The final paragraphs of this introductory section extend and develop the consequences of Deleuze's repression-repetition reversal. "I do not repeat because I repress," he writes, "I repress because I repeat, I forget because I repeat" (DR 18). What does this mean?
The normal formula—that I first repress sensory material that my psyche cannot handle; that that material is recorded in my unconscious (or as my unconscious); that that which I repress will return again and again in odd, compulsive behaviors—treats repetition a priori as a problem in need of solution and thus leaves its genetic relation to the mechanisms of repression unexplained (18). In other words, to approach compulsive behavior as a mere effect of repression does not really explain why one would have to repress sensory material in the first place. If repetition were already a primary mode of being or living, however, then this problem of repression's origin appears less urgent, for repression (at least in principle) would become a mechanism that follows from a kind of arche-repetition. In reversing Freud's formula, Deleuze hopes to carry on what Freud himself anticipates in Beyond the Pleasure Principle: to establish repetition itself as a primary mode of being, living, creating, repressing, forgetting, and thinking and, furthermore, to earn a philosophical explanation of these activities as effects of a rigorously conceptualized repetition (and difference).
In these particular paragraphs, Deleuze deviates from Freud's speculations, framing the death drive ("Thanatos") not as an instinctive tendency or desire to return to a simpler, earlier, material state of things but rather as the radical groundlessness of substance or self that serves as the condition of all repetitions or compulsions. This groundlessness "furnishe[s]" "roles and masks" for repetition and, moreover, "submits Eros"—the work of pleasure-making, fantasy-building, and game-creating—"to repetition" itself (DR 18). Wait, what? Eros? Thanatos? Groundlessness?
Although I tried to unfold some of this in my last post, it's worth dealing with this conceptual cluster again. When I need a bit of help straightening out the intracacies of Deleuze's interventions, I tend to trust Daniel W. Smith. In his essay, "The Concept of the Simulacrum: Deleuze and the Overturning of Platonism," he makes interesting use of these early pages in Difference and Repetition in order to flesh out a distinction between Deleuze's metaphysics and the pervasiveness of (a version of) Platonic idealism:
Platonism [like Freud's theory of the death drive] relies on what Deleuze calls a "naked" model of repetition (representation): the copy repeats the identity of the ideal model as the first term in a hierarchical series [this is the sense of "bare" repetition that Freud cannot think his way around; see SR 1.17]. Naked repetition thus presupposes a mechanical or brute repetition of the Same: it is founded on an ultimate or originary instance or first time (A), which is then repeated a second, third, and fourth time (A1, A2, A3, and so on.). In cases of psychic repetition, this originary term is subject to disguises and displacements, which are secondary yet necessary. In Freud, for instance, our adult loves "repeat" our childhood love for the mother, but our original maternal love is repressed and disguised in these subsequent loves by various mechanisms of condensation (metonymy) and displacement (metaphor). I repeat because I repress (amnesia), and the task of therapy, through transference, is to recover this hidden origin (not to eliminate repetition, but to verify the authentic repetitions). (Essays on Deleuze 21-22)
This is a nice summation of the slow reading in my last post and relates in a surprisingly resonant way to Jacques Derrida's reading of Jean-Jacques Rousseau in the Grammatology (I also suggested this connection in my last post). We'll get to Derrida in just a moment, though. First, back to Smith:
[But in] Proust's novel In Search of Lost Time, the hero's various loves [. . .] indeed form a series [. . . but] this series of loves does not refer back to the hero's mother; the childhood love for his mother is already a repetition of other adult loves [. . .] In other words, there is no first term in what is repeated that can be isolated from the series. My parents are not the ultimate terms of my individual subjectivity, but rather the middle terms [like all terms] of a much larger intersubjectivity [what Deleuze and Guattari would later call an aggregate or an assemblage]. (Essays on Deleuze 22)
It is instructive to look at Derrida here as an analogue to Deleuze. In the famous chapter of the Grammatology, "...That Dangerous Supplement...," Derrida isolates several instances in which Jean-Jacques Rousseau relies upon a logic of supplementarity to theorize some sort of natural/cultural or natural/artificial binary. The most famous of these binaries is, of course, speech and writing, but the second section of this chapter—which chains various binaries together—takes up the issue of mothers and lovers and masturbation. In the final paragraphs of this section, Derrida writes,
Jean-Jacques could [. . .] look for a supplement to Thérèse [in onanism] only on one condition: that the system of supplementarity in general be already open in its possibiity, that the play of substitutions be already operative for a long time and that in a certain way Thérèse herself be already a supplement [just a]s Mamma was already the supplement of an unknown mother [. . .] Through this sequence of supplements a necessity is announced: that of an infinite chain, ineluctably multiplying the supplementary mediations that produce the sense of the very thing they defer: the mirage of the thing itself, of immediate presence, of originary perception. Immediacy is derived. (156-57)
Despite all the crucial differences we might identify between Deleuze's project (which establishes a strict division between substitution and modes of repetition) and Derrida's, the resonance between a Derridean chain and a Deleuzian series should be noted. For both philosophers, their respective theories imply a primary condition for repetition that challenges conceptual paradigms that insist on fixed identities, substances, individuals, psyches, intentions, meanings, relations, or systems as their condition of analysis or empirical study. Derrida and Deleuze do not simply insist on the possibility of deviations or exceptions to "the way we think things are." Rather, they make ontological arguments that "the way things are" is itself always already on the move (often behind our backs!). For Derrida, we can best approximate being as differance: the dual (or supplemental) operation of  the establishment of an identity or meaning as "a difference" and  the deferment of that difference's self-identity or -meaning. In the case of Rousseau, each substitute for an "original" or "more natural" love (Mamma for an archetypal mother; Thérèse for Mamma; onanism for Thérèse) establishes a new source of pleasure felt as a separate presence even as it simultaneously defers or renders inaccessible the fullness of that pleasure's full presence. More importantly, each supplement establishes that there was never anything but a chain of supplemental pleasures (traces without origin, primary source, or full presence). Immediate, true, and lasting love or intimacy is always already derived.
For Deleuze, being appears to be a dance of difference and repetition that has less to do with the substitution of traces (which are falsely experienced as presences) and much more to do with the conscious or unconscious selection and construction of activities, behaviors, pleasures, ideas, styles, tastes, etc. It matters far less to Deleuze whether or not things are present . . . what matters is that the activity of things constitutes the repetition of something, carrying on a difference, a singularity that can only be carried on by a mode of repetition. Deleuze's reading of Proust relates to Derrida's reading of Rousseau insofar as neither of them find an of originary love in these respective texts; however, in Deleuze's Proust the point is less one of distinct derivations along a chain of supplements and more, as he puts it in Proust and Signs, of a "personal series of our loves [that] refers both to a vaster, transpersonal series and to more restricted series constituted by each love in particular. The series are thus implicated within each other, the indices of variation and the laws of progression enveloped within each other" (72). As seemingly resonant as Derrida and Deleuze are, then, Deleuze's sense of series (and series of series!) seems far more fractal, lateral, frayed, and non-sequential than Derrida's chain (even if Derrida, elsewhere, rejects the sequentiality of differance).
But let's get out of this rabbit hole. It's useful, before moving on to the last paragraph of this section, to look at Smith's own conclusion:
What is the "content" that is being affected or modified within these series [of loves]? In clothed repetition [as opposed to the bare repetitions of Plato or Freud], what is repeated is not a prior identity or originary sameness, but rather a virtual object or event [. . .] There is indeed, one might say, an "essence" that governs the series of loves, but this essence, Deleuze insists, "is always difference," and this difference differs from itself every time it is repeated. [It seems to me that the retainment of a virtual "essence" in Deleuze's metaphysics would make Derrida uneasy. But moving on...] [That which is repeated] is a virtuality [what Deleuze has been calling "a singularity" throughout this introduction] that is differenciated every time it is actualized. The variations [of this essence], in other words, do not come from without, but express differential mechanisms which belong to the essence and origin of what is repeated. There is not an orginary "thing" (model) which could eventually be uncovered behind the disguises, displacements, and illusions of repetition (copies); rather, disguise and displacement are the essence of repetition itself, which is in itself an original and positive principle. (Essays on Deleuze 22-23)
Oof. Getting back to the first Deleuze passage I quoted above, we might say that the "I" that first repeats and then represses does so not as an individual, original, or essential "I." Rather, the "I" that repeats is a potentiality or virtuality that subsists beneath, beyond, or beside each distinct actualization. For Derrida, there is no "present" self but, rather, a chain of supplements that have an identical ontological status. For Deleuze, it would seem, there may be no actual or individual self, but there would still be a real, virtual "I" that is ontologically distinct from any actualized mask or role that this "I" takes on. But perhaps (again) I'm getting ahead of myself.
What about the death drive? I've let that slip away . . .
To get back to Thanatos, let's turn to the final paragraph of this section. As Smith mentions, Deleuze does not dismiss psychonanalysis as a clinical practice in Difference and Repetition. (Even his later work with Guattari is less of a dismissal and more of a paradigmatic critique.) Rather, Deleuze reevaluates its effectiveness, why (in some cases) it seems successful. He writes:
We are not [. . .] healed by simple anamnesis, any more than we are made ill by amnesia. Here as elsewhere, becoming-conscious counts for little. [I love this sentence!] The more theatrical and dramatic operation by which healing takes place—or does not take place—has a name: transference. Now transference is still repetition: above all it is repetition. If repetition makes us ill, it also heals us; if it enchains and destroys us, it also frees us, testifying in both cases to its 'demonic' power. All cure is a voyage to the bottom of repetition. [. . .] In transference [. . .] repetition does not so much serve to identify events, persons and passions as to authenticate the roles and select the masks. Transference is not an experiment but a principle which grounds the entire analytic experience. The roles themselves are by nature erotic, but the verification of these roles appeals to the highest principle and the most profound judge, the death instinct. [. . .] In this sense, repetition constitutes by itself the selective game of our illness and our health, of our loss and our salvation. [. . .] It seems that the idea of a death instinct must be understood in terms of three paradoxical and complementary requirements: to give repetition an original, positive principle, but also an autonomous disguising power; and finally, to give it an immanent meaning in which terror is closely mingled with the movement of selection and freedom. (DR 19)
While Deleuze certainly offers no empirical evidence concerning the effectiveness of transference, he nevertheless provides an interesting speculation: that transference is yet another mode of repetition. It does not provide the analysand flight from repetition or compulsion. Rather, it is a practice of re-selecting roles, masks, and ideas according to which one might to learn to continue repeating or living without doing harm to oneself or others. In short, Deleuze does not deny that some compulsions are harmful. However, he does challenge us to accept that what we might call psychological healing is less a matter of the recovery of an original event or the cessation of repetition and more the invention or composition of a new mode of compulsion that will provide different effects: healing, joy, pleasure rather than psychosis, depression, or destruction. So what does this have to do with the death drive? With Thanatos? It seems to me that Deleuze (at least in this section) wants to frame the death drive as a drive of becoming-other that results from a kind of radical formlessness, groundlessness, or emptiness at the root of all things. "Death" here is not the telos of all individual life forms but, rather, the ontological blankness or darkness at the root of existence. He will later call this "the pure and empty form of time" that serves as the condition not only of ennui or angst or sadness or gloom but also—as in Proust, Woolf, and so many of the literary figures he loves—of the cultivation of joy, epiphany, love, creation, arrangement, signification, and thought (DR 122). One must learn to work and play, it would seem, in this deathly, empty condition, in which the only thing we can do is repeat differences, either to our health or our illness, our loss or our salvation. Or both.
It seems to me that philosophers or theorists who have tried to draw some sort of ethics or experimental politics from Difference and Repetition would do well to recall Deleuze's "three paradoxical and complementary requirements" of repetition and its relation to a "death instinct." Too many Deleuzians tend to forget that the purported road to freedom or cure or joy "is a voyage to the bottom of repetition" and, just like the most harmful of compulsions, is grounded in groundlessness and operates within a radical emptiness.
In my next post, we move to the penultimate section of Deleuze's introduction... which contains some of my favorite passages! Until then, reading on. Slowly.