Daniel W. Smith “Deleuze, Kant, and the theory of immanent ideas”

Smith, Daniel W. 2006. Deleuze, Kant, and the theory of immanent ideas. In: Boundas, Constantin V. (ed.). Deleuze and Philosophy. Edinburgh: Edinburgh University Press, 43–61.

In using the term ‘Idea’, Deleuze is not referring to the common-sense use of the term, or the use to which empiricists like Hume or Locke put it, for whom the word ‘idea’ refers primarily to mental representations. Rather, Deleuze is referring to the concept of the Idea that was first proposed by Plato, and the modified by Kant and Hegel. (43)

For Deleuze, Plato created the concept of the Idea in order to provide a criterion to distinguish between (or to ‘select’ between) things and their simulacra – for instance, between Socrates (the true philosopher) and the sophists (the simulacral counterfeits). If Plato failed in his project, it is because he assigned his Ideas a transcendent status. But Deleuze will more or less take up Plato’s project anew in order to rejuvenate it: Ideas, he argues, must be made immanent, and therefore differential. Yet this was already Kant’s project: in fascinating text at the opening of the ‘Transcendental Dialectic’, Kant criticises Plato for assigning to Ideas a ‘transcendent object’ – even though Kant is in the process of justifying his own appropriation of the Platonic concept of the ‘Ideas’. (43–44)

In the first critique, Kant distinguishes between three types of concepts: empirical concepts, a priori concepts or ‘categories’, and Ideas. Empirical concepts are concepts like ‘table’ and ‘rose’ that give us genuine knowledge. Such concepts are applied to a multiplicity (or manifold) of sensations: through the imagination I synthesise these perceptions, and in applying the concept ‘table’ to them, I can recognise the object before me. But Kant also identifies a second type of concept, which are a priori concepts or what Kant (following Aristotle) calls ‘categories’. Categories are concepts that are applicable, not just to empirical objects such as tables and chairs, but to any object I could ever come across, ever, for all time, in my experience. […] the concept of ‘cause’ is a category, precisely because I cannot conceive of an object that does not have a cause. […] Categories are thus a priori concepts that are applicable to every object of any possible experience. Indeed, Kant’s notion of ‘possible experience’ is derived from his notion of the category: it is the categories that define the domain of possible experience. (45)

Finally, there is a third type of concept in the Critique of Pure Reason, which Kant calls ‘Ideas’ (in the Platonic sense). An idea is the concept of an object that goes beyond or transcends any possible experience. There are various kinds of transcendent concepts: for instance, anytime we speak of something ‘pure’ or ‘absolute’ – for example, the ‘pure gift’ in Derrida, or ‘absolute zero’ in physics – we are almost certainly outside the realm of possible experience, since experience presents us with impure mixtures and non-absolutes. (45)

Kant […] famously focuses on three transcendent Ideas, which constitute the three great terminal points of metaphysics: the soul, the world and God. These Ideas go beyond any possible experience: there is no object – anywhere, ever – that could correspond to such Ideas; we can never have a ‘possible experience’ of them. (46)

To use the famous Kantian distinction, we can think the world as if it were real, as if it were an object, but we can never know it. Strictly speaking, the world is not an object of our experience; what we actually know is the problematic of causality, a series of causal relations that we can extend indefinitely. It is this problem, Kant says, that is the true object of the Idea of the world. (46)

[…] ‘transcendent’ and ‘transcendental’ are not identical terms, and in fact are opposed to each other. The aim of Kant’s transcendental project is to discover criteria immanent to the understanding that are capable of distinguishing between two different uses of the syntheses of consciousness: legitimate immanent uses, and illegitimate transcendent uses (the transcendent Ideas). Transcendental philosophy is a philosophy of immanence, and implies a ruthless critique of transcendence (which is why Deleuze does not hesitate to align himself with Kant’s critical philosophy, despite their obvious differences). (47)

This is also why Kant can assign to Ideas a legitimate immanent use as well as an illegitimate transcendent use. The immanent use is regulative: ideas constitute ideal focal points or horizons outside experience that posit the unity of our conceptual knowledge as a problem; they can therefore help regulate the systematisation of our scientific knowledge in a purely immanent manner. The illegitimate transcendent use is falsely constitutive: it falsely posits or constitutes an object that supposedly corresponds to the problem. At best, reason can simply postulate a harmony or (in Kant’s terminology) an ‘analogy’ between its Ideas and the material objects of experience: Reason here is the faculty that says, ‘Everything happens as if …’ (as if there were a world, or a soul, or a God …). (47)

Put simply, whereas Kantian Ideas are unifying, totalising and conditioning (transcendent Ideas), for Deleuze they will become multiple, differential, and genetic (immanent Ideas). (48)

Maimon’s primary objection is that Kant ignored the demands of a genetic method, by which Maimon means two things. First, Kant simply assumed that there were a priori ‘facts’ of reason (the ‘fact’ of knowledge in the first critique, and the ‘fact’ of morality in the second critique) and then sought the ‘condition of possibility’ of these facts in the transcendental. Against Kant, Maimon argues that one cannot simply assume these supposed ‘facts’ of knowledge or morality, but must instead show how they are engendered immanently from reason alone as the necessary modes of its manifestation. In short, a method of genesis has to replace the Kantian method of conditioning. (49)

Second, Maimon says, such a genetic method would require the positing of a principle of difference in order to function: whereas identity is the condition of possibility of thought in general, he argues, it is difference that constitutes the genetic condition of real thought. (49)

[…] one could say that these two exigencies are the primary components of Deleuze’s ‘transcendental empiricism’. (49)

There are two ways of overcoming the concept-intuition duality in Kant: either concepts are sensible things, as in Locke; or sensibility itself is intelligible, as in Leibniz (there are Ideas in sensibility itself). In effect, Deleuze takes this latter path. (51)

Two of these components of Deleuzian Ideas are worth mentioning here: the differential relation and the notion of singularities. The differential relation is a relation that persists even when the terms of the relation have vanished. It is thus a pure relation, a pure relation of difference; it is what Deleuze means by ‘difference-in-itself’. Moreover, not only is the differential relation external to its terms, it is also constitutive of its terms: the terms of the relation are completely undetermined (or virtual) until they enter into the differential relation; on their own they are simply determinable. Once such elements enter a differential relation, their reciprocal determination determines a singularity, a singular point. Every multiplicity (that is, every thing) is characterised by a combination of singular and ordinary points. (51)

For Deleuze, this is exactly how our life is composed or constructed – from singularity to singularity. The point where someone breaks down in tears, for example, or boils over in anger is a singular point in someone’s psychic multiplicity, surrounded by a swarm of ordinary points. (51)

One could say that these are the two poles of Deleuze’s philosophy: ‘Everything is ordinary!’ and ‘Everything is singular!’6 Your reading of this paper, here and now, is a singular moment, never to be repeated; but at the same time it is a completely ordinary event. Yet there is a complex temporality at work in Deleuze’s theory of Ideas; and there is a complex temporality at work here as well: your reading of this paper may be ordinary, yet in retrospect it may appear singular because, perhaps, it changes the way you think, or sparks an unrelated idea in you that brings you lasting fame and fortune. We never know such things in advance, of course. (52)

[…] Deleuze’s theory of Ideas is an attempt to answer Plato’s question: what is a thing, what is its essence? His answer, put briefly, is that every thing is a multiplicity, which unfolds and becomes within its own spatiotemporal co-ordinates (its own ‘internal metrics’), in perpetual relation with other multiplicities. (52)

[…] the fundamental questions Deleuze links with Ideas are questions such as: how?, where?, when?, how many?, from what viewpoint?, and so on – which are no longer questions of essence, in the old sense, but questions of becoming, of the event (although Deleuze himself does not hesitate to use the term ‘essence’). (52)

First, the elements of an Idea are completely undetermined (or virtual); second, these elements are nonetheless determinable reciprocally in a differential relation (dx/dy); and third, to this reciprocal determination there corresponds the complete determination of a set of singularities (values of dx/dy), which defines a multiplicity (along with their prolongation in a series of ordinary points). (52)

Kant’s definition of desire is extraordinary: desire is ‘a faculty which by means of its representations is the cause of the actuality of the objects of those representations’ (Kant 1952). It is hard to overemphasise the importance of this Kantian definition for Deleuze, since it breaks with a long tradition in philosophy that defined desire in terms of lack. Desire, says Kant, is a faculty that, given a representation in my head, is capable of producing the object that corresponds to it. (53)

The fundamental thesis of Anti-Oedipus is a stronger variant of Kant’s claim: ‘If desire produces, its product is real. If desire is productive, it can be productive only in the real world and can produce only reality’ (Deleuze and Guattari 1977: 26). (54)

What would a purely immanent theory of desire look like in the domain of practical reason? What if one did not appeal to the moral law – and the transcendent Ideas that serve as its necessary postulates – and instead synthesised desire with a conception of Ideas as purely immanent? This is precisely what Deleuze does in the opening two chapters of Anti-Oedipus: the three syntheses by which he and Guattari define ‘desiring machines’ are in fact the same three Ideas that Kant defines as the postulates of practical reason (soul, world and God), but now stripped entirely of their transcendent status, to the point where neither God, world, nor self subsist. (55)

Generally speaking, Deleuze gives a purely immanent characterisation of the three syntheses – connection [world], conjunction [self], and disjunction [God] – and then shows how desire itself is constituted by tracing out series and trajectories following these syntheses within a given social assemblage. (55)

The new dialectic (theory of Ideas) whose formal components Deleuze develops in Difference and Repetition, could be said to receive its practical determination in Anti-Oedipus – with the difference that, in Deleuze, the determinations of the Ideas are practical from the start (hence the importance of such questions as: how?, where?, when?, how many?, from what viewpoint?, and so on). (56)

Deleuze suggests that philosophers should start from the obscure: a clear perception emerges from the obscure (or the virtual) by means of a genetic process (the differential mechanism). Yet at the same time, my clear perceptions are constantly plunging back into the obscure, into the virtual Idea of minute perceptions: by its very nature, perception is clear and obscure (chiaroscuro). (58)

To say that perception is by nature clear-obscure (chiaroscuro) is to say that it is made and unmade at every moment, in all directions, constantly extracting the clear while constantly plunging back into the obscure. (58)

Even in such a simple example, making a decision is never a question of choosing between x (staying in) and y (going out), since both inclinations are multiplicities that include an unconscious complex of auditory, gustative, olfactory and visual perceptions – an entire ‘perceptio-inclinatory ensemble’, a multiplicity of minute perceptions, minute tendencies and inclinations. It is this state of constant disquiet that Locke called ‘uneasiness’, and which Leibniz, in the New Essays, termed the fuscum subnigrum, the dark background of the soul. (59)

The aim of presenting these examples from lived experience is to demonstrate that, despite the abstract Nature of his language, Deleuze is attempting to get at something concrete with his theory of immanent Ideas. Immanent Ideas do not constitute the condition of possible experience, but rather the genetic conditions of real experience. ‘This is what it’s like on the plane of immanence,’ Deleuze writes, ‘multiplicities fill it, singularities connect with one another, processes or becomings unfold, intensities rise and fall. . . .’ (Deleuze 1997: 146–7). (59)