the problem of the Self’ (p. x). There is ample material for a study which discussed the development of the concept of habit in Deleuze's works from Difference and Repetition on.

50. See the conclusion to Empiricism and Subjectivity (“Purposiveness’); esp. pp. 130-33.