Martin Hägglund. CR: The New Centennial Review. Volume 9, Issue 1. Spring 2009.
“Let us again return to the good we are seeking, and ask what it can be.” ~ Aristotle, Nicomachean Ethics
On an occasion when i have been given so many reasons to feel joy and gratitude, it seems appropriate to begin my response with a reflection on happiness, or more exactly on the possibility of being happy. At the outset of his Nicomachean Ethics, Aristotle identifies happiness as “that which is always desirable in itself and never for the sake of something else” (2001, 1097a). Even when we seek the most elevated virtues, we do not seek to attain them only for their own sake but also because we believe that they will make us happy. In seeking happiness, on the other hand, we do not seek any state of being beyond happiness itself. Happiness is thus the highest good because we desire it as an end in itself. When Aristotle analyzes this end, however, he runs into the problem of temporality. Aristotle himself defines happiness as an activity, which means that it takes time to be happy. But in being temporal, happiness is exposed to a future that may shatter it. Even at the moment one is happy, the moment is passing away and opens the experience of happiness to loss. According to Aristotle, happiness therefore requires a “complete life” (1098a), namely, a life that has come to an end. As long as one is alive happiness is never secured, which would seem to mean that one only achieves happiness through the completion of life in death. “Surely this is a paradox,” Aristotle remarks, “that when a man is happy the attribute that belongs to him is not to be truly predicated of him because we do not wish to call living men happy, on account of the accidents that may befall them” (1100a).
The problem outlined by Aristotle is at the core of the deconstructive logic I develop in Radical Atheism. If one defines the highest good as a state of absolute immunity—for example, as a happiness that is immune from accident and loss—it is inseparable from a state of death, since nothing can happen to it. Inversely, happiness or any other state of being requires a process of survival that takes the time to live by postponing death. On the one hand, to survive is to retain what passes away and thus to keep the past in resistance to loss. On the other hand, to survive is to live on in a future that separates itself from the past and opens it to being lost. The movement of survival is thus characterized by a double bind. Life bears the cause of its own destruction within itself, so the death that one defends against in the movement of survival is internal to the life that is defended. No matter how much I try to protect my life, I can only do so by exposing it to a future that may erase it, but which also gives it the chance to live on.
According to Derrida, the common denominator for all religions is that they promote an ideal of absolute immunity beyond the condition of survival, namely, the ideal of something that would be unscathed by time and loss. Atheism has traditionally denied the existence of such an ideal, in accordance with one of three different models. The first model is what I will call melancholic atheism. This type of atheismrenounces the belief in the unscathed, but laments its absence and holds that we are doomed to disappointment because we can never attain the timeless transcendence we desire. One of the contemporary proponents of this view is Simon Critchley, who defines his atheism in terms of “religious disappointment: disappointment that what I desire but lack is an experience of faith, namely, faith in some transcendent God” (2004, xviii). The second model is pragmatic atheism. This type of atheism argues that, while the fullness of being is an illusion, the religious desire for fullness is operative in all our commitments to the world. A prominent representative of this view is Ernesto Laclau, whose work I analyze at length in the final chapter of Radical Atheism. For Laclau, to be committed to an ethical or political cause is to invest it with the desire for fullness. We must consequently learn how to engage this desire in a progressive rather than conservative fashion. Whether we are struggling for secularism or fundamentalism, democracy or totalitarianism, we are driven by the desire for fullness. The pragmatic question becomes how to channel this desire in the best possible direction. Finally, the third model is therapeutic atheism. Unlike pragmatic atheism, therapeutic atheism does not seek to activate but rather to treat the religious desire for fullness, which is seen as the source of both individual psychological neurosis and collective political fanaticism. Beginning in the materialism of Epicurus and Lucretius, therapeutic atheism has a rich, differentiated history and today its most sophisticated defenders can be found in psychoanalytic philosophy.
The common denominator for all these models of atheism is the assumption that the religious desire for absolute immunity is operative. When we desire the good we desire an absolute good that is immune from evil, and when we desire life we desire an absolute life that is immune from death. The fundamental drama of human existence is thus seen as the conflict between the mortal being that is our fate and the immortal being that we desire. In contrast, radical atheism seeks to demonstrate that the limitation of mortal life is not a lack of being that we desire to overcome. Rather, I argue that the attachment to mortal life precipitates every positive and every negative affective response. If one were not attached to mortal life, one would not care either about oneself or any other. Only a mortal being requires care, since only a mortal being can be lost, injured, or violated.
The desire for survival-namely, the desire for a mortal being to live on—is thus presupposed in all our engagements with the world. For example, all the virtues that traditionally are assumed to be based on religious faith-compassion, love, responsibility for the other, and so on—in fact presuppose the radically atheist desire for survival. To feel compassion or take responsibility for the other, I have to be invested in a being who is susceptible to suffering and who consequently is mortal. The same goes for the experience of love. To love someone or something is to distinguish it from everything else as singularly precious. If the beloved could not die, however, it would not be possible to distinguish it as precious in the first place, since the beloved would not be irreplaceable. It is because the beloved can be lost that we seek to keep it. Or more generally: it is because things can be lost, because they have not always been here and will not always be here, that we value them.
This argument is one aspect of my notion of the unconditional affirmation of survival. A recurrent question in the responses concerns the exact status of this affirmation, so I want to be very clear on this point. Contrary to what Adrian Johnston holds in his paper, I am not proposing a life-affirming atheism. Rather, I argue that the affirmation of survival is presupposed in all responses to life. Without the affirmation of survival there would be no compassion and love (since one would not be committed to anything), but there would also be no resentment and hate (since one would not be threatened by anything). Even the most suicidal desire to end all survival presupposes the affirmation of survival, since one would not care to end all survival if one did not care about what will happen and thus cared about survival. The only way to be truly indifferent to survival is to be dead, which is to say that it is impossible for a living being to be indifferent to survival. This is the exact sense of the unconditional affirmation of survival. It spells out that whatever one does one is invested in the fate of a mortal temporal being. If one were not invested in the fate of a mortal, temporal being, one would not be invested in the fate of anything at all, since one would not care about anything that has happened or anything that may happen.
Thus, I argue that affectivity in general presupposes the care for survival. If one does not care for survival, one does not care about what happens. And if one does not care about what happens, one is not affected. In my work on psychoanalysis, then, I propose to rethink the constitution of the libidinal economy on the basis of the drive for survival. Arguing against the notion of the death drive in Freud and Lacan, I seek to demonstrate that the drive for survival allows for a better account of phenomena such as mourning, trauma, and repetition compulsion.
In his response, Johnston does not address these analyses in their own right but rather proceeds from the general claim that my logic cannot capture the unconscious processes that are the source of desire. Displaying the philosophical acumen that has established him as an outstanding reader of Freud and Lacan, Johnston provides an incisive analysis of the logic of life, death, and religion in psychoanalysis. Indeed, Johnston’s rich account of the notion of mortality in psychoanalysis goes right to the heart of the debate concerning radical atheism and chronolibido.
Johnston’s main argument is that the unconscious displays fantasies of being “undead”: a state of being that according to Johnston adheres neither to the condition of temporal survival nor to the state of immortality. However, when Johnston goes on to describe these fantasies-in a fascinating series of examples—it is clear that they are all examples of survival. Johnston addresses the afterlife of ghosts in the psyche, the refusal of death in the experience of mourning, and our attempts to live on through images and words. Far from challenging my logic of survival, these phenomena confirm it. Indeed, Johnston defines being undead as “surviving without foreseeable end, living on indefinitely” (2009, 174) and he defines repetition in fantasy as “an intra-temporal resistance to time itself, a negation of time transpiring within time” (175). These definitions answer exactly to my definition of survival.
Furthermore, Johnston’s examples of the undead pose problems for his own argument rather than for mine. His main claim is that I cannot account for the unconscious because it operates without regard for time and death. The fantasies generated by the unconscious would thus be fantasies without an awareness of temporality or mortality, but it is clear from Johnston’s own examples that this is not the case. All the fantasies he describes involve a process of mourning that is unthinkable without an awareness of temporality and mortality. To be sure, unconscious fantasies do not have to obey the chronology of linear time, but this does not mean that they can be exempt from the succession of time. On the contrary, the retroactive temporality of the unconscious that Johnston posits as a challenge to my notion of time, itself presupposes the notion of time that I derive from the implications of succession. The deferral and delay that Freud calls nachträglichkeit is on my account characteristic of temporal experience in general. A temporal event can never be present as such, since it comes into being only by becoming past and becoming related to the future. The experience of the event is always given too late (in relation to what is no longer) and too soon (in relation to what is not yet). Every experience is thus characterized by a retoractive temporality, since what happens exceeds any given anticipation and can be apprehended only in retrospect, when it has already passed. If unconscious fantasies were not marked by this succession of time, nothing would happen in them and they would not be fantasies of anything.
No doubt Johnston would here retort, as he does in his paper, that my arguments are “too logical” (2009, 151) for the unconscious and that I proceed from the position of an “armchair-philosopher” (181) who ignores the evidence offered by psychoanalytic practice. Such an opposition between theory and practice is misleading, however, when it comes to the metapsychological questions that are at stake. When psychoanalysis asserts that the unconscious is ignorant of time and death, or that we desire immortality, it does not rely on empirical evidence but on speculative concepts through which the empirical evidence is interpreted. To challenge the coherence of these concepts, as I seek to do, is thus also to challenge the interpretation of the empirical evidence. The most important move here is my argument that the concept of immortality is inseparable from the concept of death. If we fantasize about living on after death we do not fantasize about being immortal, since to live on is to remain subjected to temporal finitude. Inversely, the state of immortality cannot answer to the desire to live on, since it would put an end to the time of life.
With this clarification in place, we can pinpoint why Freud makes the mistaken inference that the unconscious is ignorant of temporal finitude. In a famous essay to which Johnston appeals, Freud argues that because we cannot imagine our own death, we are unconsciously convinced of our own immortality (Johnston 2009, 160). Freud is certainly right that we cannot imagine our own death, because to do so we have to imagine ourselves as surviving to witness our own death. For the same reason, however, we cannot imagine our own immortality, since immortality is the same as death. Hence, it does not follow from Freud’s argument that we are all convinced that we are immortal, since in that case we would all be convinced that we are dead. What follows from Freud’s argument is that even in our relation to death we fantasize about survival, but since Freud fails to distinguish between immortality and survival, he fails to make the point.
Similarly, Johnston conflates the problem of relating to oneself as mortal and the problem of relating to oneself as dead. To be sure, it is “impossible for subjects genuinely to envision their own non-being” (2009, 173), but this point does not contradict my logic of temporal survival. I explicitly argue that one cannot envision or experience one’s own death, since if one were to experience one’s own death one would not be dead. The only death I can experience is rather the death of an other whom I survive. Inversely, my relation to my own death marks my exposure to a future that will survive me and never can be appropriated by myself.
Consequently, Johnston’s powerful account of “a life-long process of mourning, a never-finished project of continually recognizing and misrecognizing one’s status as a death-bound being” (2009, 173) can be systematically reinterpreted in terms of my notion of a constitutive drive for survival. I could subscribe to Johnston’s insight that “the psychoanalytic subject of desire cannot but view itself as surviving without end, as living on interminably” (174), but that is precisely why I argue that the constitution of the libidinal economy should be rethought on the basis of a drive for survival rather than the drive for fullness posited by Lacanian psychoanalysis.
Accordingly, the passage from F. H. Jacobi that Johnston puts forward as a counterexample to my radically atheist logic of desire is in fact better explained by that logic than by the Lacanian logic of lack. Jacobi confesses that he cannot accept either the prospect of immortal life or the prospect of final death (Johnston 2009, 178). This is precisely the constitutive fear of death that I theorize under the heading of radical atheism and chronolibido. My argument is that the fear of death makes it impossible to accept either immortality or death because they are the same in that both terminate the time of mortal life. Far from advocating that we come to terms with finitude, the logic of radical atheism explains why there is no way to approach life that would allow one to accept death resolutely or immunize oneself from the traumatic impact of being mortal.
Thus, contra Johnston’s critique, I maintain that my account of desire is descriptive rather than prescriptive. At the end of my response I will return to the question of how this description of desire changes the conception of atheism, but for now I want to turn to Samir Haddad’s paper, which helpfully focuses on the relation between the descriptive and the prescriptive in developing arguments from the important series of essays on Derrida he has previously published. Haddad’s first question is how we can value anything at all if everything is subjected to the chance and threat of temporal finitude. Specifically, Haddad argues that it is impossible to value something if the chance of what is desired equally and at the same time constitutes the threat of what is feared (2009, 137). This argument presupposes that the chance is identical to the threat. Earlier in his paper, however, Haddad himself carefully shows that according to my analysis the chance is not identical to the threat. “Rather, they are two different possibilities—the chance of living on and the threat of not living on—that necessarily arise together because of their source in spacing” (129). Consequently, there is never a situation where there is an equal distribution between the chance and the threat or where the two amount to the same thing. My argument is only that the chance necessarily is haunted by the threat, in accordance with the structure of temporal finitude that enables anything to be valued in the first place. Without the threat of loss, there would be no impetus to attain or keep that which we value. This is a descriptive claim because it does not promote any particular value but seeks to account for why everything we value is threatened from within.
Accordingly, I argue that injustice is intrinsic to justice, hostility intrinsic to hospitality, and so on. In tracking these arguments in Derrida’s work I am not “indifferent to language,” as Haddad claims, since I make explicit that Derrida focuses on the positively valued term (“justice,” “hospitality,”) rather than the negatively valued term (“injustice,” “hostility”) that is shown to be complicated with the positively valued term. If Derrida’s analysis did not focus on the term that is assumed to be valuable and desirable, it would not accomplish the deconstruction of our inherited assumptions about what is valuable and desirable. The motivation for focusing on the positively valued term, then, is not to promote justice or hospitality as a value, but to demonstrate that even the values that are regarded as the highest are essentially corruptible and can have negative eff ects just as well as positive effects. Rather than disregarding that deconstruction is engaging with language and values, I argue that the deconstructive analysis intervenes in our conception of values by demonstrating the inherent undecidability of any given value.
Nevertheless, Haddad hones in on seminal issues that for too long have been neglected in writings on Derrida. He is certainly right to point out that the terms we use are infused with value and his analysis makes clear that we need to come to terms with how Derrida negotiates this fact. For Haddad’s argument to have critical purchase on my analysis, however, he would have to show that the latter is compromised by the inherited values that are latent in the terms I analyze. That is, Haddad would have to show that my analyses promote certain values despite themselves. Alternatively, Haddad could show that the inherited nature of language leads Derrida to promote values that cannot be justified on the basis of his own logic, and that Derrida’s avowed commitment to democracy, for example, has to be questioned, developed, or transformed for that reason. Of course, Derrida is not necessarily wrong to promote a certain value of democracy, but this valuation cannot be justified on the basis of deconstructive reason. Thus, one may argue that Derrida’s use of the term requires a more careful historical and political analysis of the inheritance of democracy. Such an argument would be different from but consistent with the logic I develop in Radical Atheism and could be a powerful contribution to Derrida scholarship.
In his paper, however, Haddad pursues neither of these alternatives. Instead, when he turns to a discussion of democracy, Haddad formulates two different objections. First, he takes issue with the argument (made by both Derrida and myself) that the term “democracy” has the descriptive merit of highlighting the relation to an undecidable future. According to Haddad, the term “fraternity” could be “equally mined” to emphasize “an openness to change and transformation through self-engagement” (2009, 140). To make this argument, however, Haddad must neutralize his own insight about the significance of language. Fraternity and democracy do not off er equal resources for deconstruction precisely because they are terms with different histories, and on the level of language democracy is by definition a more open category than fraternity because it does not restrict the potential community to “brothers” but to an indeterminate “people” that can include anyone whosoever. Haddad’s second point is that Derrida does not limit himself to a constative analysis of the concept of democracy but also engages in a performative commitment to democracy. This claim would only have critical consequences for my position if Haddad could show that Derrida’s performative commitment to democracy is grounded in his constative analysis of democracy. But Haddad’s own discussion of democracy shows that this is not the case, since it shows that Derrida’s analysis just as well can be used to deduce reasons for being against democracy. As Haddad points out, Derrida “underlines at length the negative characteristics of democracy, and one could imagine Derrida or his heirs coming out forcefully against it on this basis” (142).
Hence, it is unclear how Haddad’s discussion of democracy is supposed to support his general assertion that there cannot be a descriptive level of analysis because of the inherited nature of language. Indeed, Haddad himself cannot avoid presupposing a descriptive level to make his argument. Take, for example, Haddad’s central claim that “undecidability is inescapable” and that “this is precisely what is entailed by the inherited nature of language. Derrida cannot control the value of his discourse any more than he can control its meaning” (2009, 142). This is clearly a descriptive statement. To establish the conditions for valuation, Haddad has to lay claim to an argument that itself is not a valuation but a description.
My argument, then, is twofold. First, while many statements that claim to be descriptive in fact are prescriptive, this does not mean that descriptive statements are not in principle possible. Even the argument that there can be no strict separation between the descriptive and the prescriptive must claim to describe the fact that there can be no strict separation between the descriptive and the prescriptive, so Haddad himself must appeal to a descriptive level to make his argument. Second, if there were a given link between deconstruction and a prescriptive commitment, whether through language or anything else, Derrida’s argument about undecidability would be invalidated. Derrida may indeed oscillate between an analysis of the concept of democracy and a performative commitment to democracy, but the latter cannot be grounded in the former. Rather, Derrida argues that we make commitments because of the exposition to a future that may come to question or undo these commitments. To insist on this condition is not to assert the possibility of “value neutrality” in politics. On the contrary, it is to insist on what I call the “hyperpolitical” nature of values (2008, 181-87, 202-3). For a hyperpolitical thinking, no set of values can be posited as good in itself or as immune from critique but is always given over to political negotiation. It follows that the value of a given commitment only can be decided from time to time, in accordance with strategic considerations rather than an a priori rule.
Thus, Derrida draws a clear distinction between performative acts of language and the structure of the event that he describes as unconditional. We necessarily commit ourselves to values through performative acts of language, but Derrida maintains that these acts are exceeded from within by the event that makes them possible. As he puts it in “Typewriter Ribbon”: “What happens, by definition, what comes about in an unforeseeable and singular manner, could not care less about the performative” (2002, 146). Derrida’s point is that even the most stable commitment can betray itself or turn out to be misguided because of the exposition to unpredictable events. This does not mean that commitments or values are “arbitrary in their justification,” as Haddad suggests (2009, 137); it only means that they are based on reasons and considerations that are not grounded in deconstruction. The role of deconstruction is not to ground anything but to think through the implications of the unconditional exposition to time.
More exactly, the unconditional is for Derrida the coimplication of time and space that he calls spacing. Exactly how spacing should be understood is the question raised by Henry Staten’s paper. Staten himself was the first to highlight the importance of time for deconstructive logic in his brilliant study Wittgenstein and Derrida, and his response to my work is crucial for clarifying a number of issues. According to Staten, Derrida’s proposition that time is spacing is a stipulation rather than a demonstration. Furthermore, Staten claims that Derrida’s proposition only pertains to “the structure of lived experience in time” (2009, 76). I begin by recalling these claims because they diverge from my position in Radical Atheism. I agree with Staten that the conception of time as spacing “should be submitted to as much pressure as it will bear” (2009, 79), but that is precisely why I seek to develop the logic of Derrida’s proposition. Specifically, I argue that the traditional philosophical concept of time as succession allows us to demonstrate why time must be thought as spacing. Thus, I maintain that the spacing of time does not merely pertain to the structure of lived time but follows from the concept of succession itself. This is also Derrida’s point in “Ousia and Grammè,” where he argues that already Aristotle’s and Hegel’s discussions of time show that it must be thought as spacing (1982, 29-67). Indeed, in an analysis that is central for Derrida, Hegel explicitly emphasizes that the transition from space to time is not made subjectively by us but follows from the constitution of space itself. Inversely, the transition from time to space follows from the constitution of time itself (Hegel 1970, 34).
How can we demonstrate this? The classical distinction between space and time is the distinction between simultaneity and succession. The spatial can remain the same, since the simultaneity of space allows one point to coexist with another. In contrast, the temporal can never remain the same, since the succession of time entails that the now is immediately replaced by another now. Time posited for itself is thus, as Hegel reminds us, nothing but negativity (1970, 34). For the now to be succeeded by another now, it must negate itself as soon as it comes to be. By the same token, however, it is clear that time is impossible without space. Time is nothing but negation, so in order to be anything it has to be spatialized. There is no “flow” of time that is independent of spatialization, since time has to be spatialized in order to flow in the first place. Indeed, everything we say about time (that it is “passing,” “flowing,” “in motion,” and so on) is a spatial metaphor. This should not be understood as a failure of language to capture pure time but as an eff ect of what Derrida calls the originary becoming-space of time. The very concept of duration presupposes that something remains across an interval of time and only that which is spatial can remain. Inversely, the simultaneity of space is impossible without a temporalization that allows one point to be related to another. As Derrida puts it, “simultaneity can appear as such, can be simultaneity, that is a relating of two points, only temporally. One cannot say that a point is with another point, there cannot be an other point with which, etc., without a temporalization” (1982, 55).
In Radical Atheism, I seek to explain at length how the becoming—space of time and the becoming—time of space should be understood. When I use the example of empirical writing in the passages that Staten incisively examines, it is in order to account for why Derrida thought writing was an instructive metaphor to explain the transcendental structure of spacing. Staten’s critical point is that one must carefully distinguish between the transcendental claim that experience is a form of inscription, on the one hand, and the empirical inscription of experience on the other. This is correct and helpful insofar as one limits empirical inscription to signs and languages, but neither Derrida nor I do that. Empirical inscription includes all forms of inscription. The deconstructive argument, then, is that empirical inscription is transcendentally necessary. If this were not the case, we would be back to a traditional distinction between the transcendental and the empirical, where the former has an integrity that is immune from mutations of the latter. The transcendental structure of spacing, however, explains that there is no such integrity. If experience (transcendentally) is a form of inscription, it follows that every experience must be (empirically) inscribed somewhere.
The “pincer movement” of deconstruction is thus, as Staten rightly points out (2009, 74), to establish the transcendental necessity of empirically contingent inscriptions. Staten emphasizes how difficult it is to explain cogently this pincer movement, but I would suggest that there is a possible solution to the problem in Radical Atheism. As I argue, one cannot explain the trace structure of the now by merely appealing to its constitutive relation to past and future nows. This appeal is insufficient, since it does not explain why the now is not past, present, or future. My argument seeks to provide the required explanation by showing that the now never is because of the structure of succession that constitutes the now itself. And it is precisely because the now never is—because it passes away as soon as it comes to be—that it must be inscribed as a trace in order to be at all. Hence, the necessity of inscription follows from the structure of succession.
For the same reason, however, the structure of the trace cannot be restricted to the structure of lived experience but is an “ultratranscendental” condition for everything that is temporal. Although there may be good reasons for focusing on the question of life—and I will return to my reasons for doing so later on—the trace structure cannot be limited to the living but pertains to succession in general. Everything that is subjected to succession is subjected to the trace, whether it is alive or not.
Let me specify the stakes of this argument by considering the challenge to transcendental philosophy posed by Quentin Meillassoux’s treatise After Finitude, which appeared after Radical Atheism was written. There are striking parallels between the notion of the “ultratranscendental” that I develop and the critique of transcendental philosophy articulated by Meillassoux, but as Aaron F. Hodges points out in his contribution to this issue, there are also important diff erences that need to be addressed.
Meillassoux’s point of departure is the empirical phenomenon of what he calls “arche-fossils,” namely, objects that are older than life on earth and whose duration it is now possible to measure: “for example an isotope whose rate of radioactive decay we know, or the luminous emission of a star that informs us as to the date of its formation” (2008, 10). Such arche-fossils enable scientists to date the origin of the universe to approximately 13.5 billion years ago and the origin of life on earth to 3.5 billion years ago. According to Meillassoux, these “ancestral” statements are incompatible with the basic presupposition of transcendental philosophy, which holds that the world cannot be described apart from how it is given to a thinking and/or living being. The ancestral statements of science describe a world in which nothing was given to a thinking or living being, since the physical conditions of the universe did not allow for the emergence of a life or consciousness to which the world could be given. The ensuing challenge to transcendental philosophy “is not the empirical problem of the birth of living organisms, but the ontological problem of the coming into being of givennness as such” (21). Rather than being able to restrict time to a form of givenness for consciousness, we are confronted with an absolute time “wherein consciousness as well as conscious time have themselves emerged in time” (21).
Meillassoux is well aware that he could here be accused of conflating the empirical with the transcendental. Empirical bodies emerge and perish in time, but the same cannot be said of transcendental conditions. The transcendental subject is not an empirical body existing in time and space, but a set of conditions through which knowledge of bodies in time and space is possible. Thus, a scientific discourse about empirical objects or the empirical universe cannot have purchase on the transcendental subject, since the latter provides the condition of possibility for scientific knowledge.
Meillassoux’s rejoinder to this objection is one of his most ingenious philosophical moves. He grants that the transcendental subject does not exist in the way an object exists, but insists that the notion of a transcendental subject nevertheless entails that it must take place, since it presupposes the existence of a physical body that limits the perspective on the world. The transcendental subject—as both Kant and Husserl maintain—is essentially finite, since it never has access to the world as a totality but is dependent on receptivity, horizon, perceptual adumbration, and so on. It follows that although transcendental subjectivity is not reducible to an objectively existing body, it must be incarnated in a body in order to be what it is. As Meillassoux puts it, “that the transcendental subject has this or that body is an empirical matter, but that it has a body is a non-empirical condition of its taking place” (2008, 25). Consequently, when scientific discourse “temporalizes and spatializes the emergence of living bodies,” it also temporalizes and spatializes the basic condition for the taking place of the transcendental (25). Thus, Meillasssoux argues that the problem of the ancestral “cannot be thought from the transcendental viewpoint because it concerns the space-time in which transcendental subjects went from not-taking-place to taking-place-and hence concerns the space-time anterior to spatiotemporal forms of representation” (26). Far from confirming the transcendental relation between thinking and being as primordial, the ancestral discloses “a temporality within which this relation is just one event among others, inscribed in an order of succession in which it is merely a stage, rather than an origin” (10).
Meillassoux’s argument is significant for my notion of the ultratranscendental, since he does not rely on an empiricist, positivist, or metaphysical discourse to attack transcendental philosophy. Rather, Meillassoux turns the central argument of transcendental philosophy against itself. For my part, I seek to traverse the texts of transcendental philosophies of time to show that they presuppose the structure of the trace that contradicts them from within. By virtue of its own temporality, transcendental subjectivity is dependent on a material support whose necessity cannot be reduced to its own constitution. This is what I call the “arche-materiality” that precedes the relation between thinking and being, just as it precedes the relation between the animate and the inanimate. Thus, the non-living matter of Meillassoux’s arche-fossils clearly presupposes the arche-trace and arche-materiality of time. If the events to which the arche-fossils testify (e.g., the origin of the universe and the accretion of the earth) were not inscribed as a trace when they happened, nothing would remain of them and it would be impossible to date them.
Meillassoux himself does not draw this conclusion, and he is strangely silent on the question of how ancestral time recorded itself. For Meillassoux, the important point is that mathematics can access the data of ancestral reality, but the mathematical calculations of ancestral time in turn depend on the material support of arche-fossils, which presuppose the trace structure of time. Hence, while Meillassoux’s point that time cannot be reduced to a transcendental form of intuition is well taken, it does not follow that there is no common denominator between ancestral time and phenomenological time. Succession is not only operative in the consciousness of a transcendental subject but also in what Meillassoux calls “absolute time,” since the latter both precedes and exceeds the existence of thinking/living beings.
The concept of succession is perhaps the most undertheorized notion in Meillassoux’s ontology, and it is here that decisive differences between our respective arguments begin to emerge. Meillassoux argues that the principle of noncontradiction must be “an absolute ontological truth” (2008, 71) for temporal becoming to be possible. If a contradictory entity existed, it could never become other than itself, since it would already contain its other within itself. Given that it is contradictory, it could never cease to be but would rather continue to be even in not-being. Consequently, the existence of a contradictory entity is incompatible with temporal becoming; it would eliminate “the dimension of alterity required for the deployment of any process whatsoever, liquidating it in the formless being which must always already be what it is not” (70). This argument is correct as far as it goes, but it fails to consider that the same problem arises if we posit the existence of a noncontradictory entity. A noncontradictory entity would be indivisibly present in itself. Thus, it would remove precisely the “dimension of alterity” that is required for becoming. Contrary to what Meillassoux holds, the movement of becoming cannot consist in the movement from one discreet entity to another, so that “things must be this, then other than this; they are, then they are not” (70). For one moment to be succeeded by another—which is the minimal condition for any becoming whatsoever—it cannot first be present in itself and then be affected by its own disappearance. A self-present, indivisible moment could never even begin to give way to another moment, since what is indivisible cannot be altered. The succession of time requires not only that each moment be superseded by another moment; it also requires that this alteration be at work from the beginning. Even the most immediate moment must negate itself and pass away in its very event. If the moment did not negate itself there would be no time, only a presence forever remaining the same.
This argument—which I pursue at length in Radical Atheism—does not entail that there is a contradictory entity that is able to contain its own nonbeing within itself. On the contrary, I argue that the constitution of time entails that there cannot be any entity (whether contradictory or noncontradictory) that contains itself within itself. The succession of time requires that nothing ever is in itself; it is always already subjected to the alteration and destruction that is involved in ceasing-to-be.
The problem of succession is directly relevant for the main argument in After Finitude, which seeks to establish the necessity of contingency. As Meillassoux formulates his guiding thesis: “Everything is possible, anything can happen-except something that is necessary, because it is the contingency of the entity that is necessary, not the entity” (2008, 65). This notion of contingency presupposes succession, since contingency requires the unpredictable passage from one moment to another. To establish the necessity of contingency, as Meillassoux seeks to do, is thus also to establish the necessity of succession.
Meillassoux himself, however, does not theorize the problem of succession, and this comes at a significant cost for his argument. According to Hodges, Meillassoux’s critique of the principle of sufficient reason is potentially damaging for my notion of radical destructibility, which holds that everything that comes into being must pass away. But in fact, it is rather my notion of radical destructibility that allows us to locate an inconsistency in Meillassoux’s argument. Let me quote in full the passage from Meillassoux to which Hodges calls attention:
To assert … that everything must necessarily perish, would be to assert a proposition that is still metaphysical. Granted, this thesis of the precariousness of everything would no longer claim that a determinate entity is necessary, but it would continue to maintain that a determinate situation is necessary, viz., the destruction of this or that. But this is still to obey the injunction of the principle of reason, according to which there is a necessary reason why this is the case (the eventual destruction of X), rather than otherwise (the endless persistence of X). But we do not see by virtue of what there would be a reason necessitating the possibility of destruction as opposed to the possibility of persistence. The unequivocal relinquishment of the principle of reason requires us to insist that both the destruction and the perpetual preservation of a determinate entity must equally be able to occur for no reason. Contingency is such that anything might happen, even nothing at all, so that what is, remains as it is. (2008, 62-63)
While emphasizing that a necessary entity is impossible, Meillassoux maintains that it is possible for nothing to happen, so that the entity remains as it is. But as soon as we take into account the intrinsic link between contingency and succession, we can see that the latter argument is untenable. If nothing happened and the entity remained as it is, there would be no succession, but by the same token there would be no contingency. An entity to which nothing happens is inseparable from a necessary entity. To be subjected to succession-which is to say, to be contingent-the entity must begin to pass away as soon as it comes to be and can never remain as it is. Consequently, there is a reason that necessitates destruction, but it does not re-import a metaphysical principle of reason. On the contrary, it only spells out what is implicit in the principle of unreason that Meillassoux calls the necessity of contingency. Contingency presupposes succession and there is no succession without destruction. If the now were not destroyed in being succeeded by another now, their relation would not be one of succession but of coexistence. Thus, to assert the necessity of contingency is to assert the necessity of destruction.
For the same reason, Meillassoux’s opposition between destruction and persistence is misleading. Persistence itself presupposes an interval of time, which means that nothing can persist unscathed by succession. The destruction of the now makes any persistence dependent on the spacing of time, which inscribes what happens as a spatial trace that remains while exposing it to erasure in an unpredictable future. The erasure of the spatial trace is indeed a possibility that is not immediately actualized, but it already presupposes the necessary destruction of the now that is intrinsic to succession, so even in the most ideal persistence there is destruction at work.
Hence, I can answer Hodges’s demand that I demonstrate how “Meillassoux himself unconditionally requires spacing in his account of absolute contingency” (2009, 103). Given that contingency presupposes succession, and that the spacing of time follows from the structure of succecssion, Meillassoux needs the spacing of time for his account of absolute contingency to work.
The coimplication of persistence and destruction is what I analyze as the general condition of survival. One may then legitimately wonder why I focus on the question of life, since survival is operative even in the nonliving matter of ancestral time. Indeed, both Haddad and William Egginton appositely raise the question of the scope of “life” in my argument: does it include all things that exist in time or only a subcategory of them? This issue is not directly treated in Radical Atheism and calls for clarification, so I am grateful to have been confronted with the question. My answer is that everything in time is surviving, but not everything is alive. Accordingly, there is an asymmetry between the animate and the inanimate in the arche-materiality of the trace. As soon as there is life there is death, so there can be no animation without the inanimate, but the inverse argument does not hold. If there were animation as soon as there is inanimate matter, we would regress to a vitalist conception of the universe, where life is the potential force or the teleological goal of existence. The conception of life that follows from the arche-materiality of the trace is as far as one can get from such vitalism, since it accounts for the utter contingency and destructibility of life. As Staten cogently formulates it in an important recent essay: “the strong naturalist view, from which Derrida does not deviate, holds that matter organized in the right way brings forth life, but denies that life is somehow hidden in matter and just waiting to manifest itself … Life is a possibility of materiality, not as a potential that it is ‘normal’ for materiality to bring forth, but as a vastly improbable possibility, by far the exception rather than the rule” (2008, 34-35).
What diff erence, then, is at stake in the advent of life? In “Typewriter Ribbon,” Derrida speaks of his fascination with the archive of events that happened 54 million years before humans appeared on earth, in particular the archive of two midges who were immobilized in amber when they were surprised by death as they made love. “We have there,” Derrida writes, “consigned to a support, protected by the body of an amber coffin, the trace, which is itself corporeal, of an event that took place only once … the archive of a singular event and, what is more, of an event that happened to some living being, affecting a kind of organized individual, already endowed with a kind of memory, with project, need, desire, pleasure, jouissance, and aptitude to retain traces” (2002, 131). There is a crucial difference between this archive of an event from before humans appeared on earth and the archive of events from before life in general. The isotope that has a rate of radioactive decay across billions of years is indeed surviving, since it remains and disintegrates over time, but it is indifferent to its own survival, since it is not alive. The midges, on the other hand, have a project, need, and desire. Like any other living being, they cannot be indifferent to their own survival. This distinction is decisive for the definition of life in Radical Atheism. The reason I focus on life is because only with the advent of life is there desire in the universe. Survival is an unconditional condition for everything that is temporal, but only for a living being is the affirmation of survival unconditional, since only a living being cares about maintaining itself across an interval of time.
The unconditional affirmation of survival is at the core of my rethinking of desire. Over the years, I have had a productive exchange with Eggington concerning this notion of desire and in his contribution he provides a lucid synthesis of our debate. The core of Eggington’s critique is that I fail to assess how Lacan’s notion of the drive anticipates my argument. In both Radical Atheism and the essay on chronolibido, however, I do discuss Lacan’s distinction between desire and drive, and I explain why it does not answer my critique of psychoanalysis. When Egginton quotes my chronolibidinal argument about the drive and claims that no self-identified Lacanian would disagree with it (2009, 199), he draws on a passage where I explicitly argue against Adrian Johnston, who in my view has put forth the most systematic and compelling version of Lacanian drive theory. The Lacanian notion of the drive that I criticize has also been developed by Alenka Zupancic and Joan Copjec, so I do take into account the ‘canonical’ interpretation to which Eggingtion appeals. Indeed, Egginton’s own account of the drive in Perversity and Ethics follows the schema with which I take issue. Although Egginton is well aware that the drive is always inhibited by the restrictions that follow from temporal being, he holds that it is ruled by a “conservative impulse” (2006, 18) that seeks to return to a state of being that precedes the strictures of time. Thus, Egginton distinguishes between “the mythical state of an uninstitutionalized primary process” that he aligns with the drive and the process of establishing “barriers to the free flow of energy” (18). Only the latter process introduces temporality, whereas the drive in its mythical pure state “could not know time” (18). This is certainly a viable interpretation of Freud’s and Lacan’s notion of the drive, but it is not compatible with my chronolibidinal conception of the drive.
The common denominator for both desire and drive in Lacan is the conflict between the mortal being that we are and the immortal being that we lack.In contrast, I locate the fundamental conflict of libidinal being in the attachment to mortal life itself: in temporal finitude as the source of both our desire and our fear, both our dreams and our nightmares. When I argue that the desire for absolute fullness is not operative, it does not mean—as Egginton charges—that I think there are no illusory beliefs or that these beliefs cannot eff ectively instigate actions. Rather, I argue that they are effective because they engage a desire for survival and not a desire for immortality. Even the illusions of permanence or immortality—and the tenacity with which one can hold on to these beliefs—are an effect of the desire for survival.
Let me elaborate the stakes of this argument by turning to Michael Naas’s precise and beautifully articulated paper. Like Egginton, Naas is concerned that I do not take into account the force of illusions and, more specifically, the force of phantasms. With his characteristic, admirable attention to the rhetorical aspect of philosophical writing, Naas analyzes the different senses of the term “desirable” in my work and makes a sophisticated case for a prescriptive dimension in deconstruction. According to Naas, we must both assess the “fatal attraction” of metaphysical phantasms and combat their effects. Thus, Naas argues that we can employ the logic of deconstruction in order to criticize politically pernicious phantasms of sovereignty. Deconstruction would then involve “the minimal prescription to aid in the exposure and the deflating or simply the deconstruction of the phantasms of any and all absolutes-any absolute purity, any claim to autochthony, any call for salvation, any attribution of sovereignty” (2009, 65). Despite the nuances in Naas’s account, the problem with this argument is that it introduces a criterion for decision that is at odds with the deconstructive analysis of sovereignty. Naas’s discussion of political phantasms focuses on the examples of racial purity and ethnic autochthony, which leads him to say that we do “the most abominable things” in the name of sovereignty (63). As Nass himself is well aware, however, phantasms of sovereignty do not only inspire genocide and ethnic cleansing; all sorts of political struggles that we may want to support are fought in the name of sovereignty. Indeed, no political struggle can legitimate itself without appealing to sovereignty-whether it is the sovereign right of a country not to be invaded, the sovereign will of a people to change existing conditions, or the sovereign right to bodily integrity of a given individual. The deconstructive analysis shows that absolute sovereignty is impossible and undesirable, since it would cancel out life, but it does not follow that appeals to sovereignty are bad or that we should seek to deflate any attribution of sovereignty. It is not necessarily better to be less sovereign than more sovereign, and it is not necessarily better to fight against a given claim to sovereignty than to support it.
Furthermore, when I insist that one cannot derive prescriptions from deconstructive descriptions, it is not to make deconstruction apolitical but to press home the hyperpolitical logic of deconstruction. This hyperpolitical logic spells out that the question of sovereignty cannot be settled on the basis of an opposition between the phantasmatic and the real, which would justify a critique of phantasms a priori. If the deconstructive analysis of sovereignty provided such a justification, it would depoliticize the question of sovereignty, since there would be a criterion for decision that is exempt from possible critique. In contrast, the hyperpolitical logic of deconstruction recalls us to the exigencies of deciding whether or not a given phantasm of sovereignty is for better or for worse. As Derrida points out, “according to the situations, I am an antisovereignist or a sovereignist—and I vindicate the right to be antisovereignist at certain times and a sovereignist at others” (qtd in 2008, 183).
For the same reason, it does not follow from the logic of radical atheism that we should oppose political struggles that are fought in the name of religious ideals. As I emphasize in the book, it is not a matter of renouncing struggles for health or of denouncing hopes for salvation. Rather, it is a matter of demonstrating that these struggles and hopes were never concerned with the absolute immunity that is promoted as the religious ideal. The struggle for health and the hope for salvation have never been driven by a desire to be immortal but by a desire to live on as mortal. It follows that one may have good reasons to support a certain religious phantasm in order to support the survival of a certain individual or community.
The logic of radical atheism can thus be clearly distinguished from the three models of atheism that I delineated in the beginning of this response. Against melancholic atheism, I argue that there is no reason to lament the absence of the unscathed. The religious desire for timeless transcendence is not the truth of desire but dissimulates our primary and irreducible attachment to mortal life. Against pragmaticatheism, I argue that there is no intrinsic necessity that we organize our individual or collective struggles in terms of religion or religious substitutes. Given that the commitment to a cause is not driven by a desire for fullness but by a desire to live on, we can invent forms of representation that channel this desire in nonreligious ways. Against therapeutic atheism, finally, I argue that religion is neither the ultimate cause of individual psychological neurosis nor of collective political fanaticism. Consequently, there is no guarantee that it is better to remove than to keep religion in a given individual or collective imaginary.
The project of radical atheism, then, is not to convert anyone. It does not seek to teach us how to live or what to do, but to describe the exigencies that are at work whatever we do and however we live. In his paper, David E. Johnson demonstrates how these exigencies of survival can be read in Borges’s texts. I want to conclude by turning to a text that Johnson does not discuss but that confirms his argument that the logic of radical atheism is legible in Borges’s writing.
In “The Immortal,” Borges recounts the fate of those who have transcended time and mortality. Far from having attained the highest good, they have been deprived of every possibility of value and virtue. The immortals can have no courage, since they cannot risk anything, and they are immune to compassion, since they have no interest in their own fate or those of others. As Borges puts it, “nothing is preciously precarious” for the immortals, since for them “nothing can happen only once” (2007, 116). In contrast, “everything among the mortals has the value of the irretrievable and the perilous,” and their capacity to be moved by what happens results from “their phantom condition: every act they execute may be their last; there is not a face that is not on the verge of dissolving like a face in a dream” (115).
In Radical Atheism I attempt to articulate this spectral condition of survival—and the affective force of the “preciously precarious”—in the register of ontology, ethics, and politics. I am deeply grateful for the way in which the contributors to this issue have challenged me to pursue my project further. This unique occasion to think and to respond has made me very happy, even though it is impossible to know how long this happiness will last.