(On St Anselm of Canterbury’s feast day I thought I would honor him by addressing some of the issues that have troubled people about various “theories” of substitutionary atonement. Anselm himself is often blamed for views he never held. This article is excerpted from an essay which was later included, in an edited form, in my book on atonement, When I Survey the Wondrous Cross: Reflections on the Atonement. Wipf and Stock, 2010}

Even if one accepts that the interpretation of Christ’s death as a substitutionary atonement is thoroughly biblical, there remain any number of problems and objections to understanding it this way, especially if one is putting forth a view that claims for the cross both objective divine activity and universal implications for human salvation. I turn to the following works which lay out some of the criticisms of substitutionary atonement theories:  Vernon White, Atonement and Incarnation: An Essay in Universalism and Particularity, (Cambridge: Cambridge University Press, 1991) Gerald O’Collins, The Calvary Christ (London:  SCM Press, 1977), especially Chapter VI, Atonement for All, and Richard Bauckham, “Criticisms of Penal Substitution”(unpublished paper, no date).

A theology that claims that universal reconciliation came about through an historically particular event  or person faces many difficulties from outside as well as from within the Christian faith.  The expanded view of the universe in the Twentieth Century makes the universal claims of the Christ event incredible to many people. Widespread travel and world-wide communications make religious pluralism a fact of modern life and put a human face on the previously abstract “unsaved heathen” (now our neighbors) of former centuries.  In addition, the conceptual difficulties of ideas like atonement and incarnation puts pressure from within theology to abandon its traditional claims. The tendency in much recent theology is to solve these problems and objections by giving ground on both objectivity and universality. These theologies counter the outside pressures facing atonement by proposing that atonement is merely illustrative of rather than constitutive for salvation, and that salvation is particular and limited rather than universal in scope.  White summarizes from Maurice Wiles The Remaking of Christian Doctrine to give a typical example of a theology that has  responded to these pressures:

 (Wiles) proposes that there need be no more to the meaning of the passion of Christ than the following: first, a revelation of the character of God; secondly, an historical phenomenon effective in the transformation of people’s lives.  Thus he has no wish to deny that it concerns only the comparatively small proportion of mankind which has heard of and responded to the preaching of the cross; to claim anything more would be “chasing a will o’the whisp”. (White, p 3)

White himself offers an atonement theory that attempts to preserve objectivity  on God’s part and the universal implications of God’s act by using the language of recreation  rather than retribution. White wants to answer the modern objection to a theology that claims universal implications from a particular event.  He phrases the question he wants to answer thus:  “What is it about any particular act that could constitute possibilities for the effectiveness of every other act the agent untertakes in relation to other agents throughout time and space?” He rightly notes that such a form safeguards both the universality of divine action and the uniqueness of the particular event.  (p 52)  As White’s title implies, it is an incarnational Christology that provides the locus for divine activity with universal human implications. To support his view he points to Paul’s soteriology, expressed in terms such as “one , in “one spirit”, “Christ’s” and “in Christ”, referring to E.P. Sanders recent work on participation in Pauline atonement theory. (p 58)

White draws attention also to C.F.D. Moule’s writings on corporate personality  and to the Patristic idea of recapitulation as articulated by Irenaeus, where Christ is understood to be the representative of the whole human race. (p  `59) White’s soteriology here hangs on his Christology: “It requires that the man Jesus and the eternal God share a common personal identity, as subject of the same incarnational experiences.” He sees that “the life, death and resurrection of Jesus Christ constitute a particular experience for God which is then offered throughout all time and space with the potential to “save” all peoples. (p 87) The cross then, as here construed,  is primarily an event in God’s “life” and only secondarily an act in history. That lack of historical grounding shows elsewhere in White’s essay, as for example, in the implications of his theory for personal faith and for the church, which seem strangely irrelevant. Since his position needs no human agency (such as the church) to witness to the Christ event, which seems to have accomplished all that is necessary  in eternity, he avoids any charge of religious imperialism in the face of the world religions. To his credit he holds fast to the two elements that one wants in an atonement  doctrine, objectivity and universalism, and tries to deal with the pressures facing atonement from the modern world.

But these more general pressures on theology from the corrosive effects of the “acids of modernity” are not the only  objections faced by an atonement doctrine. There are objections from within theology as well.  There are any number of ways to organize these.  Gerald O’Collins, for example, summarizes these objections in three categories:  (1) God misrepresented,  (2) Christ misrepresented, (3) mankind misrepresented.  Let us look at each in turn: (1)  O’Collins wants to distance himself from atonement views in which God’s character or nature is misrepresented. For example, some articulations of the atonement  can too easily  speak of God as a vindictive punisher. For O’Collins, Christ death on the cross is better understood as retributive suffering freely undertaken for others rather than punishment imposed by God as a substitute for guilty humanity. (2)  Similarly, O’Collins thinks Christ is misrepresented by substitution language which inclines toward the impersonal and lacks the intentionality of the language of “representation.”

Representation implies conscious acceptance by free moral agents on both sides. He also wants to carefully limit the way we talk of sacrifice, avoiding talk of an expiatory sacrifice that propitiates the anger of God and wins forgiveness for humankind. He admits that Paul uses sacrificial language (as in 1 Cor. 5:7;11:25; Romans 3:24f) but  claims that Paul doesn’t make much of it as either an expiatory sacrifice or as a sacrifice that brings a new covenant with God. The reason for this reluctance is because Paul see God as the initiator of the Christ event. It was God who “put forward” his Son to expiate human sins and usher in the “new covenant.” Once we see the crucifixion as an act of God toward mankind, we can hardly turn round and speak of God sacrificing to himself.  Likewise, any stress on Calvary’s consequences “for us” tends to exclude the theme if its consequences “for God” and hence its sacrificial quality.  Paul knows the cross to be an effect of God’s saving will,  not  its cause.  And that belief restrains the apostle’s readiness to proclaim Good Friday as an atoning sacrifice which establishes a new relationship between God and Man.. (O’Collins, p 108) Paul’s understanding of Christ’s role in the crucifixion  includes both obedience to the Father and suffering undertaken for humankind.  In two senses, then, he acted as our representative, becoming obedient unto death (Philippians 2:8) and dying for us (Galatians 3:13). (p 108-109) (3) Finally, O’Collins wants to avoid articulations of the atonement  in which mankind is misrepresented in regard to human involvement.  He asks, “Does belief in atonement (i) fail to produce a sense of commitment and (ii) suggest a world that smells of masochism?” (p 109)  These are in fact opposite problems. In the first, believers in atonement reverently refer to amends made in the past by Christ and become complacent about the world around them, and in the second case, a neurotic preoccupation with self-inflicted suffering is fostered.  This latter is similar to the criticisms of the cross offered by some feminists, that it fosters a victim mentality among those already prone to victimization.  While admitting that atonement theories can go wildly astray, O’Collins concludes that

. . . a healthy atonement theology invites well-founded action and acceptance.  Just as the reconciliation and liberation brought by Christ impels believers to act in genuinely reconciling and liberating ways, so the atonement he made on our behalf should alert us to our responsibility for the good state of the moral order.  (p 109)

While O’Collins’ caveat about the language of substitution is a good one in the light of the many impersonal and mechanical atonement theories that have been proposed, and his highlighting of the language of representation reminds us of important and often neglected dimensions of Christ’s solidarity and identification with us and his freely-chosen way of obedience, I am not persuaded that we can dispense with substitution language altogether. Christ’s death is a death for us and does for us what we cannot do for ourselves. That is substitution. Representation adds some important notes, but it cannot carry the whole melody.

Likewise, O’Collins’ warnings about expiation are helpful in protecting against a notion that Jesus was sacrificed to the Father in some way , yet again I think expiation language is biblical and retains a place in any doctrine of atonement as  long as we keep in view O’Collins’ reminder that God is the initiator and not the vindictive punisher whose anger gets assuaged by the act. I wonder too about O’Collins’ statement, “Paul knows the cross to be an effect of God’s saving will and not its cause.” On one hand, that is right, that the cross can be seen in retrospect to be entirely consistent with God’s nature and character as seen throughout the biblical narrative to that point. On the other hand, to speak of the cross as an “effect” of God’s saving will could be construed to support an illustrative soteriology rather than one in which the cross is understood as constitutive for salvation. Richard Bauckham’s paper outlines the four classical criticisms of the doctrine of penal substitutionary atonement as put forward by Socinus (Fausto Sozzini, 1539-1604), who was criticising the logic of the Reformers. The four criticisms are (1) Satisfaction and forgiveness are incompatible, (2)  Substitution is unjust, (3) Christ’s sufferings are not equivalent to the punishment that sinners deserve, (4) Substitution fosters antinomianism.

Bauckham notes that Socinus makes his criticisms apart from the context of the Reformer’s theologies and that he neglects two significant aspects of the Reformers views of the atonement, (a) that the work of Christ is not the activity of a third party, but rather the divine Son of God become man, who has come into the world to do the Father’s will for human salvation, and (b) that the purpose of the atonement was not merely to save sinners, but to reconcile them to God.  He then discusses each criticism in reverse order. (4)  Substitution ought to lead to antinomianism.

Bauckham notes that this charge ignores (b) above, that the atonement puts sinners into relationship with God, thereby not leaving them unchanged. Such criticisms stem from a moralistic misunderstanding of Christianity. We do not first need to be good, we need first to be in a right relationship with God. (3) Christ’s sufferings are not equivalent to the punishment required.  Bauckham concedes that this charge is unanswerable, that attempts to answer it have failed in the past, and that if anything like penal substitution is to be retained it must be shown that this quantifying of amounts of punishment is unnecessary. (2)  Substitution is unjust. Socinus says that substitution is doubly unjust in that the guilty party goes unpunished, and an innocent party is punished undeservedly. Socinus did not believe in the incarnation, but according to it, it is not that God spares sinners and inflicts their penalty on an innocent third party, but rather in Christ, God himself takes on the suffering instead of inflicting it on sinful humanity. This answers Socinus’ charge about punishing the innocent, but leaves the charge that the guilty go unpunished. Some sort of articulation of Christ’s identification or solidarity with the sinner goes part way to meeting this objection and most atonement theories have elements of this in them, but by the standards of human law courts to which Socinus refers, these motifs of identification cannot be strong enough to fully overcome his criticism.

Bauckham refers to  Wolfhart Pannenburg’s defence of this criticism where he charges Socinus with excessive ethical individualism. Bauckham goes on to illustrate this point by drawing an analogy from human courts of law. We can see in the criminal justice system, for example, how the punishment of an individual in some very real way punishes their family. In some cases this shared suffering, through active sympathy, might even be considered a kind of vicarious suffering of punishment that we could consider to not be unjust. This is outside the competence of the courts, of course, where “each person must bear the penalty he deserves” is the proper ideal for administering justice. But in real life the interdependence of humanity is such that the innocent do as a matter of fact suffer for the crimes of others. Though sometimes we see this as evil and seek to prevent it as far as possible, in other cases we applaud it. The cases in which we applaud it are those of voluntary fellow-suffering in love. We recognise at this point that love goes beyond the ethical individualism in which the courts must largely deal, and that the vicarious suffering of love may and can go beyond the ethical individualism of the courts because it corresponds more fully to the reality of human life and relationships than the justice of the courts is able to do. (p 8) To speak of Christ bearing our punishment is only possible because God in Christ goes beyond the justice of human law courts. Since Socinus’ criticism stays within the analogy of human courts, those theories of atonement that have accepted that framework are most vulnerable to his criticism. But if God’s justice in the cross transcends the justice of human law courts than the criticism is avoided.

Bauckham says that he has hinted at how this happens by his description of what he has called “the voluntary fellow-suffering of love.” “God’s love in Christ enables him to accomplish what, as a judge in a human law court, he could not accomplish.” (p 10)  To do this means going beyond the merely retributive understanding of justice which both Socinus and his opponents shared. (1)  Satisfaction and forgiveness are incompatible. Socinus uses the analogy of a debt, where if the debt is paid, the creditor is obliged to renounce any claim on the debtor. Neither mercy nor forgiveness come into play. Again, Socinus takes no account of the incarnation. Since it is not a third party but God himself who pays the debt, he balances the books, so to speak, by paying himself off, at a cost to himself. For Socinus, God is free to do as he wills in response to human sin: justice and mercy are seen as alternatives. Penal substitutionary theory, however, makes two points against this view of God’s freedom:  (a) God is not free without cost to himself, and (b) The cost is necessary because God does not set aside justice when he exercises mercy; rather he forgives in such a way as to satisfy justice. Justice and mercy are not alternatives, and in the cross, God administers both without contradiction.

To sum up, the problems and objections to a substitutionary atonement theory come from both  the outside world of modernity, and from within the theological circle. The former seem either to be conceptual, such as how the particular can impact the universal, or socio-political, such as the “imperialism ” of universal religious claims or the negative impact of such views on victims or on people’s mental health. These outside pressures need to be addressed in formulating an adequate atonement theory, but, in my view, they are not decicisive and must not be allowed to pre-empt the discussion. The modern theologies that have  responded to these external pressures by giving ground on important features of traditional soteriology are uniformly unsatisfactory  and in the end raise as many questions as they answer.

More  challenging, I think, are the criticisms from within the theological circle.  An attempt to make a credible case for an atonement theory that does justice to both the nuances of the biblical narrative and the experience of real people is difficult at best.  Some of the pitfalls we have reviewed are as follows: views of God which are morally offensive, that see him as a vindictive punisher (or, on the other hand, failing to deal with sin and evil, which we did not mention); views which emphasize retributive justice at the expense of other elements; views which emphasize sacrifice so as to imply that  Christ died to propitiate God’s anger;  views which separate God’s justice from God’s mercy; views that are moralistic or legalistic; views in which substitution language is used mechanically and impersonally, neglecting the intentionality of the cross and the dimensions of Christ’s obedience; views which either emphasize the finished nature of atonement so strongly that they invite human moral complacency, or, conversely, views which develop a morbid preoccupation with self-inflicted suffering.   To read this list is to quickly realize that there are partial truths imbedded within all these various misconceptions and distortions.  The complexity of the biblical materials insures that no one theory will be comprehensive.  But  awareness of the problems prepares us for the important task of asking which elements are profitable for a credible atonement theory and which should be avoided.