The Ethics of Self–Sacrifice


John Milbank


Copyright (c) 1999 First Things 91 (March 1999): 33-38.

If I were to say that the highest imaginable exemplification of the good consists in dying sacrificially on behalf of an other or others, I imagine that many people, religious or otherwise, would concur. And in some recent ethical thinking, this understanding of the highest good has been given a philosophically systematic and rigorous expression. For such thinkers as Jan Patocka, Emmanuel Levinas, Jacques Derrida, and, to a certain extent, Jean–Luc Marion, the highest ethical gesture is a sacrificial self–offering which expects no benefit in return. The good is, paradigmatically, a purified sacrifice, the purest sacrifice imaginable.

Anyone who thinks of this pure self–sacrifice more closely must answer four questions: How is giving to be understood? What is the reality of death? What is the appropriate concept of the self? What are the background ontological circumstances against which the sacrificial gesture would be situated?

Recent ethical thinkers have certain characteristic answers to these questions. The only real gift, they claim, is one that expects no counter–gift in return. Unless a gift is in this fashion sacrificial—the giving up of something—it is argued, a gift reduces to a hidden contractual agreement, governed by a principle of self–interest; and actions out of self–interest, as Kant pointed out, are not pure gifts.

Secondly, they hold that death, far from being complicit with evil as religious traditions have often taken it to be, is the very circumstance that makes it possible to act ethically at all. This claim further breaks down into two complementary parts: 1) We are radically and ultimately vulnerable only because we might die—an immortal would be in the most crucial aspects invulnerable. Hence it is the fact of death alone that lends serious gravity to the ethical demand which vulnerability imposes upon us. 2) At the limit, the ethical agent might die for the vulnerable other person. This readiness to die alone guarantees the ultimate disinterest of his ethical gesture, since it would seem that a good one is prepared to die for cannot be the secret vehicle of one’s own power or (presently enjoyed) glory. In this sense, readiness to die precludes the will to power.

Thirdly, in the trend of ethical thinking we are investigating, it is characteristically assumed that what makes us aware of the self in the first place is just this double intrusion of death: the cry of the vulnerable other eliciting our preparedness to negate our own life. Combine this understanding of self with a common epistemological belief, and we bring God into the picture. The epistemological belief is that when something appears to us, when it is present to our consciousness, we can see only what we understand and are able to grasp; we reduce the "other" to the sphere of our awareness. If this is the case, then for the vulnerability of another to place an ethical demand on us greater than ourselves, the other must be greater than ourselves. Thus, the demand of the other with a small "o" passes mistily over into the claim of the other with a big "O," the demand of transcendence, of deity.

Finally, in the fourth place there is the question of ontology, of just what kind of world it is in which gift without return and the death of the other linked to my own death gives rise to subjectivity and ensures that as subjective beings we are first and foremost ethical creatures—even before we are erotic creatures or curious creatures. Recent thought has it that ours is a world in which death, the passing away of life beyond being into nothingness, is an ultimate horizon. It is suggested that only within this horizon does ethics acquire an ultimate seriousness. For if we are all terminally fragile, then our temporary lives assume an ultimate value, since we can offer our own lives for the sake of others. A death without return ensures that the choice of the good exceeds any self–interest, and that the good lies, as Levinas says, "beyond being [including our own]." With God reduced to a shadow of the human other, and no longer seen as the source of compensating heavenly rewards, the ultimate religious and ethical imperative of pure sacrifice is therefore fulfilled within a secular and symbolically drained sphere, harboring no illusions. Common to all these thinkers (with the exception of Marion) is an attempt to make nothingness or the continuous disappearance of life into the void the precondition for morality, rather than an obstacle in its path. Death in its unmitigated reality permits the ethical, while the notion of resurrection contaminates it with self–interest.

So is it true that death undergirds ethics? I want to argue against this, instead proposing the opposite position that only with faith in the resurrection is an ethical life possible. However, I believe that recent thinkers are rigorously consistent when they argue that self–sacrifice is supremely good only if death is final and unrewarded. So in exalting resurrection, I will have also to deny that self–sacrifice is most paradigmatic of the good. And this is what I shall now proceed to do, arguing that this idea is incoherent, actually unethical, and not at all a translation of the essence of monotheistic tradition as some tend to claim. To make this argument, I will examine in turn the four components of this ostensibly pure sacrifice: 1) gift without return or "unilateral" gift; 2) death as grounding the ethical; 3) a subjectivity as constituted through sacrifice and the demand of a God beyond being; and 4) ontology without resurrection or eschatological overcoming of death.

The notion that a sacrificial offering without hope of return is the only true gift suggests that to be ethical is to be prepared to lose oneself for the other. This is purported to be an improvement over the ancient Greek idea that to be ethical is to value as the only source of secure happiness that which cannot be taken away from one, such as, for example, a simple, ordered, tranquil life, passed mainly in contemplation and the enjoyment of secure friendship—a life relatively immune to disaster. I want to argue, on the contrary, that both ideas are equally solipsistic. The one thing about ourselves we know with certainty is that we are to die. When we accept this death, or prepare ourselves, if necessary, actively to appropriate it, we fulfill most rigorously the Greek demand to value only that which cannot be taken away from us. We do so, it is true, in a somewhat paradoxical manner; that which most securely defines us—death—is that which puts an end to us, while the moral gesture which supposedly establishes our subjectivity, and so is inalienable, involves our being drawn beyond our own boundaries. Nevertheless, one might suggest that pure self–sacrifice strangely turns out to be the securest self–possession, and so one might wonder whether, after all, there is something stoically solipsistic about this ethic despite its being founded upon a disinterested regard for the other.

Let us reflect further upon these claims about self–sacrifice. It is thought that one can never observe another’s subjectivity, but can only glimpse a "trace" of it in his or her pain. It would follow from this notion that one acknowledges the other as other only when one sacrificially responds to that pain. This means that we only acknowledge the reality of the other person, and then negatively, when we can no longer be in communication with him. But a person whom we cannot see or talk with is an unknown and indefinable other, and therefore only a generalized other. And here is where the problem sets in.

For a generalized other is a totalized other, an other reduced to ourselves, since we can only imagine it by projecting our own subjectivity upon it. To die for any old invisible other is the very reverse of valuing otherness, because otherness must involve not just diversity and difference but specific diversity and concrete difference. All these things have to be visibly or audibly or in some way sensorially registered.

I am also doubtful about the claim that to gaze upon something is to reduce it to the terms of our understanding or of publicly available linguistic categories. Those who hold this view are the victims of a transcendentalist dimension in phenomenology which, in the tradition of Kant, reduces what appears to that which our understanding can master. It strikes me rather that the very specificity of an object, its very ability to arrest our attention, is constituted by a depth it withholds. We can never see every aspect of a thing, nor know how it would respond in every conceivable circumstance; yet without knowing those responses, we do not know all the different truths they would disclose. If this is true of objects, then it is all the more true of other human subjects. We cannot look at anything, especially not at human beings, without ourselves being regarded in turn from an unknown depth. This depth is not necessarily contrasted with the surface of the thing, since even surfaces tend to exceed our categories: we never feel our words exactly capture a rainbow, for example. So I disagree that what is apparent, what makes itself present, is thereby reduced to what we understand it to be.

We have seen that, in much recent ethical thought, a self–sacrifice is supposed to acknowledge the other through a response to his or her pain, without reducing that other to our understanding of him or her. And yet the effect is actually the opposite—self–sacrifice is that which is most inalienable and so remains within the circle of our self–identity, as does also the generalized other which can only be a projection of our self. On the other hand, we have seen that a person with whom we can interact and whose concrete presence we enjoy is able to be genuinely other, and in a positive way, rather than just as a victim or as suffering. In fact, it is only in this positive sense that someone cannot be reduced to our self.

So, if attention to the other is central for a sense of the ethical, it would appear that convivial enjoyment of another is more important than suffering on his behalf. Moreover, if a person can only be known as other via communication, then I cannot remove myself as a participant in this situation. The German Roman Catholic philosopher Robert Spaemann has expressed this point very well: giving food to those in need, he observes, can occur as a one–way gift from those who have to those who have not, or it can occur in a feast, where all eat together. In the feast egotism is mitigated, since here one eats only if one eats along with others; and yet at the same time one does eat, and so selfhood is not eradicated. This image of the feast suggests for Spaemann that what is supremely good is the ecstatic—not in the sense of departing from life, but in the sense of living life as departing from oneself while in this very departing receiving oneself back again. In other words, beyond the ancient Greek quest for happiness in security, he proposes living convivially through generosity to the other and through receiving back again from the other.

Now all this is of course not to deny that to preserve conviviality, to preserve the spirit of feasting, one may very often have to make one–way gestures, without apparent return. Indeed, one can go further to say that in a corrupt, fallen world, the only way to the recovery of mutual interaction will pass through sacrifice unto death. But the point is that this sacrifice is not in itself the good, but rather that which sustains a road to the good in adverse circumstances. If one values every single individual as unique and irreplaceable, and if one’s image of the good is of the widest possible conviviality, then in order fully to aim for the good, even the sacrificial offering of oneself must sustain the hope of one’s own ultimate redemption. I myself am unique and irreplaceable; without oneself, as without anyone, the universe would have lost something good.

What I am suggesting here, therefore, is that if the fullness of being, or of convivial interaction, defines our vision of the perfect good, then giving can be conceived as quintessentially reciprocal; expecting a gift in return need not necessarily diminish the gratuity of a gift. But in that case, one might very well protest, what precisely distinguishes a gift from a contract? Does it not make giving into a sort of informal and somewhat self–deceptive contractual arrangement, rather like an exchange of business lunches? Not exactly—giving is more unpredictable. If we sign a contract we know what we will get back and probably when—we also take it that what we receive in return is, according to some public measure, equivalent. But if, for example, at Christmas, we exchange gifts with a friend, although there is reciprocity involved, there is also asymmetry: what we receive in return may often surprise us, and whether it is equivalent will be a matter of fine judgment. Indeed, very many different modes of equivalence might prove acceptable here. Furthermore, a return gift may be for a long time delayed, and we will require no exact guarantee of when it is to be returned—as for example with an invitation to dinner. When such a gift is returned, it will certainly in many ways repeat the initial gift—the same hour of the evening perhaps, the same sitting at a table, the same number of courses. But it will also be repeated non–identically—the menu will be different, at least in our culture. Indeed, if one were invited back to dinner immediately, the very next night, and presented with the same menu, one would be offered not a gift but an insult.

This suggests that the content of a gift may as much deem "giftness" as its circumstances (i.e., how free, how unconstrained it is, with what expectations it is given). Modern thinkers, however, tend to concentrate wholly on the formal circumstances of the gift, not on what is given. But suppose a wealthy dying man whom you knew would not live to enjoy a reciprocal invitation invited you to a meal but presented you with only a piece of stale bread and butter. Although the formal circumstances are correct, would you suppose that to be a gift?

My claim here is that asymmetrical reciprocity and non–identical repetition allow sufficiently for an element of freedom in gift exchange to distinguish it from contract. Within limits, at least, the recipient of the initial gift may choose what to give back and when. Certainly, it remains the case that the initial giver and society in general exert pressure on the recipient to give back, in the name of justice. Nevertheless the coercion is mitigated in that if he fails to give back, any punishment he would receive would also fall outside the fixed contractuality of law. For a long time he may be punished with more pressure, more gifts. But eventually the giver will judge that his generosity should be diverted to other, more promising causes, and then the defaulter will finally receive the logical consequence of his refusal: isolation.

At the same time, however, it would be true to say (and I am indebted here to discussion with Jean–Luc Marion) that the very components of non–identity and delay in gift exchange involve a certain surplus of unilateral giving over reciprocity. I do give without the guarantee of return, and if my gift differs from the return gift, then it would seem that something unique has passed from me which does not return. Hence Aquinas insists that gift involves a notion of the unreturnable, even though he also asserts that the ultimate blessedness of charity involves reception as well as giving. However, in the first case of the unreturned gift, one could say that there always remains a hope for a reciprocal gesture, as there is in an eschatological reserve: there will always be self–sacrificing in this life, but in hope of the eternal banquet.

In the second case, of giving something which I do receive back but not in the same form, there is an element of unilaterality, but this is connected to the fact that for a gift to remain a gift, it must change throughout its passage. For as soon as something passes into someone else’s hands, it is marked by their character, by their usage; it has become something different in a sense, insofar as the gift–giving succeeds in establishing understanding between giver and recipient. It follows that a return gift (which may only be that of gratitude) would further unfold a mutual understanding, so although the thing which one receives back is in the most obvious sense different from that which was first given, in a deeper sense the reciprocal gift returns the same gift of mutuality that one had first offered. A gift to remain a gift must continuously alter, and this altering is essential to exchange; but at the same time, without the exchange of gratitude a gift is unrecognized and therefore obliterated in its effective actuality.

So far I have argued that exchanging gifts more enshrines the ethical than does a one–way giving that is indifferent to return. In addition, however, I want to claim that the sheerly unilateral gift is a barely coherent notion. In some sense, at least, the free, one–way gift, although it supposedly defines the good in modern ethical thought, is impossible and cannot occur. For as many have noted, even to have the consciousness of being a giver is to reward oneself for giving and to cancel the gratuity of the gift. We may be genuinely disinterested, but we cannot escape the fact that if this disinterest is for us a value, we shall experience our disinterestedness with satisfaction. Since one would seem to fall prey to the same trap were one even to aspire to giving, it seems hard to understand how aspiring to self–sacrifice is any better. Such aspiration is equivalent to an endless deferral of the gift, except, we are told, at the point where we pass beyond ourselves in death and cannot receive our own death back again. However, the formal circumstances of the sacrificial death are not enough to make a purely one–way gift possible. For the gift to be truly disinterested, the giver of his own life must not be able even to imagine the future pleasure of its recipients.

Therefore the true gift would have to be to an absolutely anonymous other—paradigmatically the enemy, suggests Jean–Luc Marion, echoing the Gospels. In addition—and again to prevent anticipation of the other’s satisfaction—the thing given should possess no content outside the gesture of giving itself; otherwise we could be pleased that the recipient will at least possess the content we passed along (e.g., "at least he’ll always have his health"). We have already seen, however, that removing the question of the appropriate content of the gift leads to absurd results. Nevertheless, the offering of death would seem to approach most closely to the contentless gift. For example, one might ask: In dying, just what did Jesus Christ offer all those infinite numbers of people unknown to him? However, even if one is prepared to offer death for unknown others, this still does not establish a one–way gift. Before dying, the living subject will imagine that somebody benefits from his death, and will be reimbursed by the knowledge that his death is significant in some fashion or other. As a result, the giving ethical subject only becomes purely giving, and therefore ethical, once he is dead and has ceased to be a subject at all.

It has been seen then that the first and most crucial component of the notion of pure sacrifice—namely, one–way giving—not only fails to define the ethical, but is also scarcely coherent. And we have already seen how the second component, death as condition for the possibility of morality, fails to grasp the priority of intercommunion in defining the good. But one should note here, in addition, that if the ethical only arises in response to that fragility which in extremis is the death of the other, then the ethical is, ontologically speaking, something merely secondary and reactive. Far from it appearing to be the case, as some would wish, that the good lies beyond being, the good would on this construal rather seem to be distinctly less than being. In other words, only when being begins to suffer does it instigate the good.

But here one might suggest that a vision of morality as a reaction to the threat of death is less a transcription of monotheism than a reversion to the heroic morality of Homeric times. Against this morality, Socrates, in Plato’s Phaedo, insists that warriors who die for the city out of fear for their own death or the death of others in the city, or fear of loss of honor, are sacrificially trading a lesser fear of dying in battle for a greater fear of shame, loss of nobility, and the loss of the city itself. By contrast, says Socrates, the philosopher is a person who begins with absolute confidence, with a vision of eternal truth, goodness, and beauty, and with his own psychic kinship to these abiding forms. For this reason, the philosopher can act positively, truly without fear even unto death (avoiding the merely apparent fearlessness that is in thrall to an even greater fear). He is good, not primarily as acting, but as knowing, or as receiving and recognizing the realm of the forms as that which is most real. No sacrifice is involved here, since the body and lower passions given up are less intense degrees of being, truth, and goodness; the absolute degree of these things includes the reality of the lower, and so nothing is truly lost.

Thus if the good is primary—if as for Platonism and the monotheistic faiths what is, is good—the good is not first occasioned by death.

The third component of the notion of pure sacrifice is the idea that subjectivity itself is constituted through the "persecution" of my consciousness by the demands of the vulnerable other. Here again, I have already enunciated my main response: this tends to render the personal impersonal. For if my ethical response to the sorrows of another precedes my exercise of judgment, I respond in the same way to all ostensible sorrows, whether authentic or not, self–indulgent or not, self–caused or not. I would have no means of knowing whether this persecution was in fact simply a confidence trick on the part of others, universally pretending sorrow in order to win power over me. If, by contrast, one is to respond to real historical persons, then one must first distinguish the legitimacy of their claims, the specificity of their needs. This being said, it goes without saying that to regard God as a big Other shadowing the small human other is simply to make an idol out of generalized and drained subjectivity. Since God is not just another person, but the fullness of personhood and being in whom we participate, He is as much not–other—non aliud as Nicholas of Cusa put it—as other.

This leads naturally to the fourth and final component of the idea of pure sacrifice: the ontological vision which sees Being without immortality of the soul or resurrection of the body. Only this vision, according to our modern thinkers (though not, of course, the Catholic Marion), ensures that the Jewish and Christian imperative to self–sacrifice can enjoy a purified fulfillment.

What the modern presentation of this ontology tends to overlook, however, is its profound link both to the antique pagan polis and to the modern secular state. We need to comprehend the first link if we are to understand something of the process by which sacrifice was displaced from an actual bloody ritual practice to a metaphor for moral action, such that we can now say "he was very self–sacrificial" and not mean "he offered his body to Aztec priests." What, in part, assisted the transition to metaphor was the way in which the death of the hero for the city came to be construed by the Greeks and the Romans as equivalent to ritual sacrifice, and indeed as rendering the hero himself a fit recipient of sacrifices in turn.

There is a notion here that the hero’s life is subsumed in a greater social whole, losing himself without return except for posthumous praise or the celebration of his bravery. The Greeks did not believe the hero continued to live, save in the rather shadowy intimations of an afterlife in Hades. But modern secularity gets rid of even such intimations, and so perfects this pagan logic of sacrificial obliteration of oneself for some ideal, or for the State, or for both. Such a logic elevates as supreme the abstract notion of the perpetually abiding nation–state, outlasting its citizens and being more valued than the lives of individual humans. Sometimes this last aspect is disguised in the form of a "sacrifice for future generations," but since every generation should logically be subject to the same imperative to sacrifice for future generations, in no generation will the people benefit from the sacrifices of those before. Consummation of the sacrifice, then, is forever postponed.

To espouse these values is essentially to perpetuate the positivist exaltation of "altruism," of surrender of self for the future, for science, for the state. If the ethical imperative is that we should offer our lives to an other who is present to us only as a trace (as a sufferer, as we said above) and not in visibility, then, as I have shown, this other is an anonymous and therefore generalized other. Thus we live under the ethical sway of a law of abstract otherness, mirroring in the ethical realm the legal assumptions of the modern liberal state, which enjoins a merely abstract respect for the rights of the individual in general with indifference to that individual’s gender, character, or cultural specificity.

Given the assumptions of such a state two things follow: our responsibilities tend to become unlimited because we owe our lives infinitely to every other person; and the ethical good never arrives—we can never fulfil this impossible responsibility, and no one could ever legitimately relax and enjoy the benefits of the sacrifices of others. Thus the only thing that is achieved is the continued carrying out of self–obliteration. Liberals pretend that continuous self–obliteration is the demand of the moral law, but in reality it is only the demand of the liberal state, which cannot put a brake upon sacrifice because it is unable to promote any positive goals or values that would define true humanity. It follows that the exaltation of pure self–sacrifice for the other is secretly the sacrifice of all individuals to the impersonality of the formal procedural law of state and marketplace. Like the antique polis, this alone abides, this alone is eternal.

Within the ethical thinking regarding pure sacrifice that I am opposing, one’s decision to be responsible for this person rather than that appears to be entirely arbitrary. As Derrida puts it, Why look after this cat rather than all the other stray cats? However, he is surely overlooking here the limitations of liberal politics: if to be good was not, as for liberalism, to exercise a generalized responsibility, but rather, as for antiquity and the Middle Ages, to perform excellently a particular social role which helps to achieve, in coordination with other performed roles, a specific concrete social telos or end, then I can look after the cat assigned to me in the secure knowledge that other people are looking after theirs. Moreover, if this telos is taken as reflecting a transcendent reality, then the individual carrying out his role provides insight into this transcendent that others can learn from.

We can see from this that while, on the one hand, the logic of pure sacrifice upholds the law of an impersonal collectivity, on the other hand it is too individualistic and has no account of the good as achievable through coordination. For modern ethical thinkers, indeed, the tension is resolved only through death, when at last that which I alone can responsibly take on—my own death—is also recognizable as fulfilling the law of public responsibility. But this is only because, as we have seen, in fully possessing myself in death, I pass beyond myself into public, impersonal indifference. By contrast, if we allow for a good achievable through coordination, it is possible to exalt not just self–offering, but even a joyful attention to the infinite presence of a living, visible other above the social whole.

We may be confident that a society based on such coordination would be made up of an infinite series of joyful relations to each other. But that is to say that to carry out the ethical project requires a community collectively aspiring to enact charity. One needs to realize here that before early modernity, when it was reduced from a moral imperative to a private task of one–way giving, charity in the Middle Ages was a reciprocal state of being that persons had to enter into with familiar others, with adopted kin under God, and with friends with whom they were conjoined in a common purpose. Charity was not something for me, privately, to perform, but an entire network of complex reciprocity.

Of course, even a community aspiring to this would be utterly riven by fractures and failures. Much of the time one would be conscious that others were failing to perform their role, and failing to exert any sort of charitable preference. And often, indeed, failures of some would require sacrifices from others, even unto death. But it is at this point that faith in resurrection doubly sustains the project of a charitable society, founded on the widest extension of reciprocity. First of all, because of this faith, one can have hope for the victims of the failures of others; and secondly, in the case of necessary self–sacrifice, one need not surrender oneself to the consuming totality. In either case, one need not embrace the logic of ultimately necessary self–sacrifice without return, either of others or of oneself. If this is true, then only the vision of the eschatological banquet could be an image of the good, whereas the image of dying for the other—though it is the advent of the good in fallen time—cannot itself be the final good, without once more subordinating the person to an impersonal totality, in this case an abstract moral principle.

If, as in this modern ethical vision, there is no resurrected return, then we have to accept that there will be, eventually, nothing more to be said to anyone. But that means that towards all those we have harmed and wounded and then lost without reconciliation, we can only rehearse an empty gesture of private, nominal apology. They can never appear to forgive us, just as those who have injured us and vanished can never appear to be forgiven.

Without resurrection, there can never be any final reconciliation. But in the absence of reconciliation, or of hope for that, neither can there be any morality. For where I cannot be reconciled with the lost one I have injured, I owe him an infinite debt of mourning and regret. So great a debt do I in fact owe, that my energies cannot legitimately be freed up to perform my duties towards the living. But those demands of the living also are infinite and infinitely legitimate, and so, here indeed, without resurrection arises an irresolvable problem: I should not cease mourning and apologizing, and yet I should. Only the hope for an infinite community of all who have ever lived frees us from this dilemma, again to do good.

And so we must finally conclude that resurrection, not death, is the ground of the ethical. They are wrong who claim that in Luke’s Gospel a promise of eternal reward contaminates an injunction to unilateral giving. A wider reading of the New Testament, especially John and Paul, suggests that such injunctions are only a moment of eschatological delay within a wider promotion of gift–exchange beyond the fetishized limits upon such exchange imposed by most ancient societies. The name of the Holy Spirit himself as "gift" is after all bestowed not only to denote a pure one–way gratuity, but also because the Spirit expresses the infinitely realized exchange between Father and Son. And if to be resurrected is to be bodily incorporated into the life of the Trinity, resurrection is not an extrinsically added reward for successfully giving without return. Rather, for the Christian, to give is itself to enter into reciprocity and the hope for infinite reciprocity. And to offer oneself, if necessary, unto sacrificial death is already to receive back one’s body from beyond the grave. To give, to be good, is already to be resurrected.


John Milbank is Francis Gall Professor of Philosophical Theology at the University of Virginia. His most recent books include The Word Made Strange, The Mercurial Wood, and Radical Orthodoxy (coeditor).