The Work of Christ

Begin with the difference between this lesson and the last one. The person of Christ was defined. Four ecumenical councils, a creed recited every Sunday in most of the world's churches, and a set of terms the church agreed to use and not to use. The work of Christ has never been defined that way. No ecumenical council has ever promulgated a doctrine of the atonement. The creeds say that for us and for our salvation he came down and was crucified, that he suffered and was buried, and then they move on. The church that fought for four hundred years over the word homoousios has never required its members to hold any particular account of how the cross saves.

That silence is a datum, not an oversight, and it should shape how you hold your own view. It does not mean the question is unimportant, and it does not mean any answer will do. It means that the church has treated the fact as confessional and the mechanism as something Christians may argue about while remaining in communion, which is a judgement a teacher ought to respect even when he is convinced.

What is asserted is a good deal. That Christ died for our sins, according to the Scriptures (1 Corinthians 15:3), which Paul received as a tradition older than his letters. That his death was purposeful, not a miscarriage of justice God then made the best of: he was delivered up by God's set plan and foreknowledge (Acts 2:23). That God is the actor throughout, presenting Christ as an atoning sacrifice (Romans 3:25), not counting men's trespasses against them (2 Corinthians 5:19), sending his Son as the atoning sacrifice for our sins because he loved us first (1 John 4:10). That the death actually achieved something rather than making something possible in the abstract: we have been reconciled, redeemed, forgiven, justified, cleansed, brought near. And that it was for us, hyper hemon, a phrase that saturates the New Testament.

It is also asserted in six registers, and no writer apologises for the mixture. In the temple: the Lamb of God who takes away the sin of the world (John 1:29), one sacrifice for sins (Hebrews 10:12), an atoning sacrifice in his blood (Romans 3:25). In the law court: condemned sin in the flesh (Romans 8:3), became a curse for us (Galatians 3:13), so as to be just and to justify the one who has faith in Jesus (Romans 3:26). In the slave market: a ransom for many (Mark 10:45), redemption through his blood (Ephesians 1:7). On the battlefield: having disarmed the powers and authorities, he made a public spectacle of them, triumphing over them by the cross (Colossians 2:15); by his death he might destroy him who holds the power of death, that is, the devil (Hebrews 2:14). In the family and the city: reconciliation, adoption, peace, access. And in the pattern of a life: Christ suffered for you, leaving you an example, that you should follow in his footsteps (1 Peter 2:21), a sentence which stands three verses before he himself bore our sins in his body on the tree.

Six accounts follow of how those texts hang together. They are set out in the order in which each was first given a full statement, so that nothing should be read into the sequence; the oldest comes first and the newest last, and that is the only principle at work. Each is stated as its own defenders state it, with the texts they actually argue from, and each is followed by a second paragraph opening with the objection its critics press hardest and giving the answer its defenders make to it. The six are of comparable length, and you should read all twelve paragraphs before deciding anything. One of the six, penal substitution, then gets a further section to itself, and the head of that section says why and what should not be read into it.

Recapitulation. Irenaeus of Lyons, writing against the Gnostics around 180, took his key word from Ephesians 1:10, where God's plan is to bring all things together in Christ; the Greek verb there, anakephalaiosasthai, means to sum up or gather under a head. On this account the Son saves by becoming what we are and living the human story again from the inside, getting right at every point what Adam got wrong. In Against Heresies III.18.1 Christ recapitulates in himself the long line of human beings. In V.16.3 the disobedience that came about by a tree is set right by the obedience worked out on the tree of the cross. In II.22.4 Irenaeus argues that Christ passed through every age of human life, infancy, childhood, youth and maturity, sanctifying each by inhabiting it. The whole is caught in his preface to book five: he became what we are, that he might bring us to be what he is. Athanasius says the same in On the Incarnation 54: he was made man that we might be made God. The texts its defenders work from are Romans 5:12-21 and 1 Corinthians 15:21-22 and 45-49, where Adam and Christ are the two heads of the race, together with Hebrews 2:10-18, where he had to be made like his brothers in every way, and Hebrews 5:8-9, where he learned obedience from what he suffered and, having been made perfect, became the source of eternal salvation. The account is the backbone of Eastern theology, feeding the doctrine of deification from 2 Peter 1:4, and it has serious modern defenders in the West, notably T. F. Torrance and his brother James, who argued that the atoning work is the whole vicarious humanity of Christ, from conception to ascension, and not the last six hours alone.

The objection is that recapitulation explains the incarnation better than it explains the crucifixion. If salvation comes by the Son living a human life rightly, the cross becomes the last event of that life rather than the thing that saves, and it is hard to see why a violent death in particular was required. Critics add that it can slide into a doctrine where sin is a defect to be repaired rather than a guilt to be dealt with, so that forgiveness drops out of the account altogether and what is left is a therapy rather than a pardon. The defenders answer that the objection misreads the shape of the obedience: Philippians 2:8 says he became obedient to death, even death on a cross, so the death is not an appendix to the obedient life but its extremity and its proof, and Hebrews says explicitly that he was made perfect through suffering. They add that Irenaeus himself has plenty of sacrificial and ransom language and never proposed recapitulation as a rival to it, and that the charge of losing guilt is answered by Romans 5:19, where through the obedience of the one man the many will be made righteous, which is a verdict and not only a repair. Against the complaint that it is vague, the Torrances answer that it is the only account that gives the thirty years before the cross any saving significance at all, and that a doctrine which makes the life of Jesus theologically idle has a problem of its own.

Ransom and Christus Victor. The oldest popular account reads Mark 10:45 straight: a ransom is a price paid to free a captive, humanity is in captivity to sin, death and the devil, and Christ's life is the price. Origen, in his commentary on Matthew, said plainly that the ransom was paid to the evil one. Gregory of Nyssa, in the Catechetical Oration, used the image of a hook: the deity was concealed under the flesh as bait, the devil swallowed what he could not hold, and his grip on humanity was broken by his own greed. Augustine used the same idea with a different picture, calling the cross the devil's mousetrap, baited with the Lord's blood. The texts its defenders work from are Hebrews 2:14-15, that by his death he might destroy him who holds the power of death and free those who all their lives were held in slavery by their fear of death; 1 John 3:8, that the Son of God was revealed to destroy the works of the devil; Colossians 2:15; John 12:31; 1 Corinthians 15:54-57. The Swedish Lutheran Gustaf Aulen restated the motif in lectures published in Swedish in 1930 and in English the following year under the title that has stuck to it ever since, Christus Victor, arguing that this dramatic idea of God in Christ fighting and defeating the tyrants that hold humanity, sin and death and the devil and the law, was the classic view of the fathers and of Luther, and that it is not a rational theory at all but a story the church tells. It has vigorous modern defenders, including the Mennonite J. Denny Weaver, whose Nonviolent Atonement argues the victory is won without any divine violence, and N. T. Wright, whose Day the Revolution Began puts the defeat of the powers at the centre while holding that it is accomplished through a death that deals with sin.

Three objections are pressed. First, the ransom-to-the-devil form grants Satan rights over humanity which scripture never grants him, and it makes God a party to a transaction with the enemy; that objection is not modern, since Gregory of Nazianzus made it in his fourth-century Easter oration, refusing outright the idea that the blood was offered to the evil one. Second, Aulen's historical claim about the fathers and about Luther has been widely challenged, since the same writers he cites also use sacrificial and substitutionary language freely, so his account of the tradition is an argument to be checked against the sources rather than a report of them. Third, Christus Victor describes the outcome more clearly than the means: it says the powers were defeated and does not say how a crucifixion defeats them. The defenders reply that the ransom image never required a recipient, since no New Testament text names one, and that the victory's means is stated where the powers' hold is located, in sin and the law's condemnation, which is why the same sentence in Colossians 2:14-15 cancels the debt ascribed to us in the decrees against us and disarms the powers. Whether that reply keeps Christus Victor a distinct account or quietly borrows another one is exactly what its critics ask. Its defenders answer that the burden of proof runs the other way: they hold the victory language to be the most frequent way the New Testament describes what the cross accomplished, and a model has to earn the right to treat the commonest description as the derivative one.

Satisfaction. Anselm of Canterbury wrote Cur Deus Homo, Why God Became Man, in 1098, as a dialogue with the monk Boso, and set out to argue by necessary reasons, leaving Christ aside as a premise, what the situation requires. Sin, for Anselm, is not paying God the honour owed to him, and honour here means the right ordering of creation rather than a lord's wounded feelings. A debt is owed. The sinner cannot pay it, because everything he could offer he already owes; and the debt is measured by the one it is owed to, so it is infinite. Anselm's line to Boso, in the usual English rendering, is that he has not yet considered how heavy a weight sin is. The alternatives are exact: either satisfaction or punishment, and punishment would mean the undoing of the creature God made for blessedness. So satisfaction must be made. Only a human ought to make it; only God can make something of infinite worth; therefore the one who makes it must be both, which is the answer to the book's title. Christ, being sinless, did not owe death, so his voluntary death is something over and above what he owed, a gift of infinite value, and the reward due to him, needing nothing himself, is passed to those he came to save. The texts his defenders work from are Romans 3:24-26, Hebrews 9:11-14 and 10:1-14, where the offering does what the blood of bulls and goats could not, and Hebrews 2:17, to make atonement for the sins of the people (the BSB footnote offers to make propitiation).

The standard objection, which Adolf von Harnack pressed in his History of Dogma, is that this is feudalism dressed as theology: a lord's honour, a vassal's debt, a satisfaction rendered, with an eleventh-century social order projected onto God. The reply is a live argument and not a formality. John McIntyre in 1954 and, from an Eastern Orthodox side, David Bentley Hart in a 1998 article argued that the feudal reading misses Anselm's actual vocabulary, in which honour is the rectitude and beauty of the created order rather than a baron's status, and that the book's logic is about the restoration of an order, not the soothing of an offended party. A second objection concerns method: Anselm argues largely from reason, with the scriptures brought in afterwards, which makes the result look like a deduction from a concept of God. And a third, from a different direction: satisfaction on Anselm's own terms is the alternative to punishment, which means that reading Anselm as an early penal substitutionist misreads him, and that a chunk of what is often credited to him belongs to someone else. His defenders answer the charge about method by pointing out that the whole book is framed as an argument offered to unbelievers who will not grant the scriptures, so the rational form is a deliberate concession to his readers rather than a claim that the doctrine could be worked out without revelation. They also note that Anselm has very little to say about the resurrection or about the life of Jesus before the cross, and treat that as a gap to be filled rather than a fault in the argument.

Moral influence. Peter Abelard, in his commentary on Romans written in the eleven-thirties, objected to the ransom account on the ground that the devil can have no just rights, and pressed a harder question at the satisfaction account: how could the death of an innocent man please God, or how could the murder of the Son be the price of forgiving the murder of the Son? His own answer, in the excursus at Romans 3:26, is that redemption is that supreme love kindled in us by the passion of Christ, which frees us from slavery to fear and binds us to God in love, so that we do all things out of love rather than dread. The texts its defenders work from are these. Romans 5:8: but God proves his love for us in this, while we were still sinners, Christ died for us. 1 John 4:10 and 4:19: he loved us and sent his Son, and we love because he first loved us. John 15:13, greater love has no one than this. 1 Peter 2:21, leaving you an example, that you should follow in his footsteps. Galatians 2:20, the Son of God, who loved me and gave himself up for me. Modern defenders include Hastings Rashdall, whose 1915 Bampton Lectures argued the case in its frankest form, Horace Bushnell in nineteenth-century America, and, in a more careful key, Paul Fiddes, who argues that a transformation actually worked in the sinner by the love displayed is an objective effect and not a merely subjective one. Schleiermacher's account, in which Christ's God-consciousness is communicated to those drawn into his fellowship, belongs to the same family.

The objection is an old one. If the death does nothing but demonstrate love, it demonstrates nothing, because a demonstration must be a demonstration of something done. The usual illustration is that a man who leaps into a river to prove his love for a woman standing safely on the bank has shown not love but derangement; what makes the leap love is that she was drowning. Romans 5:8 says God proves his love in that Christ died for us, and the force of the proves depends on the for us being a real transaction. Critics add that the account struggles with the texts about wrath, curse and propitiation, which it has to read as figures of speech, and that it can leave sin as an ignorance to be corrected rather than a guilt to be remitted. The defenders answer that Abelard is regularly caricatured. Richard Weingart's study of his soteriology argued that Abelard retained redemptive and sacrificial language and that he was no bare exemplarist, and that the transformation he describes is God's act in the believer and not the believer's own. It is worth adding that Abelard was condemned at the Council of Sens, which historians now generally date to 1141, on charges drawn up by Bernard of Clairvaux and on a list of propositions Abelard denied holding, so the position that bears his name in the textbooks was partly assembled by a hostile prosecutor. Defenders further reply that their account is the only one that says plainly what the cross does to the person who believes it, which the others have to add as a separate doctrine of application.

Penal substitution. The Reformers took the legal imagery and made it the centre. Luther described the joyful exchange, in which Christ takes the sinner's sin and the sinner takes Christ's righteousness, and in his 1535 lectures on Galatians said of Galatians 3:13 that Christ became, by imputation, the greatest sinner, bearing the persons of all sinners. Calvin, in the Institutes 2.16, argued that Christ was condemned by a judge in a formal trial precisely so that it would be visible that he took the place of the guilty, and that he bore in his soul the judgement the guilty deserved. It passed into the confessions: the Heidelberg Catechism questions 37 to 39, Westminster Confession 8.5, and the second of the Thirty-Nine Articles, where Christ suffered to reconcile his Father to us and to be a sacrifice for original guilt and for actual sins. The texts its defenders work from are these. Isaiah 53:5, he was pierced for our transgressions; he was crushed for our iniquities; the punishment that brought us peace was upon him; and 53:6, the LORD has laid upon him the iniquity of us all; and 53:10, it was the LORD's will to crush him. Romans 3:25-26, where God presented Christ as an atoning sacrifice so as to be just and to justify the one who has faith in Jesus. Galatians 3:13, Christ redeemed us from the curse of the law by becoming a curse for us. 2 Corinthians 5:21, God made him who knew no sin to be sin on our behalf. Romans 8:3 and 1 Peter 3:18. Its modern defenders include Leon Morris, John Stott, J. I. Packer, whose 1973 Tyndale lecture argued that it is a confession of mystery rather than a rationalist theory, and more recently Simon Gathercole, who has defended substitution exegetically against the charge that the texts do not carry it.

The objection pressed hardest is that punishing an innocent man is unjust on any ordinary account of justice, and that guilt is not the kind of property that can be moved from one person to another; alongside it runs the charge that the model sets a wrathful Father against a merciful Son, and that preached plainly it does harm to people who have been on the receiving end of a father's anger. Those objections and the answers to them are worked at length in the next section, which gives this model more room than any of the other five gets, for a reason stated there. Two matters of history belong here. The first is that the clean line usually drawn between Anselm's satisfaction and the Reformers' penal substitution does not survive contact with the sources: Thomas Aquinas, between the two, treats the passion as satisfaction while also saying that Christ bore the punishment due to our sins, in the Summa Theologiae III questions 46 to 49, and Trent teaches in its sixth session that Christ merited justification for us by his passion and made satisfaction to God the Father. The second is that the position has always carried an internal warning from its own defenders: Packer's lecture argues that it must be held as a model pointing into mystery, and that stating it as a tidy mechanism of transferred units of punishment produces something the Reformers did not teach and its critics are right to reject.

The governmental view. Hugo Grotius, the Dutch jurist, published a defence of the catholic faith concerning the satisfaction of Christ in 1617, written against Faustus Socinus, who had argued in a treatise on Christ the saviour that satisfaction is incoherent, since a debt either is paid, in which case forgiveness is not forgiveness, or is forgiven, in which case no payment was needed. Grotius answered by changing the category. God in the atonement acts neither as an offended private person nor as a creditor but as a ruler, a rector, responsible for the moral government of the universe. A ruler may relax the letter of a penalty, but he may not do so in a way that teaches his subjects that the law does not matter. So God accepts the death of Christ as a penal example: not the exact penalty owed by sinners, not a transfer of their guilt, but a public demonstration, at the highest possible cost to himself, of what sin deserves and of how seriously the law is held, which makes forgiveness possible without the collapse of moral order. The texts its defenders work from are the word Paul uses twice in Romans 3:25-26, which the BSB renders to demonstrate, a public showing of God's righteousness, and Isaiah 42:21, that the LORD was pleased, for the sake of His righteousness, to magnify His law and make it glorious. It was taken up in the New England theology by Joseph Bellamy, Samuel Hopkins and Jonathan Edwards the younger, and received its fullest systematic statement from the Methodist theologian John Miley in The Atonement in Christ in 1879.

The objection is that a demonstration is not a bearing. If Christ did not actually bear the penalty, and if God could have forgiven without the cross but chose this way for its effect, the necessity of the death is rhetorical rather than real, and Galatians 3:13 and Isaiah 53:6 are being read as pictures of something that did not occur. Critics also press that the view makes God's justice a matter of policy rather than of nature. Its defenders reply that a demonstration costing the ruler his own Son is not a display in any thin sense, that the account takes with full seriousness the public and cosmic dimension of what the cross declares, and that it preserves genuine forgiveness, which a strict payment model arguably does not, since a debt paid in full leaves nothing to forgive. It should be recorded, against a common shorthand, that this is not simply the Arminian view: Arminius himself held to satisfaction, John Wesley preached substitution in the plainest terms, and the governmental account has been held and rejected on both sides of that divide. Its defenders add that the demonstration language is not imported but Paul's own, since he twice says the cross was to demonstrate God's righteousness, and that a model built on the public character of the death fits the fact that the New Testament never treats the cross as a private settlement. They press a further point, which is Socinus's original question turned to their own use: if the penalty was exacted in full from a substitute, the sinner is released by accounting rather than pardoned, and the language of not counting trespasses describes something a discharged debt does not need.

A word first about the shape of this section, so that you are not left to discover it. Penal substitution gets more room in this lesson than the other five accounts do, and the figures are worth stating rather than hiding. In the section above, each of the six has its own paragraph opening with its own named weakness, and the six run between about 530 and 580 words each, which is as even as they could be made. Then this section gives penal substitution five further paragraphs, so that in total it receives roughly three times the dedicated treatment any other model gets. That is deliberate. The reason is that this is the live argument of the present moment, the one a student will actually be asked about by someone who has read a paperback on it, and the objections are serious enough to be given in their authors' own words rather than summarised away. The extra room is not a verdict, and it does not run in either direction. A model that takes this long to answer its critics is not thereby refuted, and a model whose defenders answer at length is not thereby vindicated. Read the space as a measure of how loud the present argument is, not of who is winning it.

The current argument is about penal substitution, and it is worth having in the actual words used. In 2004 Steve Chalke and Alan Mann wrote in The Lost Message of Jesus that the cross is not a form of cosmic child abuse, a vengeful father punishing his son for an offence he did not commit. The charge had been made earlier and more radically. In a 1989 essay Joanne Carlson Brown and Rebecca Parker argued that an image of a father who requires the suffering of a son to be reconciled sanctifies obedient suffering and has been used to keep abused women in place, and Rita Nakashima Brock and Parker pressed the same point later. From a womanist standpoint Delores Williams, in Sisters in the Wilderness in 1993, rejected the language of redemptive surrogacy altogether, on the ground that surrogacy is precisely what was done to black women and cannot be the shape of their salvation. A separate line of objection comes from the Anabaptist and peace-church traditions, which argue that a salvation achieved by an act of divine violence cannot ground a church commanded to renounce violence. And the argument has run inside evangelicalism as much as against it: the Evangelical Alliance in Britain convened a symposium on the question in 2005 and published the proceedings as The Atonement Debate in 2008, with defenders and critics between the same covers. These are not frivolous objections and they are not made by people with no stake in the faith.

The answers its defenders give are these, and they should be weighed rather than waved at. First, the objection pictures three parties, an angry father, an innocent third party and a guilty race, and the doctrine has never had three. John Stott's central reply in The Cross of Christ is that this is the self-substitution of God: God was reconciling the world to himself in Christ, so the one who provides the substitute and the one who is the substitute are not two gods with different dispositions. 1 John 4:10 makes the love the origin of the atoning sacrifice and not its result. Second, the Son is not a victim of the Father's will but a willing agent of his own: no one takes it from me, but I lay it down of my own accord (John 10:18); he offered himself through the eternal Spirit (Hebrews 9:14). Third, the council of 681 confessed two wills in Christ precisely so that the human willing in Gethsemane could be a real and free consent rather than an overpowering. Fourth, the defenders grant that versions of the doctrine which do separate the persons of the Trinity are genuinely defective, and hold that the remedy is to correct them rather than to abandon the doctrine. The critics reply, not unreasonably, that a doctrine so often preached in the defective form has to answer for the form in which it is actually heard.

A second objection is about justice: punishing the innocent is unjust, and guilt is not the sort of thing that can be moved from one person to another. The answers appeal to union. 2 Corinthians 5:14 says that one died for all, therefore all died, and Romans 6:3-11 has believers crucified and buried with him, so the substitute is not an unrelated stranger but the head in whom his people are included, on the pattern of Adam in Romans 5:12-21. The willingness of the substitute is also held to change the moral analysis, since a person may voluntarily assume a liability. Critics answer that the more weight is put on union and inclusion, the less the model is a substitution at all and the more it becomes participation, which is a different account. That exchange is where much of the best current work sits, and it has not closed.

A third objection is exegetical and concerns the word at Romans 3:25. Whether hilasterion means propitiation, the turning away of God's wrath, or expiation, the removal of sin, is a genuine and unsettled dispute between the case C. H. Dodd made in the nineteen-thirties and the case Leon Morris made in 1955, and this course has already worked it in full in the Hermeneutics track, in the module on the tools, in the lesson on word studies. Do not repeat that argument here; go back and read it. What belongs here is what follows for the atonement, and it is less than partisans on either side suggest. The BSB's atoning sacrifice at Romans 3:25 and 1 John 4:10 is a compromise rendering that decides neither way, and its footnotes give or as a propitiation at both places and or to make propitiation at Hebrews 2:17, so an English reader meets the dispute in the margin. Expiation is compatible with more than one of the six accounts above, and propitiation on its own does not entail a transfer of penalty, since a sacrifice may avert wrath without the mechanism being penal. There is also a third reading with real support, including Daniel Bailey's Cambridge study, that hilasterion in Romans 3:25 is the mercy seat itself, the kapporet of Leviticus 16, which is the word's meaning at Hebrews 9:5, its only other New Testament occurrence. On that reading Paul is saying that God publicly set forth Christ as the place where God meets his people, which is a claim of a different kind again.

The traffic runs in both directions, and a lesson that gave only the objections to one model would be doing what this course exists not to do. Defenders of penal substitution press the moral influence account on Romans 5:8, as above; they press Christus Victor on its silence about how; they press the governmental view on Isaiah 53:6 (two cautions about the English of that chapter belong here: first, the BSB capitalises every pronoun for the Servant, He and His and Him, which is the translators' reading that he is Christ, since Hebrew has no capitals, and the Jewish tradition reads the Servant as Israel or a faithful remnant, so a student who holds that reading is not misreading the Hebrew, and the Christian case rests on the New Testament's own use of the chapter, as at 1 Peter 2:24; second, the BSB prints the punishment that brought us peace at 53:5 for a Hebrew word that can also mean chastisement or discipline, which critics of the penal reading stress and its defenders answer from the surrounding words crushed, pierced and laid upon him, so neither side should be settled from the English alone); and they argue that recapitulation and satisfaction both need something like penal categories to explain why death in particular was required. Defenders of the other accounts answer each of those, and have. What should be noticed is that the arguments are exegetical on both sides and that each position has adherents who know the texts thoroughly. Nobody in this argument is ignorant of the Bible, and a student who explains a disagreement by saying the other side has not read the passage has stopped thinking.

Here is what has never been seriously disputed in the church. That Christ died for our sins and that his death was according to the scriptures. That it was purposed by God and not merely permitted. That it was for us, and that its benefit reaches us rather than remaining a transaction elsewhere. That it achieved forgiveness, reconciliation, redemption and victory over sin, death and the devil, all four, since all four are asserted in the texts. That God is its author. And that it is inseparable from the resurrection, which the next lesson takes up. A Christian who holds all of that and is undecided about the mechanism is holding everything the creeds ask for.

The confessional bodies have gone further, and where they have, a student should know it rather than guessing. The Lutheran and Reformed confessions commit their churches to a substitutionary and penal account in so many words. The Catechism of the Catholic Church teaches the atonement under sacrifice, obedience and substitution, quoting 2 Corinthians 5:21 and affirming that Christ made satisfaction for us, without canonising any single theory of the mechanism, and Catholic theologians accordingly differ on how penal the satisfaction is. Eastern Orthodox theology characteristically foregrounds victory, recapitulation and deification, and a strand of modern Orthodox writing criticises what it regards as a juridical captivity in Western soteriology; but Orthodox writers differ among themselves, and sacrificial and substitutionary language is present in the Orthodox liturgies, so the contrast is often drawn more sharply by Western admirers than by Orthodox theologians. Anglican, Wesleyan, Baptist, Pentecostal and Anabaptist bodies range across the field, and the Anabaptist tradition has produced some of the most searching modern criticism of penal accounts, on grounds that come out of its convictions about violence.

Two moves are frequently made at this point, both of which present themselves as modesty and neither of which is neutral. The first is to name one model as the true mechanism and cast the rest as partial glimpses of it, usually by showing how each of the others can be derived from the one. That is a real position and it may be right, but it is a conclusion, and it has to be argued from the texts rather than assumed by the way the models are introduced. Notice that whoever makes this move has to explain why the New Testament writers, who had the opportunity, never rank their own images.

The second is to call them all metaphors, complementary pictures of a reality nobody can describe, and to take no further view. This sounds generous and it is a substantial claim of its own: it says either that the mechanism is unknowable or that the language of scripture at this point does not refer to anything in particular. It also cannot be quite true, because the models make claims that contradict each other. The governmental view says Christ did not bear the penalty owed; penal substitution says he did. Anselm says satisfaction and punishment are alternatives; the Reformers say the satisfaction was a punishment. Abelard says God required no payment; Anselm says the moral order did. Those are not different angles on one shape. At least one of each pair is false.

So what should you do? Hold the assertions of scripture firmly, learn all six accounts well enough to state each in a form its defenders would accept, and come to a considered view while knowing exactly what would count against it. Notice which account your own tradition assumes without naming, because every tradition has one and most do not know they are teaching it in their hymns. Notice too that most serious theologians hold a combination rather than a pure form, and that the honest question between them is usually not which image is true but which is doing the explaining and which is being explained. When you preach the cross, preach what the New Testament says happened, in the words it uses, in all six of its registers. And when a student asks you how it works, the best professor he will ever meet is the one who tells him what the church has settled, what it has not, and where he can go to read the argument for himself.