How does Jesus save?

When I was born 60 years ago, debates about how Christ saves us tended to divide Protestants who thought about such matters at all into conservatives who defended some form of substitutionary atonement theory and liberals who were more apt to accept a kind of moral influence theory. Both those approaches were about 900 years old. In the years since, new accounts of Christ’s salvific work have been introduced or reintroduced, and the debates have generally grown angrier, at least from the liberal side. Those who defended substitutionary atonement were always ready to dismiss their opponents as heretics; now some of their opponents complain that a focus on substitutionary atonement leads to violence against women and to child abuse.

Christians seem more divided on these matters than ever. Looking back to the history of theology to find yet another approach certainly will not magically solve the problem, but a reconsideration of the physical or mystical theory of how Christ saves us might contribute in a small way to more fruitful and civil conversation.

Shortly after 1100, Anselm (Italian by birth, French by education, appointed as archbishop of Canterbury) wrote a classic version of substitutionary atonement. He used the imagery of the feudal system of his time, in which vassals owed debts of honor to their lords. We humans, Anselm said, owe everything to our Lord and Creator. When we sin, disobey and thereby dishonor God, we fail to render what is properly God’s due. As the very foundation of justice, God cannot simply overlook our shortcomings. Moreover, since we owe God everything by way of obedience, once we have fallen behind in our account we have no way of ever catching up.

Nevertheless, such is God’s love that God will not simply abandon us (at least not all of us) to the consequences of our sins: “This debt was so great that, while none but man must solve the debt, none but God was able to do it; so that he who does it must be both God and man.” Christ’s suffering pays off what human beings owe to God’s honor, and we are thereby reconciled to God.

That we owe God more honor than we can give provides one image for what gets substituted in Christ’s death. Karl Barth notes a range of alternative themes: forensic (we are guilty of a crime, and Christ takes the punishment), financial (we are indebted to God, and Christ pays our debt) and cultic (Christ makes a sacrifice on our behalf). For various cultural reasons, I suspect, the oldest themes (honor and sacrifice) prove to have more depth than the more modern ones (payment of a debt, punishment for a crime). But in all these alternatives, the understanding of atonement has the same structure. Human beings owe something to God that we cannot pay. Christ pays it on our behalf. Thus God remains both perfectly just (insisting on a penalty) and perfectly loving (paying the penalty himself). A great many Christians would define such a substitutionary view of the atonement as simply part of what orthodox Christians believe.

Within a century of Anselm’s time, however, Peter Abelard was raising a moral objection to the idea of substitutionary atonement. “How cruel and wicked it seems,” he wrote, “that anyone should demand the blood of an innocent person as the price for anything . . . still less that God should consider the death of his son so agreeable that by it he should be reconciled to the whole world!” If an innocent Christ suffers, why should a loving God feel good about that or count it to our credit?

In our time Rita Nakashima Brock has denounced as “divine child abuse” the notion that God the Father welcomes the death of God the Son as payment. A womanist theologian once famously spoke of the “blood and weird stuff” involved in such an account. British theologian Timothy Gorringe argues that those who believe in substitutionary atonement tend to support nastier prison systems.

Such contemporary concerns make the issues more vivid, but many of the underlying issues have not changed since Abelard voiced his objections. We have become more aware of how often women are asked by their pastors to put up with abuse from their husbands and how often Jesus’ willingness to suffer abuse on the cross is offered as a model of suffering to such victims. There does seem to be something weird about how the torture of an innocent person could make things more just rather than less so.

The Presbyterian theologian Shirley Guthrie used to tell about a child who, having heard substitutionary atonement explained, burst out, “I love Jesus. But I hate God.” Not an altogether unreasonable response, it might seem, to an account in which Christ’s grace saves us from God’s unblinking justice. John Calvin, often identified with substitutionary atonement, protested that God had loved us since before creation and did not need to be led toward loving us by the events of Holy Week. But too often that is how the story makes it sound.

Abelard did not only offer criticism; he had an alternative account to propose. Christ’s love, he said, so inspires us by its unique example that “in teaching us by word and example even unto death, he has more fully bound us to himself by love; with the result that our hearts should be enkindled by such a gift of divine grace, and true charity should not now shrink from enduring anything for him.” This is the moral influence theory: in healing, teaching, suffering and dying, Christ so shows God’s love for us as to inspire us to love God and neighbor the way we should have all along.

Such optimism about human change, alas, makes the moral influence theory open to something like empirical refutation. Anselm claims that Christ’s suffering works a change in God. This is hard to disprove! But according to Abelard’s moral influence theory, Christ’s suffering is supposed to work a change—dramatically—in us. We should then be able to look at ourselves and judge.

I do not find the results of such introspection encouraging. As I understand it, Abelard and those who agree with him need to argue that Christ’s suffering has inspired Christians to shift from failure to success in our efforts to love God and neighbor as we ought. I do not see us getting dramatically better.

Nor is Abelard entirely free of the problem of valorizing suffering, for which Anselm so often gets blamed. In Abelard’s theory too, after all, the most perfect being shows his perfection by suffering. The audience benefiting from that suffering is different—it now inspires us rather than satisfying God—but, from the standpoint of many critics, that may make matters worse.

In a little book called Christus Victor, first published in 1931, the Swedish theologian Gustaf Aulén offered a third explanation of how Christ saves us—not a new theory but the recovery of a very old one. Aulén argued that for first thousand years of Christian history, until the time of Anselm, the dominant view was that by sinning we humans have fallen under the control of the devil. Christ frees us by defeating Satan in battle, thereby securing our freedom. Aulén found this theme especially prominent in the Greek fathers and then again in Luther.

The Christus Victor pattern seems more objective than the moral influence theory—it is not just that we are inspired to do something but that God has already done something on our behalf. And the risk of God the Father becoming the villain of the piece—a risk always on the edge of the substitutionary atonement accounts—disappears. Satan is the bad guy here, holding us entrapped until Christ accomplishes our liberation.

In the one image every student seems to remember from Christus Victor, Aulén contends that God does not use violence but rather cunning. Summarizing the thinking of fourth-century theologian Gregory of Nyssa, Aulén pictures how Satan saw Christ as human and consumed him, thereby also swallowing divinity and taking more than his due entitlement, so that all bets were off. “As a fish swallows the bait on the fish-hook, so the devil swallows his prey, and is thereby taken captive by the Godhead, hidden under the human nature.”

Anselm, who knew about the Christus Victor model, had already doubted whether, if we are not to imagine God as violent, we should imagine God as deceitful trickster either. He also wondered whether Satan could exercise legitimate control over creatures like us, made in God’s image. The Christus Victor account gives us a wonderfully dramatic Star Wars kind of narrative, but Aulén himself, while defending it, worried that “its mythological dress, its naive simplicity, its grotesque realism” may “awaken disgust.” At the very least, its interpreters need to specify whether its battle imagery refers to a metaphorical, spiritual battle (in which case Christus Victor theory turns into something like moral influence theory) or an honest-to-goodness swordfight battle (in which case the questions about mythological imagery become very pressing indeed). Parents who have tried to explain the religious significance of the battle scenes in the film versions of C. S. Lewis’s Narnia books may have faced the same puzzles.

The past 25 years have seen the emergence of a fourth account of how Christ saves us. This approach, based on the work of the French literary critic and philosopher René Girard, has spread widely and by now is probably more influential than Aulén’s. Born in France in 1923, Girard came to the U.S. to study literature, received his Ph.D. from Indiana University and spent most of his career at Stanford. His vision for explaining all sorts of things within the framework of a few key insights rivals that of Marx and Freud in sheer scope. In titling one of his books Things Hidden Since the Foundation of the World, Girard wasn’t exaggerating the scope of his theory. Other key works of his include The Scapegoat, Violence and the Sacred and I See Satan Fall Like Lightning.

Human beings, Girard explains, are inevitably competitive in what we desire. I may initially just vaguely want a place to come into out of the rain, something to eat, someone to hold. But then I want that particular cave, meal or sexual partner because you wanted it first—it was your choice that made it really valuable to me. In the stories and rituals studied by anthropologists, in the texts of literature and in everyday experience, Girard keeps seeing mimetic violence: our imitation (mimesis) of others’ desires leads to violence when we try to take what we want only because they wanted it first.

We come together in communities, Girard says, because we want to imitate each other, but our imitation leads to violence when we want to steal from and kill each other. Communities’ reasons for forming thus contain the seeds of their own destruction. Societies find a solution to this conundrum, Girard explains, in the practice of scapegoating. Faced with the prospect of a war of all against all, humans reformulate the conflict into a war of nearly all against a few—against the Jews or the communists or the gays or the feminists or the Mexican immigrants. Some relatively small group (maybe even an individual) is made the source of all our problems. If we could just get rid of them, the problems would go away. (If this seems an exaggeration of scapegoating, watch Lou Dobbs some night on CNN.)

Picking on scapegoats reduces the problem to manageable proportions. Everyone does not have to hate everyone else; we can all unite in hating the scapegoats. From the inside, this activity doesn’t look like picking on scapegoats; those involved in the practice think that they are simply understanding the world as it is. So, to use Girard’s most vivid literary example, the members of the chorus in Oedipus Rex do not think of themselves as picking on a clubfooted outsider (Oedipus) as the source of their problems. They truly think that his evil deeds have caused the plague that afflicts their city. Girard traces this pattern of scapegoating from primitive myths to literary classics to examples from popular culture.

The artifacts of one particular culture, however, radically break with this pattern. The Bible keeps telling us that scapegoats are innocent: Cain was wrong to kill Abel; Joseph’s brothers should not have sold Joseph into slavery; Christ was innocent. Whereas every other culture tells stories about how some small group of outsiders nearly did us in, the Bible over and over again shows us that the victims are innocent. When enough people understand that pattern, the world will change: “The cultural order cannot survive such a revelation. Once the basic mechanism is revealed, the scapegoat mechanism, that expulsion of violence by violence, is rendered useless by the revelation. . . . The kingdom of God is at hand” (The Scape goat).

Girard invites a powerful hope. He began his research as an agnostic and converted to Catholicism part way through. This provides him, fairly or not, with more credibility than those who are Christian from birth. He captures something right about that fellowship with outsiders which lay so near the center of Jesus’ ministry as recounted in the Gospels. And in a culture like ours in which the loudest Christian voices sometimes seem identified with militarism and even the justification of torture, he unambiguously separates Christian faith from violence.

Little wonder that many Christian theologians have recently been publishing books inspired by Girard: from the brilliant British Roman Catholic James Alison (Raising Abel) to the Mennonite J. Denny Weaver (The Nonviolent Atonement); from Roman Catholic author and speaker Gil Bailie (Violence Unveiled) to Andover Newton theology professor S. Mark Heim (Saved from Sacrifice). The current rate of publication runs to at least several books a year.

I wish I thought the Girardian approach works. But a theological solution needs to deal with the problem that we really have. Girard’s solution lies mostly in a realization: we realize that scapegoats are innocent, and once we have realized that, we cannot keep scapegoating them. The ability to stand up, yell “Let’s go beat up all the Xs” and get any kind of following requires both speakers and hearers who can, at least at some level, convince themselves that those Xs really are bad people whose deeds are the source of the problems afflicting the rest of us. Understanding the Bible, according to Girard, makes it impossible to keep convincing oneself of that, and hence impossible to scapegoat. And that makes the world a much better place.

Girard argues persuasively that many people who read the Bible do not understand it properly, and he also makes the case that the sheer desperation of those who want to keep scapegoating will lead them to turn to ever more extreme measures as their case gets stretched thinner. Still, the problem with Girard’s theory is that after 2,000 years of Christian teaching, we are at the end of a period that saw the Holocaust, the Chinese Cultural Revolution, Christian missionaries’ participation in the Rwanda genocide, and the ongoing division among most Christian communities over how to treat homosexuals (I am not claiming that these issues are morally equivalent). If Girard’s theory is right, we should have seen at least a little diminishment in the practice of scapegoating by now.

My problem—and I suspect it is also the problem of many others—is the one described by Paul so long ago: “I can will what is right, but I cannot do it. For I do not do the good I want, but the evil I do not want is what I do” (Rom. 7:18b-19). Helping me to realize my faults is therefore in itself no cure. I understand that when despised outsiders are over there, and people over here are speaking of them hurtfully and with contempt, then I ought to move from here to there. But it is much more comfortable over here, and the people here are often better looking and rather consistently more successful. Many days, I would rather stay put.

What do I need? I need a good cattle prod to get me moving. I need forgiveness for my past failures. And I need hope that I can continue to be forgiven. Just realizing a past pattern of mistakes is not enough. If some of you are just better human beings than I am and do not need these things, then all I can say is, God bless you! But I think I know what I need if I am to be saved. And in Girard’s account of Jesus, Jesus does not quite have it.

What follows does not involve pulling some rabbit out of a hat to solve the problem of soteriology. If I am correct in thinking 1) that adherents of substitutionary atonement and moral influence theories of atonement are increasingly at odds, and 2) that serious problems have emerged with both Christus Victor and Girardian models, then it is worth considering what new direction our soteriological reflections might take, even if one can only point very tentatively toward a different path.

Two general theological principles guide my thoughts here. The first comes from my teacher Hans W. Frei: in theology, doctrines should illuminate and clarify stories rather than stories illustrating doctrines. That is, what is most basic to Christian faith is not, for example, any soteriological theory or even doctrines about the truine God or the two-natured person of Christ but rather the stories of God’s covenant work with Israel and then the birth, life, death and resurrection of Jesus Christ. We figure out and affirm the doctrines necessary to make sense of those stories (holding them tight if they really are necessary), but the stories come first.

My second principle comes from Calvin: “Christ saves us by the whole course of his obedience.” I think this principle needs defense these days on two fronts. First, too much theology has focused exclusively on the cross or (at most) on birth, cross and resurrection. Calvin’s own Geneva Catechism jumps rather remarkably from Christ’s birth to his death “because nothing is said [in intervening matters] about what properly belongs to the substance of our redemption.” In this it only follows the pattern of the classical creeds. But any account of what Christ has done for us that leaves out his teaching, his preaching, his healing and the rest his ministry is not faithful to the witness of the New Testament and does not meet the needs of contemporary Christians.

On the other hand, Christ’s obedience led even to death. A really-nice-guy Jesus, eloquent in speech, uncontroversial in message, purpose-driven in ministry but somehow tragically misunderstood by those in power and therefore executed in a terrible mistake, is not the Jesus in whom I believe. In scripture and in the Jesus who speaks most to my condition, I find one who spoke truth to power, who at least suspected from early on that his road would lead to his death, and who understood that somehow (I would leave many paths open beyond that “somehow”) his death would be not the failure of his life but the achievement of his goal.

In short, both Jesus’ life and his death matter to our salvation, and we should not rest content with one without the other.

That said, we can take a step behind Aulén’s project, for even earlier than the Christus Victor motif came what J. N. D. Kelly called the physical, or mystical, theory of redemption, already remarkably developed in the work of the church’s first great postbiblical theologian, Irenaeus of Lyons. Drawing on Paul’s account of Christ as the second Adam and Luke’s genealogy, with its implication that Christ “recapitulated in himself all the dispersed people dating back to Adam,” Irenaeus concluded that “because of his measureless love” Christ “became what we are in order to enable us to become what he is.” This last phrase may be the single most repeated one in early Christian thought. “There is one grand theme which . . . provides the clue to the fathers’ understanding of the work of Christ,” Kelly concluded after a lifetime of research. “This is none other than the ancient idea of recapitulation which Irenaeus derived from St. Paul, and which envisages Christ as the representative of the entire race.”

Similar ideas appear in the writings of nearly all the Greek fathers. Indeed, in all the debates about Christ’s full humanity—does Christ have a human soul? a human will? human emotions?—the clinching argument is always that because Christ saves what is human by uniting it with divinity, any part of Christ which does not fully share in humanity will not be saved. In Gregory of Nazianzus’s famous phrase, “That which he has not assumed He has not healed.”

To say that Christ saves our humanity by uniting it with divinity is not to say that his saving work is accomplished at the moment of the incarnation. At that point, after all, the boundaries of that work remained far from clear. Were gentiles also saved? What about the despised and the condemned? The list of such questions could be long extended. Only when God incarnate has welcomed sinners into his table fellowship, cured those who suffered, died the death assigned the blaspheming and seditious, even gone into the realm of those who have rejected God and exist in a hell of utter isolation (I pick up at the end a theme most eloquently presented in our time by Hans Urs von Balthasar)—only when this God incarnate has been raised can we glimpse the expansiveness of God’s work of salvation. It is only the crucified One who can save us all.

The physical, or mystical, account of the atonement can indeed make room for elements of other pictures. The Christ who becomes what we are so that we might become what he is can also teach us about God, manifest self-sacrificial love, defeat on our behalf forces of evil and show us the innocence of many whom society condemns. But the process begins when he conjoins our sinful humanity with divinity.

One of the problems involved in this theory of the atonement is that accepting it supposedly involves buying into Platonic philosophy, and Christian theology should always be reluctant to tie our faith to one philosophical position. Plato famously taught that there are particular objects and universal forms or ideas. Particulars get defined by their participation in the forms. Thus, for instance, there is an eternal, unchanging idea of justice, and there are lots of particular laws—but a law is just only to the extent that it participates in the idea of justice. In the debates of the Middle Ages, “realists” argued that universal forms have a real existence—eternal justice is really out there somewhere—whereas “nominalists” believed that justice is only a name we humans use to classify laws we want to put in the same category.

For Christ to unite our humanity with divinity, the argument goes, there has to be a humanity that we all share. But that means that Platonism in its realist form has to be true if we are to be saved, and again, the history of theology suggests the dangers of tying our faith to any particular philosophy.

But I do not think this argument follows, either historically or conceptually. In the fourth century Athanasius talked about how a whole town can be transformed if a famous person becomes its citizen. Similarly, he said, how different it is to be a human being if the Creator of the whole universe has also become one like us in all things except sin. The American theologian Jonathan Edwards was, at least in some moods, a radical nominalist who thought that there was no single right way to divide up the universe—God could do it however God wanted. If for the purposes of our salvation God wanted to think of all human beings as a single unit, Edwards saw no reason why not. But neither Athanasius nor Edwards believed in the philosophical view that universals have real existence (or, if Athanasius did, he was not using it to make this argument). There are many ways, in short, to think of humanity as sufficiently one whole that Christ could transform it.

Eastern Orthodox theologians have often taught that human beings are deified in the process of salvation. Western theologians, whether Catholic or Protestant, are often nervous about such language and criticize a soteriology that claims that “he became like us so that we might become like him” because it implies a process of deification. I am not sure that deification is such a bad theological concept, but in any case I do not believe that a physical theory of salvation requires it. There may be good reason, for instance, to hold that Christ was divine by nature while the rest of us are united to God only by adoption. Adoption rather than deification would therefore be the standard soteriological word. The core idea could remain. We join God’s family, we become citizens of God’s city, our humanity becomes one with God’s divinity—these could all be, as the Presbyterian Confession of 1967 said in a different connection, “expressions of a truth which remains beyond the reach of all theory in the depths of God’s love.”

The author of Hebrews put it like this: “For the one who sanctifies and those who are sanctified all have one Father. . . . Therefore he had to become like his brothers and sisters in every respect. . . . Because he himself was tested by what he suffered, he is able to help those who are being tested.” As Paul wrote to the Romans, “Just as one man’s trespass led to condemnation for all, so one man’s act of righteousness leads to justification and life for all.” Christ became like us so that we might become like him.

“By golly, he has done it; he has solved the problem of soteriology” would be a ridiculously inappropriate response to this essay. So would any instinct to throw away all discussions of other soteriological theories. My proposal is modest: How we understand the atonement seems an increasingly contentious topic within Christian theology. Among theories old, new and revived that have received a good bit of discussion lately, I think the physical or mystical account has been unfairly neglected, and I would hope that organizers of conferences and panels as well as theologians smarter than I am might devote more attention to it.