Some questions won’t go away. It is an article of the Nicene Creed that the Lord Jesus Christ was “crucified for us under Pontius Pilate.” What do the two little words for us mean? What good is the cross?

To ask that is to ask what is in technical parlance a soteriological question. But English-language theology has long used a good old English word as the comprehensive name for what the question is about: atonement. How atonement, or the atonement, can best be understood is thus a standard and convenient way to state the basic question in regard to Christ’s suffering and death as beneficial.

It is a commonplace that no “orthodox” answer has ever been formally defined. Nor is there consensus. Gustaf Aulén famously distinguished what he called “three main types of the idea of the atonement,” and others have offered similar typologies. But however they are categorized, explanations of the cross are not only different but disparate. That is why there is a question. Of the available positions, which one(s) should be taken seriously, taught, believed and preached?

The word atonement itself is no help. In a way it is part of the problem. On the one hand, we were all taught that it wears its meaning on its face: atonement is at-one-ment, reconciliation, in half as many syllables. That does seem to have been what the word was invented to mean. On the other hand, however, the verb atone, which came later, has veered off in another direction. No one ever translates 2 Corinthians 5:19 as “God was in Christ, atoning the world to himself.” To atone is not to reconcile, either in everyday use or in theology. I atone for something, some failure of mine, some offense on my part; and my atoning consists in acting, or more especially suffering, so as to compensate for the wrongdoing. Consequently what is meant by atonement may be either of two things. It can mean being or coming to be at one—the original, etymological sense. It can mean leveling the score, redressing the balance, making reparation or restitution or the like—probably the more usual sense. The two meanings are not unrelated, and the distinction between them is often blurred, but to insist on it is by no means splitting hairs. For one way to sharpen the question at hand would be to ask: Does atonement depend on atonement? Otherwise stated, does reconciliation with God depend on compensating, making amends, paying a price? Is that what the cross is all about?

Aulén, for one, thought not. He maintained that atonement (that is, reconciliation) can best be conceived as the triumphant outcome of confrontation and conflict, with Christ as the conquering hero. Hence the title of his book, Christus Victor. Enormously influential though the book has been, however, the stirring imagery of the Christus Victor motif has yet to reclaim the primacy it lost to the scholastic tradition inaugurated by Anselm, which still predominates in Western theology. It has been from this “Latin type” that the notion of atonement takes the meaning it commonly has in relation to the cross. The default setting, as it were, continues to be that Christ’s suffering atoned, compensated, made amends for human malfeasance—somehow.

The next question is how. There is a default setting for that too, although it has never gone unchallenged. In its most clear-cut form it usually goes by the name of substitutionary penal atonement. The straightforward logic of this account runs as follows. God is just. Justice, divine or human, requires that wrongdoers, whose wrongdoing makes them liable to punishment, should be punished. Humans, one and all, are sinners. As such they incur a penalty, which in justice ought to be paid and which has, in fact, been paid—not, however, by those who owe it and deserve to pay it, but by Jesus. The verdict never changed. Sinners are guilty. But because he died, the sentence has been suspended for everyone else. Instead of punishing, God pardons. That is the good news.

There is much to be said for this traditional explanation. For one thing, it supplies a comprehensible link between Christ’s suffering and a beneficial result, forgiveness, and so also between the gospel as a chronicle of past events and the gospel as kerygma here and now. For another thing, it gives God all the credit. My reprieve is none of my own doing; it is altogether an amazing, gracious gift. And for yet another and perhaps the most important thing, substitutionary penal atonement is not just conceivable but imaginable. It can give me what Newman would call a “real apprehension” of a tremendous boon for which I may be correspondingly grateful. The greater my conscientious dread of well-deserved punishment and the more vivid my experiential awareness of myself as a sinner, the greater the blessing of being assured that despite my guilt I shall not be given my just deserts.

That is why penal substitution preaches well: it speaks to the condition of the twice-born, to the sin-sick soul; and its speech is framed not in cool theological abstractions, which can give only “notional apprehension,” but in vivid, moving, affect-laden narrative images. Penal substitution may be a theory of atonement—an intelligible explanation, that is—but its appeal is not in the first instance intellectual. It is emotional, imaginative, existential. That being so, I can, as Newman says, “believe as though I saw.” I can picture Christ taking my place, enduring the pains I ought to have felt, and my imagination of how deeply he suffered brings home to me how great the penalty is that I have been spared. It is sometimes said that theology and spirituality have parted company. Not here.

Is the undeniable emotive power of this account enough to guarantee its truth? Newman held that there is no genuine belief without real apprehension. Even so, it does not follow that the criterion of credibility for a claim is whether it packs a visceral punch. I may be inclined to accept a statement because it engages my deepest desires and fears and yet discover that upon examination its implications are intellectually bogus or morally repulsive. The standard account of atonement, for all its affective effectiveness, might be like that—convincing, but only until you start to think about it.

There are, of course, serious objections to atonement conceived in terms of substituted punishment. All of them have been raised again and again, but it is worth rehearsing the main one. Recall the beginning of the argument summarized above: God is just. That sets the context for everything else, and the sequel makes it clear that by justice is meant, more specifically, retributive justice, which consists in attaching rewards to merit and penalties to fault. Now justice, so defined, is an attribute of the God described all through the Bible. There can be no objection on that score. The problem, rather, is that penal substitution cannot be squeezed inside the same definition. To punish the guilty is just. They deserve it. The innocent do not. To punish them is not just; it is just outrageous. But Christ was innocent, tempted in every way as we are, yet without sin (Heb. 4:15). Nobody would deny that Pilate, Caiaphas and the rest acted unjustly; but if by doing what they did they were executing a divine plan—if God intended to punish his Son by their hands—then evidently God is not just after all.

From this internal contradiction there are two escape routes, one incredible, the other reprehensible. The first introduces the remarkable claim that Jesus was guilty, but only because the guilt of others was transferred from them to him. This expedient so undermines the very idea of moral responsibility that it would be better not to speak of justice at all. Guilt in the relevant sense is not the sort of thing that can be siphoned out of one person and into another. Nor is it any better to argue that punishing the innocent, though admittedly wrong as a rule, can in exceptional cases be just, provided it serves to “send a message” that dramatizes the heinousness of disobedience in order to deter those who might be inclined to disobey. There is a name for that: terrorism.

The point of these well-worn objections is that atonement, conceived in terms of penal substitution, cannot be conceived coherently. Much the same point has been made, in more robust fashion, by writers who declare that what Western tradition calls atonement is divine child abuse or the vengeful violence of a tyrannical God. Those are caricatures, which is not to say they are utterly mistaken. Retributive justice does leave something to be desired as an intelligible framework for making sense of the cross.

What it leaves out, above all, is a personal dimension. In the forensic context of strict retributive justice, rewards and punishments correspond to desert and nothing else. It does not matter who the deserver is. In the same context, being forgiven is not a positive good; it is only a double negative. Punishment, by definition, takes away from an offender something valuable—liberty, physical well-being, companionship, possessions. Forgiveness would mean the remission or cancellation or cessation of (deserved) punishment. It comes down to taking away the taking away.

But a person is more than a party at law; and among persons forgiving is not reducible to omitting retribution. Forgiveness involves a change in both the forgiver and the forgiven—in their attitudes, their motivations, their selves. Enemies they were; friends they become, or become again. Hostile interaction gives way to concord. Such a reconciling shift in personal relations does not always happen, and when it does it is difficult, painful and costly—but not because suffering is an extrinsic preliminary condition that has to be met before forgiveness can occur, but because willingness to suffer is intrinsic to what forgiveness, in the personal sense, is.

Why so? Because, in the first place, evil is like the good it undoes in that it is infectious. It propagates itself. Suppose, then, that I have injured you. As a person, you are free to choose your response. If you choose to retaliate, you perpetuate the evil by causing a new injury. The choice may be wholly justifiable, but it is no less injurious for that. If instead you choose to hold a grudge, to brood on your injury and cultivate your dudgeon, you will still perpetuate the evil, internally, by diminishing yourself, souring your character and becoming your own victim as well as mine. On the other hand, if you choose to forgive, you are choosing to absorb the infection, as it were; to contain its self-diffusion, to forgo the gratifications of revenge, resentment, self-vindication and righteous indignation. Furthermore, you are choosing to make your willingness known to me, to offer me your friendship, to accord me a status and value no less than yours, all without denying my offense or ceasing to be my victim. At the same time, conversely, until I have chosen to acknowledge you as such, to own the injury, ask for your benevolence and reciprocate your offer, the forgiveness that we must both choose if it is to occur has yet to be fully chosen.

On this very abbreviated analysis, forgiveness is a matter of honesty, humility, communication and exchange, none of which takes place automatically or effortlessly, even for saints, much less for sinners. To forgive is not to forget, the adage notwithstanding. It is to remember. In the poet-theologian Charles Williams’s words, to forgive is to know an offense as an occasion for joy, a felix culpa, a happy fault. Such an altered state of conscious engagement does not come within the scope of justice in any ordinary sense. Even-handed justice responds to evil only with evil and only to good with good. Forgiveness responds to evil with good by transforming it, by willingly accepting diminishment so as not to prolong it, and by using it as a means of introducing a new good or restoring one that was wrecked.

In that regard, to return to the initial question, forgiveness would seem to be an instance, perhaps the defining instance, of a more general, more inclusive pattern. Its reversal of roles is not only a theme that runs through much of what Jesus is reported to have taught. Also, and most important here, it is enacted in the way he is reported to have met the final surge of hostility to that teaching and to himself. The hostility was probably inevitable; in that sense it was “necessary that the Christ should suffer” (Luke 24:26). But the necessity was not absolute. Things could have gone otherwise, to judge by the Gospels. Jesus could have chosen to flee, to fight arrest, to summon 12 legions of angels. All these he chose to refuse.

By so doing he chose to bear the cross, and his choice gave the bearing of it a meaning it would not otherwise have. Among thousands of Roman executions, this one is meaningful—not in the way a quantum of suffering might be meaningful, weighed in the scales of retributive justice, but meaningful as a communication, a word, an expression of willingness consistent with what Jesus had until then been expressing in deed and speech.

Has all this got anything to do with atonement? No. Not in the sense that because Christ accepted his suffering we do not have to suffer. It is the other way around. He accepted it because we do have to. His was a cross that had always been ours, the one way open to us, in a skewed world, for putting a stop to the consequences of our own malice without adding to them. Accepting that way, the way of the cross, was an act of solidarity with us and an offer of solidarity with him—an appeal for us to follow him by willingly taking up whatever crosses the world imposes, by making them occasions for joy, by forgiving.

Any exposition of “crucified for us” along the lines drawn here is susceptible, so it seems, to criticism on the ground that it is an “exemplarist” or “moral influence” account, at best no different in principle from the one that got Abelard in trouble and at worst Pelagian. By this account, what Christ’s passion has done, the critic might object, is not a deed that we ourselves are incapable of doing; all it does is exemplify the principle that forgiveness is costly, that evil is to be met with peaceable resistance and that it is better to submit to wrong than to do it. It did not take the crucifixion of God to tell us that. Socrates said as much.

The charge of Pelagianism would stick if taking up our crosses instead of taking revenge were something we could do by ourselves, with enough effort. It is not. Nor is the life of self-donation portrayed in the Gospels an exemplar that is at all inviting or attractive to the self-regarding will to power with which it seems we are born. On the contrary. If it is true, as Christianity’s fiercest and most perceptive opponent maintained, that what is worth choosing above all else is heroic self-affirmation, then “God on the cross” can only be, as it was for Nietzsche, a ghastly negation of all that is best and noblest in humanity. Maybe that is hyperbole. Still, we are told it was Jesus’ own teaching that no one can come to him without being drawn by the Father who sent him (John 6:44), and the saying certainly suggests that it is not by unaided effort or instinct that anyone who comes to him does. The attraction, it would appear, is either unnatural or supernatural. So far, Nietzsche was right.

The point is worth a little elaboration. One of the peculiarities of Western Christianity has been a tendency to speak of God’s initiative in reconciling his human creatures as though it were entirely a matter of sending his Son into the world. But God’s Spirit too has been sent—and continues to be. On the well-founded assumption that this second divine initiative complements the incarnation, there is reason to suppose that part of the indwelling Spirit’s job description is to be the “drawing” that attracts self-sufficient persons to the self-emptying person of Christ. In other words, the motivation for choosing this exemplar is itself a gift, “the love of God poured into our hearts by the Holy Spirit given to us” (Rom. 5:5). It would follow that reconciliation—atonement, if you like—can be understood as the action of a tri-personal God, rather than a transaction between the Father and the Son.

There remains the question of divine justice. It is a Pauline preoccupation, and the theology of penal substitution is more than anything else a cumulative attempt to systematize the unsystematic profundity of the letter to Romans. There Paul is obviously struggling to maintain that divine justice is retributive (the “wrath of God”) yet somehow also, in Christ, creative or transformative. In the tradition that began with Anselm, the struggle is resolved on the side of retribution. Arguably, though, it would be more authentically Pauline to resolve it, as Derek Flood has begun to do in Healing the Gospel, on the side of what he calls “restorative” justice. The distinction, roughly stated, is that punitive justice is concerned with what may be done to evildoers and restorative justice with what can be done about evil. Taken in this latter sense, justice as a divine attribute has its supreme embodiment in Christ’s acceptance of the cross. So and not otherwise is good brought out of evil—not even by God.

Presumably God has always been able to purge the world of its evils with an apocalyptic blast of power. Instead he has chosen to conform to the same justice he requires of his human creatures, to submit to the conditions of at-one-ment with them, to become all they are and are to be. And that is good news.