Here is a common argument against the penal substitutionary view of the atonement: The penal substitutionary view of the atonement entails a retributive view of punishment, but retribution is insufficient moral justification for punishment. Therefore, the penal substitutionary theory of the atonement is false. In response, I shall argue that the retributive justification of punishment is well supported by the Bible and by commonly held intuitions about meritorious actions.
The penal substitution view of the atonement holds that sinful human beings deserve divine punishment and that Jesus Christ was punished in their place. According to this view, punishment is deserved by a sinner but can be absorbed by someone other than the person who has committed the sin:
The Father, because of his love for human beings, sent his Son (who offered himself willingly and gladly) to satisfy God’s justice, so that Christ took the place of sinners. The punishment and penalty we deserve was laid on Jesus Christ instead of us, so that in the cross both God’s holiness and love are manifested
Tom Schreiner contends that the act of sending the Son was carried out by the Father because of the Father’s love for sinners and in order to satisfy the justice of the Godhead. Jesus Christ is punished with the punishment that humans deserve in order to show both God’s holiness and his love. The crucial element of the definition is the appeal to just desert and satisfaction of God’s justice. On this view, punishment satisfies the demands of God’s justice and substitution expresses the love God has for those sinners.
Central to the penal substitutional view of the atonement is punishment. Punishment is some level of harm or suffering levied against a person who has done some wrong action. Punishment relates to actions. In the Old Testament, the Lord tells the people that he will repay them according to their deeds (Jer 17:10). Paul tells us that indecent acts are due a penalty (Rom 1:27). For someone to deserve punishment, they must have done something to deserve it. Punishment is also a kind of action that causes harm to the person being punished. Luis Pojman defines punishment as an “evil inflicted upon another person who is judged to have violated a rule.” The evil human beings deserve for sin includes death and eternal separation from God. On this view, it is often said that the punishment is necessary – if a person sins, then there must be a punishment. The latter assumption is usually based on the idea of debt, a person who sins owes some form of payment to an offended party. According the penal substitutionary view, the kind of punishment assumed is retributive. According to the retributive theory of punishment, the infliction of some harm or suffering on another person is just only if (i) the person is a free moral agent and has committed a sin for which the person is morally responsible, (ii) the punishment is given to a person who deserves it, and (iii) the punishment is administered by a person who has the authority to punish the sinner.
Retributive views of punishment have been consistently present in the history of the church and have been related to theories of the atonement. In The City of God, Augustine considers the final judgment of Christ to be a moment in which he will “give an all embracing judgment on the race of demons and on the human race, condemning them to misery as the deserved retribution for the first sins of the race.” Calvin assumes a retributive justification for punishment in his doctrine of the atonement when he writes:
This is our acquittal: the guilt that held us liable for punishment has been transferred to the head of the Son of God (Isaiah 53:12). We must, above all, remember this substitution, lest we tremble and remains anxious throughout life—as if God’s righteous vengeance, which the Son of God has taken upon himself…if Christ made satisfaction for our sins, if he paid the penalty owed by us, if he appeased God by his obedience—in short, if as a righteous man, he suffered for unrighteous men—then he acquired salvation for us by his righteousness.
Finally, Charles Hodge presented what has become a standard presentation of penal substitution among reformed Christians. Hodge argued for penal substitution from a retributive theory of punishment. For Hodge, the idea of retribution is rooted in the nature of God: “If justice is that perfection of the divine nature which renders it necessary that the righteous be rewarded and the wicked punished, then the work of Christ must be a satisfaction of justice in that sense of the term.”
Retributive theories of justice have been attacked on both ethical and theological grounds. In place of the retributive view, many people have advocated for theories of punishment that emphasize deterrence or rehabilitation. According to these views, punishment is justified only if there is a measurable good produced by the punishment. Deterrence theories defend punishment as reducing crime in a population by demonstrating what would happen to people if crime is committed. On this view, punishment is justified if it can be shown that it adequately deters people from crime. Rehabilitation theories include acts of punishment that are considered therapeutic. For example, time spent incarcerated is supposed to be time spent curing criminals of their inclinations to perform wrongful actions. Both alternative these theories of punishment appeal to the utility of punishment. On these views, punishment is justified by appeal to its consequent goods, either its therapeutic or deterrent.
Some theologians have defended punishment on the grounds that it produces a good that justifies punishment. For example, Hans Boersma rightly argues against those who say that God cannot be connected with violence of any kind, but then justifies God’s violent acts by suggesting that they are necessary for some resulting good. Punishment, a species of violent act, “serves the purpose of pure, eschatological hospitality.” Christopher Marshall argues similarly. He concludes that justice is not achieved in punishment but only in the outcome of restored relationships. Though both Boersma and Marshall espouse a justification for punishment that is consistent with a consequentialist ethic, their criticisms of retributive theories of punishment assume a consequentialist ethic. Though it is beyond the scope of this paper, there are good reasons for thinking that consequentialist theories of ethics are incompatible with basic Christian beliefs.
In contrast, a deontological theory of ethics holds that the moral status of an action is determined by some intrinsic feature (whether it is rational or breaks a moral law). Immanuel Kant defended punishment on deontological grounds arguing that to punish for any other reason apart from the intrinsic value of desert was to treat a person as a means to an end.
Punishment by a court … can never be inflicted merely as a means to promote some other good for the criminal himself or for civil society. It must always be inflicted upon him because he has committed a crime. For a human being can never be treated merely as a means to the purposes of another or be put among the objects of rights to things: his innate personality protects him from this, even though he can be condemned to lose his civil personality. He must previously have been found punishable before any thought can be given to drawing from his punishment something of use for himself or his fellow citizens.
Kant’s theory of retribution rests upon his categorical imperative and is impersonal in nature. More recently, Herbert Morris has defended a conception of retributive punishment according to which punishment is justified on the grounds that a wrongdoer has taken advantage of the community in which he or she lives. Punishment is meted out in order to secure the “equilibrium of burdens and benefits.” In contrast, a divine command theory according to which the moral status of an act is determined by God’s prescriptions, punishment is not only the violation of moral law, but a personal affront to God. Consequently, and in contrast to a Kantian theory of punishment, the justification for punishment is ultimately rooted in the divine nature rather than an impersonal moral law.
In recent decades, and in consequence to developing ethical assumptions, theories of civic justice have weakened the assumption that punishment is retributive. They argue that just as Anselm’s satisfaction theory of the atonement assumed a feudal system of society, the penal substitutionary view assumes a retributive theory of justice. Consequently, the persuasive power of the theory lies not in exegetical considerations, but in cultural assumptions about just forms of treatment for wrongdoers in society. And just as we have abandoned Anselm’s suggestion that the death of Christ restores the cosmic feudal order, we will abandon the view that justice is primarily retributive. And if we do, we should abandon a penal substitutionary view of the atonement.
In reply, proponents of a retributive view of punishment can appeal to two sources: rational, non-biblical sources and theological, biblical sources. In the following section, I will argue that non-biblical arguments for a retributive view of punishment are only weakly supportive of retributive punishment. Consequently, the retributivist should appeal primarily to biblical texts in order to make their claim. In the following section, I will argue that the Bible provides sufficient evidence for the claim that God’s nature entails a retributive form of punishment.
Non- Biblical Arguments for Retributive Punishment
According to the retributive view, punishment is deserved by the one who performs an immoral action. Though it is clear that if someone deserves punishment, they must have committed a sin, it is less clear that a sinful action is sufficient to morally justify punishment. Some theologians and philosophers have attempted to find rational grounds for accepting this view. However, it is far from clear that they succeed. In this section, I will consider non-biblical arguments for the justification of a retributive theory of punishment and conclude that it is weakly supported by intuitions about meritorious human actions.
In Cur Deus Homo, Anselm argues that sinful actions are morally sufficient to warrant punishment. Though his arguments are theological, on a plausible interpretation of what he says, Anselm attempts to ground his conclusions in rationally acceptable assumptions. In section XII, Anselm argues that it is not right for God to pass over human sin without punishment. If Anselm can make this case by appealing to reason, then he can show that the punishment of Christ was necessary in order to pay the price for human sin. Since such a view entails a retributive theory of punishment, we would have a good non-biblical argument for a retributive theory of punishment. Unfortunately, Anselm fails to present a justification for his claim without an implicit appeal to a biblical assumption. Anselm gives three arguments. First, Anselm argues that it is not right to pass over sin unpunished:
To remit sin in this manner is nothing else than not to punish; and since it is not right to cancel sin without compensation or punishment; if it be not punished, then it is passed by undischarged…It is not fitting for God to pass over anything in his kingdom undischarged…It is, therefore, not proper for God thus to pass over sin unpunished.
Anselm’s implied reason appears to be that punishment is form of compensation, something owed to God. On Anselm’s view, the atonement pays a debt owed by human beings to God for which they cannot pay. The debt in question is a life lived without sin. Since we cannot deliver such a life, Christ offers his own life in our place. It is not clear, however, that debt and punishment are synonymous in this way. If compensation is what is required, then a perfect life would suffice. What then of the cross? If punishment is not necessary for compensation, then punishment cannot be identical to compensation. If so, then, while Anselm’s argument gives a reason for compensation, it does not give a reason for punishment.
Perhaps, however, Anselm does not have in mind solely that a sinner owes God a non-sinful life, but that, given that the person has sinned, he is owed punishment. This would be so if it is part of the nature of God that he punish sin such that for every sin, there must be punishment. Anselm indicates that the reason for the necessity of punishment is not some abstract law, but the nature of God: “It is not fitting for God to pass over anything in his kingdom undischarged.” The question is: what makes passing over a sin unpunished ‘unfitting’? If doing so is unfitting to a principle assumption of justice, Anselm gives no non-circular reason to accept such a reason. If, on the other hand, Anselm intends to point to the nature of God as determinative of what is just, then he can reply that since God determines what is just according to his nature and God decides that sin cannot be passed over without punishment, then the argument does not beg the question. However, if Anselm’s argument is of this kind, it is no longer an argument from a non-biblical source. Since it is implausible that there is a good non-biblical argument that shows that God’s nature includes a retributive principle, what Anselm’s argument requires at this point is an exposition of the nature of God from the Bible. Consequently, though Anselm’s argument may be correct, it is not an argument grounded in a non-biblical source. It must ultimately be grounded in God’s revelation of himself.
Anselm’s second argument suggests that punishment is the distinguishing feature that marks out the difference between a guilty person and a non-guilty person: “if sin be passed by unpunished [then] with God there will be no difference between the guilty and the not guilty; and this is unbecoming to God.” Bruce Reichenbach rightly replies that Anselm conflates punishment with being liable to punishment. For a person to be liable to punishment they must be guilty. Thus, the guilty and non-guilty are distinguishable whether they are actually punished or not.
Anselm’s third argument suggests that the presence of law entails that someone who breaks the law will be punished:
Justice to man is regulated by law, so that, according to the requirements of the law, the measure of award is bestowed by God… But if neither is paid for or punished, it is subject to no law… Injustice, therefore, if it is cancelled by compassion alone, injustice is more free than justice, which seems very inconsistent. And to these is also added a further incongruity, that it makes injustice like God. For as God is subject to no law, neither is injustice.
Anselm appears to be arguing that if there is no punishment, then there is no law (or, the law is not applied). It is difficult to see why Anselm thinks this is so. Presumably, there is something in the law itself that requires punishment. But this is not a self-evident truth, one that we grasp by grasping the concept of law or by reflecting on the nature of justice. Rather, we would know that law entails punishment if the law said so. But in order to know this we would have to know the law. However, this again would be to appeal to a source outside of reason alone. The Bible may well teach that law demands punishment of those who break it, but we could only know this by an examination of the text itself and not by observing lawbreaking or reflecting on humanity’s relationship with the law. In conclusion, none of Anselm’s arguments succeed in establishing a justification for punishment that does not appeal to an assumption only known through biblical revelation.
Though Anselm’s arguments fail to ground a retributive view of punishment in reason alone, perhaps human intuition about just deserts is sufficient. To ground an ethical judgement in intuition is to suggest that upon reflection a fact about the world that justifies an ethical judgement is self-evident. Louis Pojman attempts to ground just deserts in intuition. Pojman argues that though no argument for the retributive theory is forthcoming, we should be able to affirm its truth by considering the symmetry of merit in the world. When we consider our emotions in response both to some service or harm rendered, we are naturally inclined either to repay a service or a harm in kind. So strong is such an instinct that perhaps it serves as an evidence for a basic assumption about the world: “Every action in the universe has a fitting response, and that response must be appropriate in measure to the original action. It follows that evil deeds must be followed by evil outcomes and good deeds by good outcomes, exactly in proportion to the vice or virtue in question.”
Pojman’s idea is initially appealing. If one wants to convince someone of the truth of retributive justice it is often highly effective to appeal to the intuition that a person who commits a heinous act is deserving of punishment. However, the problem with justifying an ethical judgment in intuitions is that it is largely dependent on there being objective moral truths, those moral truths being simple and not dependent on any other truth, and being clearly self-evident to any human being who cares to reflect on the matter. It is the final condition that is troubling for this theory. Of course, when I reflect on the matter I am disposed to agree, but, as is clear from both contemporary western cultural attitudes and a plethora of academic opinion (all of which presumably written by people who have carried out a great deal of reflection on the matter), not everyone agrees with me. Thus, unless one is already predisposed to believe that sin deserves punishment, one may just dismiss the idea. To accept Pojman’s view one would have to accept his theory of knowledge of moral facts – intuitionism. Though initially plausible (it appears that many moral facts are self-evident), intuitionism is open to strong objections by those who reply that such moral facts are not self-evident. Therefore, though intuitionism is persuasive in some cases, it is far from clear that it can fully justify the retributivist view.
So far, I have argued that reason alone is insufficient to supply a moral justification for retributive punishment. But what if we assume, for the sake of argument, that punishment is morally justified and then reason against alternative theories of justification for punishment? This would lead us to conclude that if punishment is justified, then it is justified on retributive principles. C.S. Lewis argues that if one accepts that punishment is justified, then one must have some moral reason for it. However, there appear to be only three options: retribution, deterrent, and rehabilitation. Lewis argues that neither deterrent nor rehabilitation count as reasons for punishment. Therefore, if we accept punishment as justified, we must accept retribution to be the reason. When we consider punishment in terms of deterrence or rehabilitation we don’t consider it in terms of whether it is just. Rather, we only consider it terms of whether it succeeds in deterring or rehabilitating. Whereas the concept of desert is conceptually linked to the concept of justice, the same is not true of deterrence or rehabilitation. One might reply that justice just is what deters or rehabilitates. However, Lewis asks us to consider whether we would consider ourselves as subjects with moral rights if those moral rights were reduced to whether our punishments contribute either to the deterrence of others from crime or the curing of ourselves from it. If we consider ourselves as moral subjects with rights, then we should consider whether we deserve punishment. Furthermore, Lewis comments that one need not be a judge to determine the success of deterrence or rehabilitation, one only needs to be acquainted with the data. But surely one must be a judge to determine the desert of a person’s punishment. If we consider a justice system to be the domain of a jurisprudence and not merely of data analysis, then we should prefer that the primary justification for punishment is whether we deserve it.
Lewis’ argument can only conclude that if punishment is justified, then it is justified by a retributive theory of punishment. This leaves open the option that punishment is not justified. And, for some theorists, this is the correct conclusion. Punishment, for some is an increase in evil since it doubles the quantity of suffering for every offence. At the very least, Lewis improves the claim of the retributivist given the assumption that punishment is justified.
I’m inclined to agree with all the conclusions of Anselm, Pojman, and Lewis, that sin deserves punishment that that this fact is an evident feature of the world in which we live. However, none of the arguments on offer successfully establish that fact. We may very well intuit that the grounds for punishment is the desert of the sinner, but proving such a conclusion merely from truths about the world or our intuitions is not sufficient. Rather, we need evidence from truths about God’s justice, truths only available through special revelation.
Furthermore, if there is nothing self-evident about retributivism, then the theory is open to the accusation that the only plausible grounds for a non-biblical argument appeal to the nature of vengeance. A victim is owed some suffering to be meted out to the offender. If indeed this is a basic picture of punishment, then it is well attested to by a multitude of victims. However, the justification of human to human vengeance is ruled out firstly by scripture’s clear teaching that human beings should not enact vengeance on each other (Lev 19:18; Rom 12:19) and the obvious complaint that a society operating under a rule based on vengeance is not a desirable society. Indeed, retributivists are at pains to point out that retributive punishment is justified in the absence of the desire for revenge in the victim. However, the Bible’s description of God’s desire for vengeance is coupled with his command to punish sinners. Thus, though human beings are forbidden to act out of vengeance, God enacts his own vengeance in punishing evil doers and commands sanctioned human authorities to punish sinners under their care. The latter form of punishment is carried out by human beings regardless of their desire for revenge, but as the means by which God metes out punishment and avenges offences. Thus, the Bible condemns human vengeance and extolls divine vengeance. Indeed, part of the justification for retributive punishment is God’s personal reaction.
Biblical Arguments for Retributive Punishment
A retributive view of punishment is amply supported by the Bible. The Bible shows us that not only is retributive punishment commanded by God in a civil context (Gen 9, Deut 25:3), but it also grounded in God’s nature (Rom 3:25-26; 2 Thess 2:8-9; Rev 20:11-15). Taken together the scripture provides support for a retributive view of justice that goes beyond a strong moral intuition. What the Bible supplies is evidence that retributive justice is grounded in the nature of God. If we can make this case from scripture, then we have good reason to accept the claims of the retributivist.
Genesis 9:5-6 is a standard text used to support the principle of lex talionis (a life for a life). In this case, those who murder shall be killed. Although there is some debate over whether the text indicates that the one who kills is man or God, the principle is clear: those who commit murder deserve death. As Bruce Waltke points out, such a payment is a demand of God: “This is an obligation, not an option… Blood shed through homicide must be dealt with.” Waltke argues that the obligation to punish murderers carries through to the prescriptions for atoning sacrifices in the Old Testament: “If the blood is not compensated by capital punishment or atoned for, it brings the Lord’s judgment on the land (Deut 19:13; 2 Sam 21:1; 1 Kings 2:9, 31-33).”
The central demand God makes is for lifeblood and comes in the context of a dietary restriction, namely that Noah’s family should not eat the blood of animals. Blood represents the life of the creature including human creatures made in the image of God. Consequently, to take the blood of another human warrants the taking of the blood of the murderer. This is central to the atonement as described by the author of Hebrews who argues that if there is to be forgiveness, blood must be shed (Heb 9:22) and it is only the blood of Jesus that is sufficient to pay the price for sinners (9:12).
In two New Testament passages, retributive justice is meted out by God in return for actions against his people. Paul commences his second letter to the church at Thessalonica with an encouragement for those experiencing affliction. Paul tells his readers that it is just for God to repay in kind those who afflict them. Paul tells the Thessalonians that God will “repay with affliction” those who afflict his reader (2 Thess 1:3-9). The term, ‘repay’ (ἀνταποδοῦναι), means ‘to give back’ or ‘return’ what is due. In this case, those who afflict God’s people will be repaid with affliction from God.
Paul goes on to tell his reader that such a repayment is retributive: God will deal out “retribution to those who do not know God and to those who do not obey the gospel of our Lord Jesus. These will pay the penalty of eternal destruction, away from the presence of the Lord and from the glory of His power” (2 Thess 1:8-9). Retribution, in this case, is not abstracted from God’s wrath. The term, ‘retribution’ (ἐκδίκησιν), is used for ‘vengeance’ or ‘punishment’ and uniquely the prerogative of God (Rom 12:19). Robert Thomas notes the connection punishment has with God’s vengeance but seeks to avoid the conclusion that God is prone to vindictiveness, “It has no overtones of selfish vindictiveness or revenge.” However, as Schreiner points out, God’s wrath should not be reduced to a law like reaction of the cosmos to good and evil. Rather, God’s punishment is uniquely personal. Consequently, “sin is not merely the violation of God’s law but spiritual adultery.” If so, then Paul’s point is not that desert alone justifies punishment, but God’s justice entails punishment carried out by God.
Stephen Travis argues that we should not take retribution as an action of God, but rather take Paul to be suggesting that the fact that those who are persecuted will be saved and others not as ‘poetic justice.’ In other words, this text does not teach that it is in God’s nature to repay sin with punishment but only that those who do sin end up in situations that cause them great suffering. From the perspective of those who are persecuted, this may appear to be a punishment meted out by God in retribution for the deeds of the persecutor, but it only appears this way. However, this is a strain of the plain meaning of the text and if assumed would put into question the rest of the passage. For example, Paul has just claimed that God will make those who suffer worthy of the kingdom (v. 5). Should we also take this to be a “judgment without content”? This would entail regarding both God’s judgments—to make some worthy for the kingdom and to afflict others—only appear to be actions carried out by God. Both would be merely ‘poetic.’
In Revelation 16:4-7, John records the proclamation of the angel who affirms the just punishment of those who “poured out the blood of saints and prophets” (v. 6). The angel tells us that the persecutors of God’s people deserve their punishment and that such an action displays the nature of God: “Righteous are you…because you judged these things; for they poured out the blood of the saints and prophets, and you have given them blood to drink. They deserve it” (vs. 5b-6). As Mounce comments, “In a moral universe God must of necessity oppose evil… Because they had poured out the blood of the saints, God has given them blood to drink.” The sins of the persecutors have rendered them ‘worthy’ of punishment. The strange sounding idiom—that the persecutors should drink their own blood—comes from Isaiah’s declaration that God would feed Israel’s oppressors “with their own flesh, and they will become drunk with their own blood as with sweet wine” (Isaiah 49:26).
Richard Bauckham suggests that the idea of repayment is closely related to the ability of a sinner to understand the justice of God: “Intrinsic to the idea of ‘measure for measure’ judgments is that this is the best kind of justice because its justice should be obvious even to the criminal. To receive something equivalent to what one has done is to find, as it were, one’s own sin boomeranging back at one.” The warning of retribution is supposed to lead sinful people to repentance though John tells us that even in the midst of their experience of God’s wrath, people remain unrepentant (Rev 16:9). John also records the outcry of the saints for justice. Their cry is for God’s vengeful actions against their oppressors: “How long, O Lord, holy and true, will you refrain from judging and avenging your blood on those who dwell on the earth?” (Rev 6:10b). Such an outcry assumes that justice is served through a retributive action of God who will pay back the wrong doer with affliction. Such a cry sounds harsh unless one spends a moment to consider the evils committed by the oppressor. As Schüssler Fiorenza comments: “One can adjudicate the central quest of Revelation in theological terms, however, only if one comprehends the anguish that fuels this outcry for justice and vindication, for divine revenge and restitution for so many lives taken, and so much blood unnecessarily shed.” Perhaps this might help us understand the instincts of the martyrs, but it does not justify it. Instead, what the martyrs’ prayer suggests is that their outcry is grounded in their knowledge of God. This is confirmed by the angel in 16:4-6 and in the witness of the alter in 16:7.
Fiorenza’s observation lends credence to the claims of Pojman that people can generally intuit that evil actions deserve punishment in the same way good actions deserve rewards. However, whereas Pojman’s justification lies in a moral intuition, a theological justification lies in the nature of God. What Pojman intuits is not merely a fact about the world but a fact about its creator who is clearly perceived (Rom 1:18). If so, then the retributivist can appeal to a justification for his view whereby there is a fact about reality—the divine nature—that justifies retributive punishment.
I have argued that a retributivist theory of punishment is assumed by the penal substitutionary theory of the atonement and that retribution is supported by evidence found in scripture about the nature of God and his demands for punishment of sin. There are several outstanding objections to retributive theories of justice to which we now turn.
Many objections to a retributive theory of punishment are theological and motivate extensive criticisms of the penal substitution view of the atonement. In general, theological objections appeal to another feature of God and claim that retributive punishment is inconsistent with that feature. For example, Peter Schmiechen argues that the atonement should attempt to resolve the tension between God’s love and his justice. However, he argues that the penal substitution view of the atonement succeeds by making justice the only significant theological factor:
Instead of the tension of justice and love being resolved in a way that affirms God’s agency and God’s gracious will to save, the theory leads to the necessity of punishment according to distributive justice… We end up with the domination of a particular kind of justice over the entire theological agenda.
Schmiechen argues the Bible’s storyline emphasizes both God’s faithfulness to his creation and his aversion of evil. The penal substitution view succeeds in reducing the atonement to punishment that satisfies God’s justice at the expense of an account of God’s love. Schmiechen points out that the retributive view of justice does not entail that God carry out the punishment for the purpose of loving those whom he saves. Rather, if we accept the retributive theory of punishment, then the substitution of the Son is necessary only to satisfy the justice of God. Love is simply not the issue. On the other hand, if one takes a utilitarian view of punishment—that it is justified only if it produces some good—then one can argue that the reason for punishment of Jesus is the love of those whom God saves.
In reply, retributivists argue that since the Father punishes the Son in place of the guilty party and the Son voluntarily submits himself to the punishment, the punishment both satisfies the demands of justice and demonstrates the love of God for his people. Put simply, justice is served by the punishment of the Son and love is served by the substitution of the Son. The difference with Schmiechen’s view is not that penal substitution proponents cannot resolve the tension between God’s love and his justice without reduction. Rather, the difference is between what is assumed to be a moral reason for punishment. On the penal substitutionary view of the atonement, the moral reason for punishment is solely the sin of the evil doer. On Schmiechen’s view, a punishment is justified only if there is some greater good produced by the punishment. In other words, the debate between the two sides comes down to whether one assumes a deontological or consequentialist ethic. If the penal substitution proponent can adequately account for God’s love and his justice in the atonement, then Schmiechen’s objection fails.
Several other theologians have objected to retributive punishment because they say it is incompatible with the love of God. However, the incompatibility is not between the love of God and retribution per se, but between two theories of ethics. For example, Stephen Finlan argues that the emphasis of God’s actions as detailed in the New Testament are never retributive but restorative: “Certainly there is a need for correction, training, and re-socialization, but punishment for its own sake is a wholly unworthy concept.” In other words, punishment is not justified unless it produces some consequent good – correction, training or re-socialization. Christopher Marshall argues that punishment is justified only if “the intention of the punishment is to reclaim the offender, restore relationships, and bring healing to the victim.” Such criticisms all assume that punishment can only be justified on consequentialist grounds. But if divine command theory is the right ethical theory to adopt, there simply is no force to their objections. What one would need is either an argument that shows that there is a genuine contradiction produced by holding to a retributive theory of punishment and an orthodox conception of God’s attributes, or an argument that demonstrates that a consequentialist theory of ethics is correct and that divine command theory is false. As I have briefly argued, consequentialism does not accord well with the teaching of scripture and is arguably incoherent.
Another example of a theological objection comes from Eleanor Stump. Stump’s argument is an example of an attempt to show that retribution and some feature of God are genuinely incompatible. Stump argues that God is supposed to be a merciful God and it is because of his mercy that he forgives sin. But, if the retributive view of punishment is true, then God cannot forgive sin: “To forgive a debtor is to fail to exact all that is in justice due. But according to [the penal substitutionary view], God does exact every bit of the debt owed him by humans; he allows none of it to go unpaid.” Commonly, when we think of forgiving a debt, we think of failing to exact the debt from the one who owes it. However, if the penal substitutionary view of the atonement is true, then God succeeds in exacting the debt from Christ. Stump argues that though God has the right to punish, it is not necessary that he do so. Instead, punishment is justified since it restores the relationship between God and humankind. In reply, the retributivist can argue that forgiveness does not entail that a punishment is not exacted for sin. Rather, forgiveness is removal of guilt and the consequent punishment from a sinner. Forgiveness does not entail a lack of punishment, it merely entails a lack of punishment for the one who is forgiven. God can offer forgiveness to one person and punish another on behalf of that person without thereby thwarting his own mercy.
A third species of this kind of argument rests on the assumption of revelational priority. On this view, certain parts of revelation take priority over others. According to Denny Weaver, our knowledge of God is dependent upon our knowledge of Jesus as he is revealed in the gospels. Since the gospels reveal Jesus as a non-violent person, we should conclude that the divine nature is non-violent. Jesus took suffering and violence, but did not mete it out. Consequently, God does not mete out violence in the name of punishment:
The wrath and violence of God pose a problem only if one accepts the profession that God is fully revealed in Jesus, that what is revealed in Jesus is the very character of God… each person is fully God, so that nothing can reside in God that is not present in each person of the Trinity, and no person of the Trinity can have an essential divine characteristic that is not part of Godself. Following this expectation, if Jesus is nonviolent one ought not seek to discover or to justify violence as a characteristic of God.
Weaver’s argument is valid but unsound. Specifically, we have no reason to presume that Jesus is essentially nonviolent. If Jesus is nonviolent for 30 years, it does not follow that he is essentially nonviolent anymore that it follows from being asleep for eight hours that I have the essential property of being asleep. Furthermore, it is not clear that anyone would suggest that violence can be the property of a person. Violence is a description of a kind of action. Furthermore, actions are not states of being, but occur at times. In contrast, wrath may be considered the property of persons, some consistent disposition towards evil. However, if wrath is an essential property of God, it does not follow that God would always act violently. Thus, Weaver’s objection fails.
A second kind of objection relates both to retribution and to substitution. Even if one accepts a retributive view of justice, one might object to the idea that a person can avoid punishment if another person agrees to take the punishment on their behalf. David Lewis illustrates the objection:
Imagine that an offender has a devoted and innocent friend. The offender has been justly sentenced to be punished in his place. If the friend undergoes the punishment that the offender deserved, does that render it permissible (or even obligatory) to leave the offender unpunished? Is that any reason at all in favor of sparing the offender?
The objection rests on the eminently plausible second condition for just punishment – a punishment is just only if it is carried out against a person who has committed a sin. If we accept the condition then either there is an exception in the case of human culpability for actions committed against God, or God unjustly causes his Son harm and suffering. Indeed, some advocates of penal substitution support a view that entails the injustice of the punishment. For example, D. A. Carson writes:
It is the unjust punishment of the Servant in Isaiah 53 that is so remarkable. Forgiveness, restoration, salvation, reconciliation—all were possible not because sins have somehow been cancelled as if they never were, but because another bore them unjustly. But by this adverb ‘unjustly’ I mean that the person who bore them was just and did not deserve the punishment.
Ironically, Carson’s comments serve as an argument against penal substitution. Surely, God’s actions in punishing his Son cannot count as unjust actions on pain of showing that God is not just (Rom 9:4). Surely, that is not what Carson wants to say. His definition provides some clarity: To be unjustly punished is for someone to be undeserving of the punishment he receives. However, it is not necessarily unjust to punish someone who does not deserve punishment. By suggesting so, Carson concedes too much. In reply, the retributivist should argue that a penal substitutionary view of the atonement does not entail unjust punishment for an innocent party. To remove the charge of injustice from the punishment of an innocent party either some morally sufficient reason must be provided for punishing the innocent party in place of the guilty or the innocent party must become guilty of the sins of others and thereby deserve the punishment.
How should we supply either a morally sufficient reason for punishing an innocent party or imputing guilt to an innocent party? A possible solution relates to how we are to understand the following utterance: “Jesus Christ was punished for sins he did not commit.” According to K. G. Armstrong, we can understand such a statement in two different ways. The weak interpretation according to which the one punished is treated as if he was guilty entails a legal fiction. This view asserts that God is justified in punishing Jesus only if Jesus is treated as if Jesus has committed our sin. A legal fiction is warranted as long as there is some greater good achieved by its application, in this case, the redemption of God’s people from punishment for their sin. As David Peterson puts it, “the penalty inflicted by God’s justice and holiness is also a penalty inflicted by God’s love and mercy, for salvation and new life.”
Steven Porter provides an example of the weak interpretation. According to Porter, it is morally permissible to transfer punishment to an innocent party only if (i) the one punished takes the punishment voluntarily and (ii) the one meting out the punishment is exercising a retributive right for certain intrinsic goods. For Porter, those intrinsic goods are the treating of sin with adequate seriousness coupled with expressing the value of the person offended. Such a view is a species of expressionism according to which punishment is justified only if punishment expresses condemnation of an immoral act. The expressive good of a punishment is the expression of both the moral seriousness of sin and the value of the victim, combined with the treatment of sinners as morally responsible agents. According to Porter, though God does not have to punish a substitute, he does so for the greater good of expressing this kind of condemnation of sin. Therefore, God is morally justified in punishing a substitute.
It is not clear, however, that Porter’s solution evades the objection that his view entails a consequentialist view of ethics. Porter attempts to distinguish his view from the consequentialist account of punishment by suggesting that punishment of a substitute is justified by an intrinsic good, in this case, the good of God expressing condemnation of sin. However, an intrinsic good is some good in the action itself whereas an extrinsic good is some good entailed, or produced, by the action. The question is: is the expressive good intrinsic or extrinsic? An answer to this question is not easy to produce. On one hand, God expresses condemnation in the act of punishment and thus it is an intrinsic good. On the other hand, and perhaps more plausibly, God achieves an expression of his condemnation by means of punishment. If so, then the punishment of a substitute achieves an end and is justified by that end. In other words, expressivism entails a consequentialist theory of ethics. And if so, then the reply to the objection fails.
However, there is another way to think about imputation. On a strong interpretation, Jesus is treated as the one who is guilty of our sin. In other words, there is no legal fiction, but liability to punishment is imputed to Jesus. If this is the case, then Jesus genuinely deserves punishment. If so, then the objection fails. Though it is not easy to make a positive case for the strong interpretation from scripture, it does not appear to be inconsistent with it. Paul’s declaration that Jesus Christ is “made to be sin” (2 Cor 5:21) does not rule out that our guilt for sin is imputed to Christ. Such a view of imputation is not as strange as we might think. For example, we are used to other people representing us (in a representative democracy, for example). We are also inclined to hold representatives morally culpable for the sins of those they represent (an employer, for example). If so, then thinking of Christ as our representative would include imputing to him the moral responsibility for our sin thus justifying his punishment for our sin. If so, then the objection fails.
On either the strong or weak interpretation, the retributivist can respond by rejecting the assumption that punishment of Jesus is immoral because Jesus is innocent by suggesting that the doctrine of imputation renders Jesus as guilty (according to the strong interpretation) or God being justified in punishing Jesus because of a greater good (according to the weak interpretation).
I have argued that a penal substitutionary theory of the atonement rests on a retributive theory. Consequently, if punishment cannot be morally justified on retributive grounds, then holding to a penal substitutionary view of the atonement is more difficult. However, though a retributive theory of punishment is only weakly supported by common human intuitions about human merit, it garners strong support from scripture. The Bible supports the idea that it is part of the nature of God to punish sin on the basis that sin deserves punishment and if so, then retributive punishment is morally justified by God’s nature. Since, on a divine command theory of ethics, the moral status of any act is determined by God’s rules given in accordance with his nature and God determines that it is right and necessary that sin is punished, the retributive theory of punishment is true. Finally, I responded to both theological and ethical objections to retribution as a justification for punishment and concluded that those objections fail.
 I am aware that much of the debate over punishment relates to its definition and the proportionality question. This paper focuses on the moral justification question rather than either the question of definition or the ‘penalty-fixing’ problem. In this context, the penalty-fixing problem relates to what punishment is sufficient to pay for sin. For an overview of the different questions related to a defense of retributive punishment see K. G. Armstrong, “The Retributivist Hits Back.” The Philosophy of Punishment: A Collection of Papers. ed. H. B. Acton. (London: Macmillan, 1969), 540-55.
 Luis Pojman, The Death Penalty: For and Against (Lanham, MD: Rowman & Littlefield, 1998), 5.
 Augustine, City of God, Bk. XX, ch. 1.
 John Calvin, Institutes, II.XVI.6; II.XVII.3.
 Charles Hodge, Systematic Theology Volume 2 (New York: Scribner, 1872), 490.
 Hans Boersma, Violence, Hospitality, and the Cross: Reappropriating the Atonement Tradition (Grand Rapids: Baker Academic, 2004), 175.
 Christopher Marshall, Beyond Retribution: A New Testament Vision for Justice, Crime, and Punishment (Grand Rapids: Eerdmans, 2001), 69.
 See for example, arguments against the compatibility of consequentialism and Christianity in John Feinberg and Paul Feinberg, Ethics for a Brave New World (Wheaton: Crossway, 2010); Dennis Hollinger, Choosing the Good: Christian Ethics in a Complex World (Grand Rapids: Baker Academic, 2002), 35–36.
 Immanuel Kant, The Metaphysics of Morals trans. Mary Gregor (Cambridge: Cambridge University Press, 1996), 6.331–32.
 Herbert Morris, “Persons and Punishment,” in Punishment and Rehabilitation ed. Jeffrie Murphy (Belmont: Wadsworth, 1985), 27.
 Peter Schmiechen, Saving Power: Theories of Atonement and Forms of the Church (Grand Rapids: Eerdmans, 2005), 113–116.
 J. Denny Weaver, “Violence in Christian Theology,” Cross Currents 51, no. 2 (Summer 2001): 150–176.
 I don’t mean that non-biblical sources are opposed to biblical sources. I only mean that no appeal to the Bible is necessary for the justification of a given claim.
 Anselm, Cur Deus Homo, XII.
 As Bruce Reichenbach points out, there is a circularity to Anselm’s argument. On the face of it, Anselm asks us to accept that it is not right to pass over sin unpunished because sin must be punished. Bruce Reichenbach, “Healing Response” in The Nature of the Atonement, eds. James Beilby and Paul Eddy (Downers Grove: InterVarsity, 2006), 108.
 Anselm considers the relationship of justice to God’s nature at the end of Section XII. Anselm argues that though what is just is in accordance with God’s nature, it would be wrong to suggest that injustice could have been considered just if God had deemed it so. This may imply that Anselm’s arguments are from the nature of God and not from reason alone. However, since our knowledge of God’s nature in respect to the necessity of punishment are not known through general revelation, we can only know them if God has revealed them in scripture. Thus, Anselm’s arguments ultimately require an appeal to special revelation.
 Cur Deus Homo, XII.
 It is plausible that Anselm intended his arguments implicitly to appeal to a scriptural authority rather than to a purely rational source.
 Louis Pojman, “Merit: Why Do We Value It?” Journal of Social Philosophy, Vol 30 No. 1 (Spring 1999), 96.
 Bruce Waltke, Genesis (Grand Rapids: Zondervan, 2001), 158.
 Michael Wechsler, “Genesis,” in The Moody Bible Commentary eds. Michael Rydelnik and Michael Vanlaningham (Chicago: Moody, 2014), 61–62.
 Robert L. Thomas, “2 Thessalonians,” in The Expositor’s Bible Commentary, vol. 11 (Grand Rapids: Zondervan, 1978), 312–313.
 DAG, 72.
 DAG, 238.
 Robert L. Thomas, “2 Thessalonians,” 312–313.
 Schreiner, “Penal Substitution View”, 80.
 Stephen Travis, Christ and the Judgment of God: Limits of Divine Retribution in New Testament Thought (Peabody, Mass: Hendrickson, 2008), 78.
 Robert Mounce, The Book of Revelation in NICNT vol. 17 (Grand Rapids: Eerdmans, 1977), 295.
 Ibid., 296.
 Richard Bauckham, “Judgement in the Book of Revelation,” Ex Auditu 20 (2004): 3.
 Schüssler Fiorenza, Revelation: A Vision of a Just World (Minneapolis: Fortress, 1991), 64.
 At this point I am talking about epistemic justification, the question of how one grounds a claim in some reason. Pojman’s epistemic justification is a self-evident truth, but a theological epistemic justification is found in the nature of God. Ethical theories depend on an epistemic justification. Pojman’s view is usually classed as an inuitionist view and the view I present is usually classed as a definist view. See John Feinberg and Paul Feinberg, Ethics for a Brave New World (Wheaton: Crossway, 2010), 50–51.
 Peter Schmiechen, Saving Power: Theories of Atonement and Forms of the Church (Grand Rapids: Eerdmans, 2005), 110.
 Stephen Finlan, Options on Atonement in Christian Thought (Collegeville, Minn: Liturgical Press, 2007), 97.
 Christopher Marshall, Beyond Retribution: A New Testament Vision for Justice, Crime, and Punishment (Grand Rapids: Eerdmans, 2001), 139.
 The argument for the incoherence of consequentialism usually takes the following form: Action A’s moral status is determined exhaustively by the consequences of A. If A is a grievous evil (rape or cold-blooded murder, for example) and produces great goods, it is, according to consequentialism, not only morally justified, but a good action. But this appears to lead to the conclusion that A is both a grievous evil and a good action (contradiction). Therefore, consequentialism is incoherent (and, therefore, false).
 Eleonore Stump, “Atonement According to Aquinas,” in Oxford Philosophical Readings in Philosophical Theology Volume I: Trinity, Incarnation, Atonement, ed. Michael Rea (New York: Oxford University Press, 2009), 268.
 Robert Yarbrough surveys Old and New Testament occurrences of ‘forgiveness’ and related terms and does not mention any definition including anything like ‘failure to extract payment’ R. W. Yarbrough, “Forgiveness and Reconciliation,” in New Dictionary of Biblical Theology, eds. Alexander, Rosner, Carson, and Goldsworthy, (Downers Grove: IVP Academic, 2000), 498.
 J. Denny Weaver, The Nonviolent God (Grand Rapids: Eerdmans, 2013), 48.
 David Lewis, “Do We Believe in Penal Substitution?” in Oxford Readings in Philosophical Theology, Volume I: Trinity, Incarnation, Atonement ed. Michael Rea (New York: Oxford University Press, 2009), 308.
 D. A. Carson, “Atonement in Romans 3:21-26.” The Glory of the Atonement: Biblical. Theological & Practical Perspectives: Essays in Honor of Roger Nicole. eds. Roger R. Nicole, Frank A. James, and Charles E. Hill (Downers Grove, IL: InterVarsity, 2004), 134.
 Armstrong’s example utterance is ‘he was punished for something he did not do’ and Armstrong is not discussing the atonement. Consequently, he reaches different conclusions. His distinction, however, is highly relevant to the objection I am discussing. Armstrong, “The Retributivist Hits Back”, 479–480.
 David Peterson, “Atonement in the New Testament,” in Where Wrath and Mercy Meet: Proclaiming the Atonement Today, ed. David Peterson (Waynesboro, GA: Paternoster, 2001), 38. Emphasis mine.
 Steven Porter, “Swinburnian Atonement and the Doctrine of Penal Substitution,” in Oxford Philosophical Readings in Philosophical Theology Volume I: Trinity, Incarnation, Atonement, ed. Michael Rea (New York: Oxford University Press, 2009), 320–326.
 Joel Feinberg, “The Expressivist Function of Punishment,” The Monist, 49.3 (1965): 397–493.
 This problem has been treated in light of the greater good defense in response to the problem of evil. Some philosophers argue that the greater good defense entails a consequentialist ethic. However, other philosophers argue that the greater good in view is an intrinsic good and so does not entail a consequentialist ethic. Richard Swinburne, for example, offers a greater good defense but calls himself an anti-consequentialist: Richard Swinburne, Providence and the Problem of Evil (New York: Oxford University Press, 1998), 254. David McNaughton argues that Swinburne’s view entails a consequentialist ethic. He calls it ‘Swinburne’s Lapse’: David McNaughton, “Is God (almost) a Consequentialist?: Swinburne’s moral theory.” Religious Studies 38, no. 3 (2002): 265-281.