Paimana

Fourth of Muharram, 1448 AH

I hope the connection between the two halves of my title, the necessity on the one side and the will on the other, will become clear as I go along.[^1] I want to look at one case, a very old one, and about a problem it raises that I don’t think is usually put in quite the way I am going to put it. The case is the death of al-Ḥusayn b. ʿAlī at Karbalāʾ, in the year 61 of the Muslim calendar, and the problem, as it happens, is a problem about a verb.

The tradition that remembers him leans, at the decisive point, on the word chose. On the night before the killing he gathered the people who were with him, told them that the enemy wanted no one but himself, lifted from them any obligation to remain, and had the lamps put out so that anyone who wished to slip away in the dark could do so unseen. And he stayed. Now, the staying is the thing we are supposed to honor. But notice what the honoring seems to require. It seems to require that he could have given his allegiance to the other side and did not, that there was, as people say, a road not taken. If there was no such road, one wants to know what, exactly, we are praising.

Here is where the difficulty starts. The very same tradition is absolutely insistent that he knew. He had been told, the reports say, before he was born; a handful of the earth of the place had been carried to his grandfather against the day it would turn to blood; and when he reached Karbalāʾ he is supposed to have recognized it by its smell and named it, on the spot, as the ground of his own slaughter.[^2] So he is not a man stumbling into his fate. He sees it coming from a great distance and walks toward it. And there is a second thing, which is heavier than the first. He is maʿṣūm. Now, ʿiṣma, infallibility, is a technical notion in this tradition, and I’ll come back to what it does and doesn’t mean; but on the face of it, it is the claim that the Imam does not, and on the strong reading cannot, will against the good. So now we have foreknowledge on one side and infallibility on the other, and between them they seem to close off precisely the space in which the verb chose was supposed to do its work. The freest act the tradition knows turns out, when you look at it, to be the act with the fewest alternatives. That’s the problem. I should say at the outset that I am not going to give you a theory of free action, or a theory of infallibility. I don’t have one. I want to give a picture, and I’ll try to persuade you that it is a better picture than the going alternatives, and, what is a different matter, that it is true.[^3]

Let me first set out the worry as an argument, because I find that helps. Something like this:

(1) An agent deserves the special honor we give to a free act only if he could have done otherwise.
(2) Al-Ḥusayn knew, with certainty, that he would die if he stayed.
(3) Al-Ḥusayn was maʿṣūm, so that he could not have chosen the wrong.
(4) Therefore al-Ḥusayn could not have done otherwise, and the ground of the honor is removed.

I have stated it baldly, perhaps more baldly than its friends would like, but I think this is what is bothering people, whether or not they put it this way. And I want to say, first, that premises (2) and (3) are doing very different kinds of work, and that one of them — the one that looks the more menacing — is not really menacing at all.

Take (2). The thought is supposed to be: he knew he would die, therefore his dying was unavoidable, therefore he could not have done otherwise. But that is a fallacy, and it is, in fact, a fallacy I have spent a good deal of time on in another connection. It is the very same slide that leads people to suppose that whatever can be known in advance, or known independently of how things turn out, must therefore be necessary. People run together the epistemic and the metaphysical, just as they run together the a priori and the necessary; and these really must be kept apart.[^4] That someone knows that p — even that God knows it, if you like — is a fact about knowledge. Whether the agent could have done otherwise is a fact about the agent and his powers. And there is simply no valid step from the first to the second. The knowledge tracks the act; it does not manufacture it. If I know, watching you, that you are going to raise your arm, my knowing does not lay so much as a finger on your arm. Make the knower omniscient and the point is unchanged; you have made the knowledge more certain, you have not made the deed more compelled. So I am inclined to set (2) aside altogether. Foreknowledge, by itself, closes no road. The whole weight of the problem falls on (3), on the ʿiṣma, and that is where I want to spend my time.

Now what does (3) come to? Here one has to be careful, because there are two things ʿiṣma might mean, and the difference between them is the difference between a real problem and no problem at all. On the first reading, to say that the Imam is maʿṣūm is to say that he is so constituted that he will not sin, though the power to sin remains in him; the wrong is within his reach and he never reaches for it. On the second reading, it is to say that he cannot sin, that the willing-against-the-good is not in his repertoire at all, that the alternative is closed by what kind of being he is. If the first reading is right, then he retains the alternative and simply never takes it, and the honor has something to attach to. If the second is right, then the alternative is gone, and gone by the nature of the agent rather than by any external bar, and we are back in the soup. So the question is which of these the tradition means, and, more importantly, which of them is true of the case. I’m going to argue, eventually, that the right thing to say is more interesting than either, and that once we have it the dilemma dissolves. But let me approach it the way I think one should, by looking at the two obvious ways out and seeing why neither is good enough.

The first way out is the most ingenious, and it comes from the philosophy of action rather than from theology. I have in mind the line of argument that Harry Frankfurt opened some years ago against the principle I stated as premise (1) — the principle, as it’s called, of alternate possibilities.[^5] Frankfurt asked us to imagine a man, call him Jones, who is going to do a certain thing, and a second party, call him Black, who badly wants Jones to do that thing and who has rigged matters so that if Jones should show the slightest sign of deciding otherwise, Black will step in and see to it that Jones does it anyway. But as things go, Jones decides and acts entirely on his own, for his own reasons, and Black, having no occasion to lift a finger, never lifts one. Now: Jones could not have done otherwise, because Black was lying in wait. And yet, Frankfurt says, and I think most people share the intuition, Jones is responsible for what he did, and responsible exactly as he would have been if Black had never existed, because the act came out of him in the ordinary way and Black played no part in the actual sequence of events. The intervener was there, but idle.

You can see the application, and it is closer than an analogy; it is almost the same structure. Read ʿiṣma as the idle intervener. The Imam is so kept that were he to incline toward the sin, the inclination would not be allowed to issue in the deed; but the safeguard never operates, because the Imam, of his own motion and for his own reasons, inclines to the good and does it. The infallibility is a watchman who is never needed. On this reading al-Ḥusayn is responsible for the staying, and honored for it, precisely as he would be if no ʿiṣma attended him at all, because the staying came out of him in the ordinary way; that the alternative was barred is true and beside the point. It was barred by a guard who never had to act. And I should mention that this is not some exotic theological construction. People working on divine foreknowledge in the ordinary philosophical literature have themselves been driven to put God in Black’s place, God being the one party who could have certain knowledge of what a free agent will do and stand ready accordingly.[^6] So the theological case is not off at the margin of the debate; the debate has wandered toward it of its own accord.

Now I think this argument is good, as far as it goes, and I do not want to be read as simply dismissing it. But it is not enough, in two ways, and the second is the one that matters. The first is internal trouble. The defenders of the old principle have not lain down; they say that even where the intervener guarantees the outcome, the agent keeps some shred of an alternative, if only the bare flicker of beginning to incline otherwise — the very flicker that would have set the intervener going — and that this flicker, faint as it is, is enough to keep the principle alive.[^7] And behind the flicker there is a sharper objection, a dilemma: either the tie between the agent’s prior state and his act is deterministic, in which case the example just helps itself to what it was trying to prove; or the tie is loose, indeterministic, in which case the intervener cannot know for certain what the agent will do, and a real alternative creeps back in. I won’t try to settle this here. I only want to record that the matter is not closed, and that a defense of al-Ḥusayn’s freedom that rested on Frankfurt’s cases would inherit all of this unfinished business.

But the deeper trouble is that even if the argument succeeds completely it gives us the wrong thing. Suppose it works perfectly. What has it shown? It has shown that the absence of an alternative is no bar to holding al-Ḥusayn responsible, no bar to honoring him. It has not shown the thing we actually care about, which is that the act is supreme. The Frankfurt move clears an objection out of the way; it does not put anything in the objection’s place. It tells us we may go on praising the staying despite there being no road away from it, and it leaves entirely untouched the question of why the staying is the most exalted act the tradition has to show. For that you need an account of how a particular act of will, in losing its alternatives, becomes most completely itself. There is such an account. As it happens it is also Frankfurt’s, but it comes from the other side of his work, and I’ll get to it. First let me deal with the other obvious way out, which I think is worse, though it often passes for depth.

The other way out is to refuse to choose. One says: the tension between the infallibility and the choosing is not to be resolved but lived with; the tradition wants him bound and free at once, foreknowing his end and yet meriting it by his choosing, and since these cannot both be had cleanly, the honest thing is to hold both and call it a mystery, or a paradox at the heart of the matter. Now, I have a great respect for genuine mystery, but I do not respect this, and I’ll tell you why. There is, by the way, a perfectly good theological precedent for the maneuver, and it is instructive. The Ashʿarī theologians, faced with the parallel need to keep God the maker of all things and the human being nonetheless answerable for his deeds, produced the doctrine of kasb, “acquisition”: God creates the act, the man “acquires” it, and is judged on the acquisition. And it has been the nearly universal verdict, inside the school and out, that kasb does not solve the problem but only gives it a name; that it differs from outright determinism only in its wording.[^8] I am reminded of certain theories of meaning I have discussed elsewhere, which give you a word for the thing you wanted explained and send you away feeling you have been told something. One could, by the same trick, declare al-Ḥusayn both unable to do otherwise and fully the author of his deed, and decline to say how, and dress the declining up as a refusal of false precision.

The reason I don’t accept this is not that paradox is always evasion. It is that this paradox does not fit the case. Look at the man the reports describe. He names his own death saʿāda, felicity, and says that the only weariness is life among the oppressors.[^9] That is not the voice of a man trapped between two horns, weighing alternatives he cannot escape; it is the voice of a man who wants the thing, who wants it with the whole of himself. And a reading that leaves the tension standing as a brute paradox leaves precisely this unexplained. It can tell you why he did the deed (he could not do otherwise) and why we may still credit him with it (the watchman never had to act); it cannot tell you why the deed wears the face of joy instead of the face of resignation. The unresolved reading is not too humble; it is inattentive to the very thing it claims to revere. And the feature it cannot accommodate is exactly the feature that the right account is built to explain.

So let me give the account. The instrument is again Frankfurt’s, but it is the Frankfurt who wrote about the will and what we care about, not the Frankfurt of the counterexamples; and the shift is the whole point. The idea is that a person’s will can be bound by what he most deeply cares about, in such a way that certain courses become, for him, impossible — not because anything outside forbids them, and not because he lacks the bodily or logical capacity for them, but because his will is so wholly formed around what he cares about that he cannot bring himself to act against it. Frankfurt calls this volitional necessity, and his standing example, which I cannot improve on, is Luther: Here I stand; I can do no other.[^10] Now, Luther could, in every ordinary sense, have recanted. Nothing held his tongue but his own will. When he said he could do no other he was reporting not a chain on his wrist but the condition of a will that had become incapable of the alternative because the alternative had become unthinkable for the man he was. The necessity is perfectly real, and it is a necessity of the will. And here is the thing Frankfurt saw, which I think is exactly right: a necessity of this kind is not felt as a chain at all. It is felt as freedom, because the man so bound is doing the very thing he most wants, with nothing foreign pushing him and nothing left over in him to resist.

What this displaces is a certain picture of freedom — the picture on which to be free is to stand at a fork with the will hanging undecided between the branches, equally ready to go either way. That used to be called liberum arbitrium indifferentiae, the freedom of indifference, and on the present view it is not the height of freedom but a mark of a self not yet finished, a will still shopping among the options because it has not yet become anything in particular. The freedom that matters is the other thing: the whole-hearted settling of the will on what it cares about, the closing of the gap between what the man is and what he does, until there is no daylight left in which an alternative could stand. And I want to be careful here, because this is easy to misstate. I am not saying the man is robbed of his alternatives and then handed a consoling redefinition of “free.” I am saying the alternatives fall away as a consequence of the will’s becoming entire — the way a settled love does not deliberate about betrayal, because the lover is no longer the sort of person for whom betrayal is a live road. The freedom of indifference is the freedom of not yet caring. The freedom of volitional necessity is the freedom of caring completely. Those are very different things, and I think people have run them together, and have their intuitions, as I once said in another case, exactly reversed.[^11]

Set ʿiṣma against this and the dilemma I started with turns inside out. Recall the two readings, the “will not” and the “cannot.” It turns out the tradition’s own dominant construction is, in effect, the volitional-necessity reading, and not the bare “cannot.” For the standard Imāmī view — against an Ashʿarī view on which God simply does not create the sin in the Imam — is that the Imam retains the power to sin and does not exercise it; that ʿiṣma is a kind of grace, a luṭf, which renders him immune from the sin while leaving the capacity for it intact; and the theologians are explicit that the capacity must be left intact, since an Imam from whom the power had been confiscated would have lost the freedom on which his whole standing rests, and the doctrine would collapse into the very determinism it was meant to avoid.[^12] Now that is the structure of volitional necessity and not the structure of compulsion. The maʿṣūm is not a man from whom the alternative has been taken away by an outside hand. He is a man in whom the alternative has died of the will’s completeness — a man for whom the sin, fully within his power, is simply not within his range; unthinkable for the person he is, in just the sense in which recantation was unthinkable for Luther. One of the great commentators, pressing for the inner character of the state, describes it as a condition in which there is left no desire except the desire of God, reached through an excess of love that makes the very contemplation of sin a kind of shame.[^13] It is a will so taken up by what it loves that hemming has become beside the point.

If that is right, then the foreknowledge and the infallibility are not the two walls closing off the room in which al-Ḥusayn’s freedom was supposed to live. They are the conditions under which his willing becomes volitionally necessary, which is to say the conditions of its highest freedom and not the cancellation of any freedom at all. His I can do no other, spoken not in words but in the staying, and in the naming of death as felicity, is Luther’s sentence carried as far as it will go: the sentence of a will so transparent to the good it loves that the good’s necessity and the will’s freedom have stopped being two things.

Now let me make a move that I think strengthens this, and that comes from the part of philosophy I usually work in. People will want to resist by saying: but surely he could have given the allegiance; surely there is a possible situation in which he does. And I want to ask, in my pedantic way, what such a situation would be a situation about. Imagine a blatantly fictional situation. Suppose that on the night of the ninth al-Ḥusayn had given the bayʿa to Yazīd and ridden home to Medina. Have I described a situation in which he does otherwise? Or have I described a situation about somebody else? This is not a frivolous question. We do not, after all, find out about such situations by peering at them through a telescope, discovering, out there, a faraway al-Ḥusayn doing the deed; we stipulate the situation, and then it is a perfectly good further question whether the man in it is the man we started with.[^14] And if willing the good wholeheartedly, refusing the unjust ruler, is essential to who this man is — if it belongs to him the way, I have argued, his origin belongs to him, so that this very man could not have sprung from a different source — then the situation I described, however fluently I describe it, is not a situation in which al-Ḥusayn gives the allegiance. It is a situation about a man who resembles him up to a point and then diverges, which is to say a different man, or no one in particular. The “alternative” was never a genuine alternative for him. I don’t insist that one is forced to read the essential properties this way. But I will say that those who confidently assert that of course he could have done otherwise are helping themselves to a transworld identification that they have not earned, and that the intuition cuts, if anything, the other way.

I want to forestall a misunderstanding, and the forestalling is, I think, where the real content of all this is kept. Someone will by now have a word ready, and the word is amor fati, the love of fate; and the temptation to hang it on al-Ḥusayn is strong, since here too is a man who wills what must be and seems not merely to submit to it but to love it. I think this is the wrong description, and getting the description right matters as much here as anywhere. Nietzsche’s amor fati is the formula of a man who has decided that the world has no point beyond itself and who proposes to love it anyway, to want nothing other than what is, and to take the strength to do so as the measure of greatness.[^15] The whole force of it is that there is nothing behind the necessity to redeem it; the love is a love of fate as fate, a yes flung at a universe that gives no reason for the yes, and its dignity is the dignity of a self strong enough to affirm without being given anything to affirm. The old Stoic version is gentler but built the same way: Epictetus telling you to will events as they happen rather than wishing them to happen as you will, the image of the dog tied to the cart who follows along willingly or else is dragged, but arrives either way.[^16] Both teach the willing of the necessary, and both locate its worth in the willer’s stance toward a necessity that is just there.

Al-Ḥusayn’s willing of the necessary is neither of these, and the difference is not a matter of degree. It is not the affirmation of a fate that means nothing, because the necessity he wills is the decree of a God to whom the act is owed and to whom it returns; the willing is obediential all the way down, a yes said to someone and not into a void. And it is not the Stoic’s apathy, the draining-off of the passions in favor of assent, because the man who calls his death felicity also weeps; the reports show him consecrating his grief, offering the whole of his feeling rather than emptying himself of it. The necessary, for him, is necessary because it is willed by the Beloved, and that single fact changes what the love is. Amor fati becomes, in his hands, not the love of fate but amor simply, love of the One whose decree the fate is, the necessity loved only as the shape the will of the Beloved has taken. The willing of the necessary turns out to be a mode of love and not a mode of strength. And that, finally, is why it can wear the face of joy that the brute-paradox reading could not account for. Joy is what whole-hearted love looks like when it meets what it loves, even when what it loves arrives as a death. One does not get to gladness in the face of Karbalāʾ by an exertion of fortitude. One gets there by loving completely enough that the will and the good have become a single thing — which is, give or take the theology, the same arrival Frankfurt was reaching for with his man whose will is bound by what he cannot help but care about.

Let me come back, at the end, to the verb I started with, because the question was what is left of chose once the alternatives are gone, and I think the answer is that the verb survives, and is in fact strengthened, the moment we stop taking it to assert a claim it was never required to make. When the tradition says that al-Ḥusayn chose, it does not need to mean, and ought not to be taken to mean, that he stood at a fork with the road to allegiance genuinely open under his feet and picked the other branch. It means that the staying was wholly his, that it came out of the deepest constitution of his will and carried his whole weight, that in him, at that moment, the willing and the being had become one act. The dignity is not the dignity of a refusal he was free to make and graciously declined to make. It is the dignity of an embrace: of what was already fixed, taken up and made his own so completely that the fixity and the freedom are the same fact seen from two sides. And I want to insist that this is more than the freedom of alternatives and not less. The freedom of standing undecided at a fork is the freedom of a will that has not yet become anybody; the freedom of the bound will is the freedom of a self so entire that it has passed clean out of the reach of the alternative. The first is the freedom of the man still deliberating. The second is the freedom of the man who has finished deliberating because he has finished becoming himself.

There is one last thing, and it bears on what the season is actually for. The tradition has always refused to let the defeat at Karbalāʾ count against the act — has insisted, against every appearance, that the slaughter was a victory — and this can look like piety simply overriding the evidence, since the camp was destroyed and the heads were carried to Kūfa, and by every worldly measure it was a loss. But the philosophy of what gets called moral luck gives the refusal its warrant. Nagel and Williams pointed out how much our judgment of a man is hostage to how things happen to turn out, to a luck in the result that he does not control: the reckless driver who happens to hit no one judged more gently than the one who happens to hit a child, though their wills were identical.[^17] The tradition’s verdict on al-Ḥusayn is, in effect, the flat denial of luck in the result. It holds that the worth of the act was never in its outcome and never in its alternatives, but only in the willing — in the whole-heartedness of a will that desired nothing but the desire of God — and that a thing whose whole worth is in the willing cannot be touched by what becomes of it in the world. The defeat does not reach the act, because the act was never the kind of thing a defeat could reach.

I said at the start that I would not give you a theory, and I haven’t. I have given a picture, and I think it is a better picture than the ones on offer, and I think, for what it is worth, that it is true. What is mourned, on this picture, is not a man overtaken by a fate he could not escape, and not a man who spent on his death a freedom he might have spent otherwise. It is a man so entirely himself that his self and his death and his God had become a single act; who could do no other, not because the road was barred, but because he had loved his way past the point at which the other road still existed; and who called the end of that road felicity, and, I am inclined to think, meant it.

---

[^1]: A note on the footnotes. Several of them were afterthoughts, and a few carry as much of the point as the body does, which I am aware is a vice, but one I seem unable to give up.

[^2]: The traditions of foreknowledge are many, and they come through both the Sunnī and the Shīʿī collections — the vial of earth entrusted to the Prophet and to be turned to blood on the day of the killing (the so-called ḥadīth al-qārūra); the recognition of the ground on arrival at Karbalāʾ on the second of Muḥarram. The chains are contested and some are weak, and I am not going to enter into the textual questions here; the point I need is only that the tradition takes the foreknowledge to be overdetermined, established many times over rather than resting on any one report. For the body of material see al-Majlisī, Biḥār al-anwār, vols. xliv–xlv, and, on the Sunnī side, the relevant notices in Aḥmad’s Musnad and al-Ḥākim’s Mustadrak.

[^3]: I labor the distinction between giving a picture and giving a theory because I think a great deal of trouble in this area comes of demanding theories where pictures are what is wanted, and then, not getting a theory, concluding that nothing has been said. A picture can be correct or incorrect. It can be better or worse than a rival picture. These are not small things to be able to say.

[^4]: I have argued at length elsewhere that “a priori” and “necessary” are not synonyms, are not even coextensive, and that an argument, not a definition, is needed to connect them; that there are necessary truths known only a posteriori, and probably contingent truths knowable a priori. The present point is the cousin of that one. “Known in advance” belongs to epistemology; “could not have been otherwise” belongs to metaphysics; and the inference from the first to the second is invalid, however natural. Determinists about history make a version of the same slide when they argue that since a great man’s deeds were, in some sense, bound to be done by someone, the man himself was dispensable; the slide is not refuted by, but it is also not licensed by, any correct account of the matter.

[^5]: Harry G. Frankfurt, “Alternate Possibilities and Moral Responsibility,” Journal of Philosophy 66 (1969). The principle he attacks — that one is responsible for what one has done only if one could have done otherwise — he calls the principle of alternate possibilities, and the counterexample with the idle intervener is his. I am going to put the counterexample to a use he did not, I think, have in mind, which is to model a theological notion; I hope he would not object.

[^6]: The maneuver of replacing the human intervener with God, precisely so as to secure certain advance knowledge of what a free agent will do, is made in the literature on foreknowledge and the principle of alternate possibilities — see, among others, Vihvelin and Hunt, and the use Fischer makes of the device against the indeterministic horn. I find it striking that the free-will debate should have arrived, on its own and for its own reasons, at the structure I want for ʿiṣma. The application to infallibility specifically I have not seen made, and I offer it for what it is worth.

[^7]: The “flicker of freedom” is Fischer’s phrase for the residual alternative the defender of the old principle points to; he has himself tried, in several places, to put the flicker out. The sharper objection I mention is the so-called dilemma defense (Widerker, Kane, Ginet). I do not pretend the dispute is settled, and an argument that leaned on its being settled would be in trouble.

[^8]: That kasb renames the difficulty rather than removing it — that it stands to determinism roughly as a synonym stands to its word — is, I gather, close to the consensus, and was felt by critics within the kalām tradition itself. The doctrine is al-Ashʿarī’s. I cite it not to score against a school but because it is the clearest available illustration of a maneuver one is tempted to make in the case of al-Ḥusayn as well, and ought, I think, to resist.

[^9]: The words ascribed to him — that he sees in death nothing but felicity (saʿāda), and in life among the oppressors nothing but weariness — are reported in the martyrdom and epistolary material; one finds the cognate sentiment in the testament to his brother on leaving Mecca, where he frames the rising as the seeking of reform and the enjoining of right. I am not resting anything on the precise wording of a particular report; I am resting on the unmistakable register of the whole, which is the register of a thing wanted, not a thing borne.

[^10]: Frankfurt develops volitional necessity in “The Importance of What We Care About” and “Rationality and the Unthinkable” (both in the volume of that name, 1988) and pursues it in Necessity, Volition, and Love (1999). Luther’s Hier stehe ich, ich kann nicht anders is his example, and the crucial observation about it is that the man retains the capacity to do the thing he says he cannot do, so that the “cannot” reports the constitution of his will and not a constraint laid on it from outside.

[^11]: The remark about reversed intuitions I borrow from myself; I once said it of those who find the notion of an accidental property unintuitive, and I think it applies here with equal force to those who find the freedom of the bound will a contradiction in terms. I should add, in the spirit of full disclosure, that I regard a thing’s having intuitive content as very heavy evidence in its favor — about as heavy, ultimately, as evidence gets — and I am unmoved by the fashion for treating it as nearly worthless.

[^12]: That ʿiṣma is a luṭf leaving the power to sin intact, and that it must leave it intact on pain of collapsing into compulsion, is the dominant Imāmī position; al-Mufīd is explicit that the Imams “are still free to choose between good and evil and are not compelled to the good,” and the later systematizers, Naṣīr al-Dīn al-Ṭūsī and al-Ḥillī after him, say in so many words that the infallibility does not exclude the capacity to sin. The contrasting Ashʿarī construal, on which God simply does not create the sin in the agent, is of a piece with that school’s determinism, and is, for my purposes, the picture to be avoided.

[^13]: The commentator is al-Majlisī, characterizing the inner state of ʿiṣma: a condition in which no desire is left save the desire of God, reached through an excess of the love of God that makes the contemplation of sin a kind of shame, the capacity for disobedience notwithstanding. I lean on this because it supplies exactly what the merely structural account leaves out, namely that the necessity in question is a necessity of love.

[^14]: This is the point I have pressed about possible worlds generally: that they are stipulated and not discovered, that we do not identify an individual across them by inspecting his qualities through a glass but by stipulating, of a counterfactual situation, that it concerns him, after which it is a substantive question whether the stipulation can stand. Applied here together with a suitable essentialism — the doctrine that certain properties belong to a thing in every situation in which it figures at all, as I have argued origin does — it yields the conclusion that a situation in which “al-Ḥusayn” submits may not be a situation about al-Ḥusayn. I put this forward as a strengthening of the main line, not as something the main line depends on.

[^15]: Nietzsche, The Gay Science §276, and Ecce Homo (“my formula for greatness in a human being is amor fati: that one wants nothing to be other than it is … not merely to bear the necessary, still less to conceal it, but to love it”). The essential thing, for my contrast, is that the love is directed at a necessity behind which there is, by hypothesis, nothing.

[^16]: Epictetus, Encheiridion 8 (“do not seek to have things happen as you wish, but wish them to happen as they do, and your life will go well”); the image of the dog tied to the cart, who comes along willingly or is dragged but arrives regardless, is reported of the early Stoa. The Stoic and the Nietzschean differ in temper but agree in locating the worth of the willing in the willer’s relation to a necessity that is simply given, which is just where al-Ḥusayn’s case, as I read it, differs from both.

[^17]: Thomas Nagel, “Moral Luck,” in Mortal Questions (1979); Bernard Williams, “Moral Luck,” in the volume of that name (1981). The variety at issue is what Nagel calls resultant luck, luck in the way one’s actions turn out. The tradition’s insistence that the defeat does not diminish the act is, in this vocabulary, a refusal to let resultant luck bear on the worth of the deed — a refusal I have tried to ground in the location of that worth wholly in the willing.