Christ, Our Indemnitor

I am grateful to Aaron Davis and Danielle Jansen for organizing this symposium on my book Atonement and the Death of Christ (ADC) and to the editors of Philosophia Christi for publishing our exchange. I think that ADC may well prove to be theologically the most important work I have written, and so I am gratified that it has become the subject of discussion and evaluation.

In ADC I articulate and defend a multi-faceted, if incomplete, atonement theory. There may be additional facets that should be added to form a full-orbed atonement theory, but the facets I deal with, including penal substitution, satisfaction of divine justice, Christus Victor, Christ’s representation (or our participation in Christ), and the moral influence of Christ’s passion are essential to any biblically adequate atonement theory in view of the many biblical atonement motifs they capture, such as sacrifice, the Suffering Servant of Isaiah 53, God’s justice, our redemption, and Christ’s representation. Thus, my theory already looks far beyond what Oliver Crisp calls “the narrow confines of penal substitution.”

I do think that penal substitution is the “table” or central facet anchoring the other facets of the jewel that is a biblically adequate theory of the atonement, so it is not surprising that most of our symposiasts should largely pass over the other facets I deal with to focus on penal substitution. I want to note simply in passing for the sake of those who have not read ADC that I do not neglect but seek to justify and defend these additional facets as well.

Focusing, then, on penal substitution, I want in this response to address four questions concerning penal substitution: (1) Is penal substitution biblical? (2) Is penal substitution coherent? (2) Is penal substitution just? And (4) Is penal substitution satisfactory?

(1) Is Penal Substitution Biblical?

I was initially disappointed when I learned that our symposiasts included no biblical theologians, though upon reflection I saw the concentration on philosophical theology to be entirely appropriate for a philosophical journal. Still, as a Christian, inspired Scripture is for me, as for all the great confessions of Christendom, whether Catholic, Coptic, Orthodox, or Protestant, the ultimate and infallible authority for Christian doctrine, including the doctrine of the atonement. Perhaps I am what Joshua Thurow calls a “biblical Moorean”—which is, I suspect, just a philosophical sobriquet for “fundamentalist”—but so be it: I will privilege what I take to be the clear teaching of Scripture over our moral intuitions any day, especially since the latter reflect relations between human persons rather than between divine and human persons and require nuancing in any case. Indeed, Thurow is arguably more Moorean about his moral intuitions than I am about penal substitution, since I am open to being shown, not that Scripture teaches false doctrine, but that I have misinterpreted Scriptural teaching. Settling that question requires the sort of close engagement with the biblical text that characterizes the first part of ADC.

We can be grateful that Jonathan Rutledge does engage with the biblical material in his response, though he does not challenge the biblical basis of penal substitution. The doctrine of penal substitution is grounded, primarily, in Isaiah’s Suffering Servant of the Lord in Isaiah 53 and the widespread deployment of this motif in the New Testament (NT) and, secondarily, in the Levitical animal sacrifices, whereby expiation/propitiation was achieved by the animal’s suffering the fate that was the offerer’s just desert as the punishment for his sin. Rutledge passes over these texts without comment. If I understand him correctly, then, his position seems to be that while penal substitution is sufficiently “tethered” to the biblical text, my theory will need to be adjusted in light of needed revision of other motifs that are “insufficiently tethered” to the text.

Rutledge names three such atonement motifs that I work with: (i) divine justice, (ii) propitiation and Passover, and (iii) sacrifice. Rutledge does not deny that these are, indeed, biblical atonement motifs, but he thinks that I have misinterpreted them, so that my account requires “further development” in light of the correct interpretation of these motifs. Now I agree with Rutledge that it would, indeed, be interesting to see “what shape a penal substitutionary model of atonement should take” once these revisions are made. Prima facie, it seems to me that the revisions could be fairly easily accommodated. Taking God’s justice to be purely consequentialist, as favored by Rutledge, would still allow us to say that divine justice is essential to God and that penal substitution aims to satisfy divine justice. Indeed, as noted in ADC, adopting consequentialism would allow one to avoid completely the objection to the justice of penal substitution, since, notoriously, on a consequentialist theory of justice punishment of the innocent may be justified in light of its extrinsic benefits. Taking Passover to be non-propitiatory would be unproblematic, since the propitiatory function of the Levitical sacrifices remains. Construing Christ’s death, not as the atoning event, but as the initiatory event in the process of atonement, completed by Christ’s presenting his sacrifice in heaven, is more jolting but still leaves Christ’s sin-bearing death as an indispensable element of his atonement.

But I am uninterested in pursuing these further developments, since I do not think that Rutledge has made his case that such revisions are necessary in light of the biblical text.

(i) Divine Justice. Rutledge denies that “God’s justice includes an essentially retributive element.” I find this denial incredible. Certainly, it is easy to point to biblical passages where God’s punishment can be interpreted consequentially as aimed at the repentance and restoration of the wrongdoers punished. But those passages do not exclude a retributive element to God’s justice as well. The case of Nineveh is not in any event a good example for Rutledge’s view, as he imagines. For had the Ninevites not repented, they would, much like the earlier Canaanites, have been destroyed, not restored. God did not punish the Ninevites because they repented; but had they not done so, God’s punishment would have been retributive, not restorative, for most of them! Indeed, Rutledge’s discussion of the case of Nineveh exhibits a major misunderstanding of retributive justice. Rutledge seems to think that God’s justice’s being retributive implies that punishment is inevitable, which is mistaken (ADC, pp. 179-82). It is therefore misconceived to try to play off God’s “merciful character” against his retributive justice and to portray Jonah as the retributivist and God as the consequentialist in the story. They are both retributivists, but God, unlike Jonah, is also merciful.

The words ekdikēsis and antapodoma just are NT words for retribution, as Rutledge admits. He correctly reminds us that what counts is not the lexical meaning of words but their contextual use. But Rutledge gives no good reason for taking NT use of these words in any non-retributive sense. He tries to soften the import of Rom 12.19 by appealing to the case of Nineveh. But the background to Paul’s statement is not Jonah but Deut 32.35, a fearsome passage on divine retribution; and at best what the case of Nineveh shows is that we may leave retributive justice to God, since he knows whether his retribution will be necessary, while we do not. Paul, like the author of Hebrews, explicitly says that the wicked deserve punishment (Rom 1.32; Heb 10.29), which entails that God’s justice is in some measure retributive.

The fact that divine punishment is ultimately eschatological (Rom 2.5) underlines its retributive nature, for then it is too late for restoration. Rutledge’s fanciful suggestion that “eschatological punishment could be construed as a means to preventing them from becoming more estranged from God than they already are” fails to commend itself as a plausible interpretation of the biblical descriptions of the grim fate of the damned. Reflecting on Paul’s statement that the impenitent are “storing up wrath” for themselves on the day of judgement, Rutledge asks, “Who are those who have been storing up wrath against themselves until the day of judgment in Romans 2? All people, for all have sinned.” No! Those who are united with Christ have escaped the wrath of God; it is only the impenitent who await judgement and punishment.

(ii) Propitiation and Passover. Rutledge then disputes my claim that the initial Passover sacrifice was propitiatory. I agree that it was not so clearly propitiatory as were the later Levitical sacrifices, about which Rutledge has little to say; but it did serve to avert God’s judgement and can in that sense be called propitiatory. Israel at this point in its history was as yet unredeemed and, hence, alienated from God and lost in sin, so God’s judgement had to be averted. Rutledge’s claim that “Passover should be seen primarily as a defeat of the angel of death,” whom he identifies as Satan, is not only anachronistic but fails to do justice to the text, which does not speak of an angel at all. Rather God says, “I will pass through the land of Egypt that night, and I will smite all the first-born in the land of Egypt, both man and beast; and on all the gods of Egypt I will execute judgments: I am the Lord” (Ex 12.12). Rutledge also points to Heb 2.14, which interprets Christ’s death as destroying “him who has the power of death, that is, the devil.” Right; this statement shows that Christus Victor must also be a facet of an adequate atonement theory; but it does nothing to negate the interpretation of Jesus’ death as the fulfillment of the provisional animal sacrifices, which “can never take way sins” (Heb 10.11). Christ “has appeared once for all at the end of the age to remove sin by the sacrifice of himself” (Heb 9.26). Need I point out that Rutledge’s assertion “Christ’s Passover death is . . . an act of obedience that shields Jesus—and those he represents—from an ultimate death and the devil, for it corrects Adam’s original sin” is wholesale eisegesis, untethered to the biblical text?

I heartily concur with Rutledge that God’s deciding to give Israel the Passover ritual to follow “is certainly not incoherent, but for modern sensibilities, it is bound to seem a bit strange.” As I emphasize in ADC, the whole Levitical system of blood sacrifices is bizarre for modern sensibilities; but that requires us to shed those sensibilities and to enter into the horizon of these ancient peoples if we are to understand these texts and the practices they prescribe.

(iii) Christ’s Death as an Atoning Sacrifice. Finally, Rutledge argues that Christ’s sacrifice should not be identified with the cross, but “takes place after the resurrection and ascension to the heavenly temple.” On the basis of the book of Hebrews, he reasons as follows: “if the resurrection is indeed the pre-requisite to becoming a high priest, then Jesus’ sacrifice, which he offers as the high priest, clearly is subsequent to his death on the cross, not simultaneous with it.” This is a non-sequitur. A Levitical animal sacrifice took place prior to the priest’s presentation of its blood in the Most Holy Place. What takes place in heaven, then, is at most the presentation of Christ’s blood, not his being sacrificed. This distinction was nicely captured by Protestant scholastics in opposition to Faustus Socinus in their distinguishing within Christ’s priestly office (officium sacerdotale) his rendering satisfaction (satisfactio) and his intercession (intercessio). Whether the author of Hebrews thought of Christ’s assuming his priesthood at his incarnation or at his ascension is ambiguous (Heb 1.1-13; 2.14-17; 3.14-15; 5.5-6; 9.26; 10.5). But the author of Hebrews thought of Christ’s death as redemptive: “he is the mediator of a new covenant, . . . since a death has occurred which redeems them from the transgressions under the first covenant” (Heb 9:15). Although Rutledge can cite David Mofitt in support of his view, the majority view of Hebrews is that the author describes Christ’s death and ascension as the analogues to the killing of a sacrificial animal and the presentation of its blood in the Most Holy Place. One can appreciate Moffit’s emphasis on Jesus’ resurrection as an entailment of his ascension to heaven without denying the atoning significance of Christ’s death.

In any case Hebrews is but a part of the NT witness. As pointed out in ADC, the substitutionary, punitive nature of Christ’s death, already foreshadowed in the Levitical sacrifices, comes to the fore in the NT authors’ identification of Christ with Isaiah’s Servant of the Lord, who “bore our sins in his body on the tree” (1 Pet 2.24). The common Hebrew idiom of bearing sin/iniquity means to be liable to punishment or to endure punishment. This Christ did in our place and on our behalf. Paul believed that it was through Christ’s death on the cross that we obtained reconciliation with God: “We were reconciled to God through the death of his Son” (Rom 5.10; cf. Col 1.19-22). Thus, “the cross” came to be a metaphor epitomizing the Gospel message, such that Paul could call the Gospel “the word of the cross” (1 Cor 1.18), reminding his Corinthian converts that “I decided to know nothing among you except Jesus Christ and him crucified” (1 Cor 2.2). Hence, Paul would glory in nothing “except the cross of our Lord Jesus Christ” (Gal 6.14).

(2) Is Penal Substitution Coherent?

In ADC I interact at length with attempts to show that Christ’s substitutionary punishment is incoherent. We may be thankful that none of our symposiasts endorses such objections; indeed, some explicitly affirm the coherence of Christ’s penal substitution. Still, Danielle Jansen and Oliver Crisp, while affirming the coherence of holding that Christ endured the suffering that we deserved as the punishment for our sins, also affirm that according to the standard definition of legal punishment, “punishment must be upon the guilty for the act to be considered punishment” and that it is a “category mistake to speak of the ‘punishment of the innocent’,” which comes dangerously close to endorsing a so-called “definitional stop” that rules out punishment of the innocent by definition (ADC, p. 154), thereby fostering an incoherence objection against Christ’s being punished for our sins. They are, however, mistaken in this, as an examination of Jansen’s own definition of punishment, borrowed from David Boonin, reveals. Jansen herself notes, “Some legal theorists claim that an innocent may be punished (that is, an innocent may suffer punishment unjustly),” though she chooses to “ignore that complication here.” If, however, it is possible to punish the innocent, as almost all theorists agree, then there is no incoherence in Christ’s being punished, and the question becomes whether his punishment is just.

(3) Is Penal Substitution Just?

Joshua Thurow’s paper is devoted to pressing the objection to penal substitution based on the immorality of punishing an innocent person for another’s wrongdoing. It is intriguing that Thurow backs away from the typical, general claim that an innocent third party cannot be justly punished for another’s sin to the more modest claim that an innocent person cannot be justly punished for another’s “deep sins,” like murder. This is a noteworthy re-trenchment on the critic’s part, for it tacitly allows that penal substitution may in very many cases be just. It is easy for the critic simply to cite examples like punishing Johnny’s mother Beth for a murder he committed as powerful intuition pumps that such punishment is unjust. But it is much more difficult to give a principled account of why substitutionary punishment is acceptable in some cases but not in others. If God can justly punish Christ for a wide array of human sins, why not for all of them? It seems to be all or nothing here.

In ADC I offer a number of possible replies to the injustice objection.

(i) Penal Substitution without Punishment

Thurow notwithstanding, it seems to me that the injustice objection fails for what is often called a “penal consequences” view, namely, the view that Christ is not punished for our sins, but suffers the harsh treatment that would have been our punishment had it been inflicted on us. Christian philosophers have learned from discussion of the problem of evil that God may justly inflict horrible suffering upon innocent people. That is his sovereign prerogative. So if God chooses to regard some such instance of suffering as sufficient to atone for the sins of others, as in the case of Christ, how is that unjust? After all, Christ died long before the sins of billions of people were even committed; so why is God unjust to regard his suffering as sufficient for atoning for those sins? As Jansen rightly recognizes, the problem for the penal consequences view is not the justice of penal substitution but its satisfactoriness.

Now in light of misunderstandings in the literature (and even among our symposiasts), I want to say clearly that while I defend the penal consequences view against the accusation of injustice, I myself do not advocate such a view but believe Christ to have been punished for our sins. But I want to be open to the viability of various options.

(ii) Metaethical Contextualization

A crucial question for those pressing the injustice objection is, who or what determines what is just or unjust? Answering this question requires an account of our moral obligations and prohibitions, and I can think of no more plausible source of our objective moral duties than God’s commands. Given a Divine Command Theory (DCT) of ethics, the injustice objection has difficulty even getting off the ground. Since, presumably, God does not issue commands to himself, he literally has no moral duties to fulfill. Even if God has established a system of justice among human beings that forbids the punishment of the innocent (which would explain Thurow’s deeply held moral intuitions), he himself is not so forbidden. If God wills to take on human nature and give his own life as a sacrificial offering for sin, who is to upbraid him? He is free to do so as long as it is consistent with his nature. And what could be more consistent with our God’s gracious nature than that he should condescend to take on our frail and fallen humanity and give his life to satisfy the demands of his own justice? Seen in this meta-ethical context, the injustice objection becomes almost ludicrous.

In reply, Thurow says that this response is not terribly effective dialectically, since DCT is widely disputed. Never mind that secular ethicists must, of course, dispute the theory, despite its wide reception among theists. More importantly, Thurow’s reply stands things on their head. The point is that given DCT God cannot coherently be accused of acting unjustly. So to defeat this response critics are going to have to defeat DCT itself. Does Thurow think he can do so? Then let him have a go at Robert Adams, William Alston, Philip Quinn, David Baggett, and many other eminent meta-ethicists. Good luck! So long as DCT is a plausible meta-ethical account of moral duty, the objection falters.

Second, Thurow points out that a plausible DCT will not be voluntaristic but will hold that God’s commands must be consistent with his nature (a view Adams calls theistic Platonism). I concur wholeheartedly and shall have more to say on this head under our next point. But Thurow says nothing further to show that God’s essential goodness is inconsistent with Christ’s being a penal substitute. On the contrary, as I say, it exalts God’s goodness.

Third, Thurow charges that DCT eliminates justice as a motivation for Christ’s atoning death. God could have willed anything to satisfy the demands of divine justice. So why did God choose Jesus’ crucifixion as the means of atonement? This reply raises the question of so-called acceptation. Let us assume that retributive justice is essential to God’s goodness. John Duns Scotus suggested that God might have accepted any sacrifice he pleased as satisfactory for the demands of his retributive justice. Defenders of penal substitution have not been sympathetic to acceptation accounts. The objector to penal substitution would find a sympathetic ear among penal substitution theorists if he affirmed that retributive justice, as we know and understand it, is essential to God’s nature and so could not be satisfied by mere animal sacrifices. But retributive theories of justice also require that the punishments be proportionate to the crime if justice is to be satisfied, what Oliver Crisp calls a “proportionality understanding” of the atonement, according to which God must accept an act that has an objective moral value at least proportional to the demerit of the trespass it atones for. Christ’s death has unlimited value, more than sufficient to outweigh the demerit of humanity’s sins. The question remains why God chose Christ’s crucifixion to satisfy the demands of his justice. I can think of no better answer than the one given by Abelard and Grotius, that the passion and crucifixion of Christ exhibit in an unparalleled manner both the holiness and the love of God, a display of those attributes which has proved tremendously attractive in drawing billions of people to faith in Christ. Here we see the importance of the moral influence facet of a full-orbed atonement theory.

(iii) Retributive Justice and the Divine Nature

If we suppose that retributive justice is essential to the divine nature, the question arises as to what retributive justice is. As I explain in ADC, theorists have distinguished between negative retributivism, which holds that the innocent should not be punished because they do not deserve it, and positive retributivism, which holds that the guilty should be punished because they deserve it. What distinguishes retributivism as a theory of justice is the positive thesis that punishment of the guilty is an intrinsic good because the guilty deserve it. God is a positive retributivist “who will by no means clear the guilty” (Ex 34.7). But the penal substitution theorist may maintain that God is only qualifiedly a negative retributivist, since even if he has prohibited human beings from punishing innocent persons (Deut 24.16), and even if he is too good to himself punish innocent human persons (Gen 18.25) (which, again, explains Thurow’s deep-seated moral intuitions), still he reserves the prerogative to punish an innocent divine person, namely, Christ, in the place of the guilty. This extraordinary exception is a result of his goodness, not a defect in his justice.

Thurow thinks that this response is unavailing because it just builds right into DCT that morally God can punish the innocent if he wills it; but that contradicts our moral intuitions that an essentially good being would not punish the innocent. Not at all! This response is willing to concede that an essentially good being would not only prohibit human persons from punishing the innocent but would not himself punish innocent human persons, thereby accounting for our moral intuitions. But God would be able to punish an innocent divine person, so that he is only qualifiedly a negative retributivist. We have no insight into the divine nature that would allow us confidently to preclude this possibility.

(iv) Prima Facie vs. Ultima Facie Justification

The prima facie demands of retributive justice, whether positive or negative, can be overridden by weightier moral concerns. Even the staunchest of contemporary retributivists, Michael Moore, recognizes that the demands of retributive justice are prima facie demands that can be and are overridden in specific cases. In the extreme case where one must punish an innocent person or else the world will be totally destroyed, says Moore, one should punish the innocent person. The penal substitution theorist could similarly claim that God, by waiving the prima facie demands of negative retributive justice and punishing Christ for our sins, has mercifully saved the world from total destruction and was therefore acting compatibly with moral goodness.

Thurow replies that “It’s not clear to me that the duty to not punish the innocent is prima facie.” Ah, good! We finally see some flexibility regarding his moral intuitions. That should make him less confident that God cannot punish Christ in our place.

But Thurow goes on to say that if my response is to succeed, I must show both that the great good of redeeming humanity outweighs the deep badness of unjustly punishing someone and also that there was no other better way of redeeming humanity. This puts the shoe on the wrong foot. It is the objector who must show that the great good of redeeming humanity does not outweigh the alleged badness of God’s punishing, not someone, but Christ in our place. That burden of proof strikes me as far too heavy for the objector to bear. Neither can we say with any confidence that God’s choice of Christ’s penal substitution was not as good a way as any of saving mankind. True, Anselm suggests compensation rather than punishment as the means of satisfying divine justice; but Anselm’s theory is insufficiently tethered to the biblical text to be acceptable. In any case, how can Thurow know that if God had chosen some other means, there would have resulted a more optimal balance between saved and lost? It is obvious that we are now in the realm of speculation rather than solid argument.

(v) Punishment and the Imputation of Sins

We finally come to what I consider to be the heart of the matter. In virtue of the imputation of our sins to Christ, Christ was legally guilty before God and therefore legally liable to punishment. Imputation is a biblically based doctrine that should be part of any adequate atonement theory.

Thurow rejects imputation as absurd: “Imputation seems impossible, full stop.” This bold assertion not only fails to take account of God’s unique role as Legislator, Judge, and Ruler, but is also flatly mistaken. Imputation of wrongdoing and guilt is not merely possible; it is actual. Imputation permeates the Anglo-American system of justice. It takes place constantly across a wide array of crimes. And it is regarded as just.

In ADC I give two illustrations of imputation in our justice system. The first involves the use of a legal fiction by the court for the sake of a particular action. God, as supreme Judge, can adopt the legal fiction that Christ actually committed our sins, so as to find him legally liable to punishment. Thurow objects that my two examples of legal fictions are not sufficiently analogous to the imputation of our sins to Christ to be relevant. This objection is misconceived. One need not show similarities between specific cases in which the court adopts a legal fiction for there to be a common principle employed across a diversity of cases. Moreover, the case of ship personification in Admiralty Law is striking precisely because it does involve the imputation of guilt to an entity other than the person who committed the wrong. The fiction of ship personification was adopted for the express purpose of imputing wrongdoing to the ship herself. Thurow is right that the motivation in so doing is different than God’s motivation in adopting the legal fiction that Christ committed our sins. The ship is found guilty in order to right a wrong, whereas Christ is found guilty for the sake of our salvation. Nonetheless, the legal principle employed remains the same.

I am grateful for Crisp’s detailed interaction with this alternative. My earlier criticism of Crisp was that he takes the legal fiction employed by God to be treating Christ “as if he were guilty.” I am not aware of any penal substitution theorist who has held such a thing, and it is not the alternative I advocate. Rather the legal fiction that God employs is that Christ has committed the sins for which we are condemned. Therefore, he is legally guilty, as Calvin claimed.

Now Crisp admits that Christ can be held to be legally liable to punishment for our sins by means of a legal fiction, but he objects that Christ cannot be found guilty of our imputed sins because guilt just is the property of having committed the relevant crime. But my contention is that this equation is profoundly mistaken, both philosophically and theologically. If guilt just is the property of having committed the crime, then guilt can never be removed. This is philosophically problematic because according to a retributive theory of justice, the guilty deserve punishment. In that case a criminal who has fully served his sentence or who has been granted a full executive pardon is still guilty and therefore still deserves punishment, which is absurd. I propose that we can explain Crisp’s intuitions by “taking tense seriously.” In ADC I cite various court cases that do precisely that. Rather than say of the old man released from prison, “he is the person guilty of the murder 25 years ago,” we should say of him, “he is the person who was guilty of the murder 25 years ago.”

Theologically, it is deeply objectionable to equate guilt with the fact of having committed the sin, for that makes guilt ineradicable, despite God’s pardon. That contradicts the teaching of the Bible that God expunges our guilt and constitutes us perfectly righteous in Christ. I frankly shudder when Crisp intones, “There is something indelible about sin. It stains the soul. . . . even when the liability to punishment is removed, the stain remains like a birthmark or tattoo. . . . Indeed, it is impossible to remove it once the sin in question has been committed precisely because I am the one personally responsible for having carried it out in the first place.” On this view not even God can remove our guilt.

These objectionable consequences should lead us to reflect upon the nature of guilt itself. If guilt is not the property of having committed the crime, it is tempting to think that guilt is culpability (or blameworthiness) for the crime. But that cannot be right either because people are found guilty all the time for crimes of strict liability, which involve no culpability. It seems to me, therefore, that the best definition of guilt is simply liability to punishment. When a person is found guilty in a court of law, he is found to be liable to punishment. Once a person has served his sentence or received an executive pardon, he is no longer liable to punishment and therefore no longer guilty. I was more than mildly surprised when Crisp invokes “a longstanding theological distinction between a liability to guilt and a liability to punishment,” for this is a Roman Catholic distinction, firmly rejected by Reformed theologians like Turretin, that was employed in the Canons of the Council of Trent to justify the church’s imposing temporal punishments upon pentinent Christians even though their liability to guilt had been expunged by a divine pardon. As Turretin explained, once a person’s guilt has been expiated, it is impossible for a liability to punishment to remain.

Although I find this understanding of guilt as liability to punishment to be a plausible account of the matter, it can, nonetheless, remain an open question. For what is crucial is simply that guilt is not the property of having committed a crime and that, therefore, Christ can be found guilty in virtue of the imputation of our sins to him.

My second illustration of imputation in our justice system involves the finding of vicarious liability in accord with the principle of respondeat superior. On the contemporary scene this principle has given rise to a widespread and largely uncontroversial finding of the vicarious liability of employers. Cases typically involve employers’ being held liable for the illegal sale of items by employees but may also include “deep sins” like assault and battery, fraud, manslaughter, and so on. It is futile to assert that this is impossible, for it happens – all the time. Thurow again tries to show that vicarious liability is disanalogous to the imputation of our sins to Christ. First, he says, a finding of vicarious liability requires a special relationship between the superior and the subordinate, so it cannot serve as a model for how guilt in general could be imputed to others. Right, nor is it intended to serve as a general model, but as a specific example of imputation. Ironically, however, as I note in ADC (p. 201), Christ, pace Thurow, does in fact stand in relation to us in precisely that special relation required by the courts for a finding of vicarious liability. Second, he says, vicarious liability is not subject to prerogative, but imputation is a divine prerogative, something God chooses to do. In fact, however, pursuing vicarious liability is a prerogative of the state that it might refuse (ADC, p. 202). But never mind. Any disanalogy here is incidental. God’s freely choosing to find Christ vicariously liable for our sins exploits the same mechanism used by the court. So much for the denial of the possibility of imputation!

Now Jansen asserts that “legal precedence for the transference of liability for murder in the case of a ‘liable superior’ or corporate entity is without precedence.” I take this statement to mean that there is no legal precedent for a finding of vicarious liability for the crime of murder.

But as Aaron Davis shows (and as Jansen acknowledges in a footnote), in the case of felony murder we do have a dramatic example of someone’s being found vicariously liable for murder. For example, the driver of a getaway car waiting at the curb while his accomplices rob a liquor store can be convicted of murder if one of his accomplices should, whether intentionally or accidentally, kill the shopkeeper in the course of the robbery. Indeed, even if the victim dies later as a result of the robbery, the driver may be charged with felony murder. He may even be sentenced to death for the crime, even though he did not himself kill anyone. “Therefore,” Davis concludes, “we have an example of legal guilt as replicable in circumstances so serious as murder and punishable so severely as by death.”

The reason I overlooked the example of felony murder was because I was focused on applications of respondeat superior, which principle is not applicable in cases of felony murder. I have since come to appreciate that imputation is even more pervasive in our justice system than I imagined, playing a role as well in the law of agency, the worker’s compensation system, fidelity, performance, and related bonds/title insurance, and the law of indemnity (I owe these examples to law professor Melvin Cockrell). I also overlooked felony murder because I was looking for the imputation of guilt to a party who was entirely innocent, which is obviously not the case with felony murder. But what Davis helps us to see is that the minor accomplice found guilty of felony murder is, apart from vicarious liability, innocent of the crime of murder. He can be found guilty of murder only because the guilt of the killer is imputed to him.

God forbid, however, that we should draw an analogy between a felony murderer and Christ, as Davis essays to do! The fact that Christ fulfills the legal requirements for standing in a relation to us of respondeat superior is a happy coincidence. What is at issue, rather, is whether we have any experience of the imputation of the guilt of a wrongdoer to a third party who did not commit the wrong. Remember: we are not trying to construct our theology on the model of the American justice system. Rather, we are responding to an alleged defeater of the biblical doctrine of the imputation of our sin and guilt to Christ. In Davis’ words, we can accept legal analogies for the atonement “as truly and merely analogies: they do not describe what God must do to redeem humanity at all.” Those who blithely assert that we have no analogies of imputation, even for the most serious of crimes, are plainly wrong.

Davis raises two concerns about appealing to vicarious liability for felony murder as analogous to the imputation of sins. First, “defenders of retributive theories of punishment are among the severest critics of strict liability offenses in criminal law.” There is less to this statement than meets the eye. Of course, the critics are retributivists, for consequentialists cannot object to strict liability, since consequentialism allows that the innocent may be justly found to be guilty. Moreover, the statement says nothing about the percentage of retributivists who criticize strict liability. The fact is that there are many thousands of statutory offenses involving elements of strict liability (ADC, p. 157). Davis’ cited statement: “if criminal punishment is only justified if the criminal is deserving of it, and if his deserving punishment depends on his having acted culpably, then criminal punishment should never . . . be a matter of strict liability,” is trivial, for in cases of vicarious liability the criminal deserves punishment without his having acted culpably with respect to the crime.

Davis also fears that my allowing that God may not be an unqualified negative retributivist will put me outside the mainstream of retributivists because retributivists usually accept both positive and negative retributivism together. There need be no fear, however; for allowing God the prerogative of punishing an innocent divine person constitutes no exception to the union of positive and negative retributive justice among human persons, which is what legal theorists are concerned with. Moreover, in the dialectical development of my defense of penal substitution, my final position does not qualify God’s negative retributivism: rather, because our sins are imputed to Christ, Christ is not an innocent person, so that in punishing Christ God in no way compromises negative retributive justice.

Finally, Davis worries that lower courts may have interpreted the Supreme Court opinions in an overly mechanical way, assuming that “anyone causing death in a predicate felony is death-eligible, regardless of culpability.” But Davis has conflated the finding of guilt with the sentencing of those convicted. In Cabana v. Bullock (1986) the U.S. Supreme Court clarified that the culpability required for a death sentence is not required for a finding of guilt for a felony murder accomplice. Justice Byron White explained that the finding of culpability need not even be made by the jury but can be made later by the trial judge in a sentencing hearing. Neither Enmund v. Florida nor Tison v. Arizona challenged the legality of imputing murder to the non-killer accomplice in a felony murder; rather the entire debate was about whether such an accomplice could be sentenced to death given the Eighth Amendment’s prohibition of “cruel and unusual punishment.” The controlling assumption is that since capital punishment is the maximum punishment the court can prescribe, it would be inappropriate if the actual killer did not receive a worse punishment than the mere accomplice. So Enmund required that in order to be sentenced to death for felony murder, the criminal had to have killed, attempted to kill, or intended to kill. Tison relaxed the requirement to permit the execution of major participants in a felony murder who had acted with reckless indifference to human life. The “overly mechanical interpretation” alleged by Binder et al. has to do entirely with lower courts’ relaxing the conditions required for the imposition of the death penalty. It has nothing to do with the imputation of felony murder to an accomplice who does not himself kill anyone. Seeing no need, then, for Davis’ suggested final options, I shall leave the pursuit of those options to others.

Thurow has a second general objection to the justice of penal substitution, namely, it would be a roadblock to people’s coming to salvation in view of its moral repugnance. I find this objection utterly implausible. I bet that the doctrines of the Trinity and incarnation have been far greater obstacles to faith than penal substitution, yet God revealed them. Can Thurow cite any sociological studies showing that people refuse to come to faith in Christ due to their revulsion at penal substitution? “When properly understood,” as we have seen, penal substitution does not “involve something humans regard as deeply morally wrong.” And how many of those who are mistakenly revolted by penal substitution allow it to be a roadblock to union with God? Do they rather not, like Thurow, just reject penal substitution than refuse to come to God?  I suspect that Abelard and Grotius are more likely correct that the message that Christ bore the penalty for our sins has inspired humility, contrition, and love in untold millions of those who hear it.

(4) Is Penal Substitution Satisfactory?

Jansen devotes her paper to this interesting question. While conceding the “coherence and justification” of penal substitution, she disputes the satisfactoriness of the so-called penal consequences view of penal substitution.  Unfortunately, she basically just reiterates what I say in ADC about my own misgivings concerning the penal consequences view of penal substitution. I prefer the punitive construal of penal substitution, first and foremost, because it best explains Isaiah’s Suffering Servant of the Lord and the NT identification of the Servant with Christ, but, secondly, because the punitive construal makes better sense of the satisfaction of divine justice.

So I shall leave the defense of the satisfactoriness of the penal consequences view to its proponents. But I shall register my surprise that this view is by no means indefensible. As noted in ADC, p. 196, the doctoral dissertation of Blaine Swen, “The Logic of Divine-Human Reconciliation” (2012) provides a surprisingly plausible account of the satisfaction of divine justice on a non-necessitarian theory of vicarious non-punitive atonement. Jansen herself in the end acknowledges,

It is unclear to me why, if God so chose to, the penal consequences endured by Christ on the cross could not also then satisfy justice, as in the same way, Craig argues for in the punishment view of PSA. And if Craig wants to maintain that there is a divine concession to the demands of retribution. . . which allows for the death of Christ to pay such debt according to the punishment endured, then it is strange to not consent to the idea that God could relax the demands of retributive justice to accept the suffering or penal consequences endured by Christ.

This is exactly Swen’s position, and it seems to me not indefensible. But since I do not in the end relax or qualify God’s negative retributive justice, preferring instead the doctrine of imputation, the question does not concern me.

I shall rather comment here on the satisfactoriness of punitive penal substitution. In dealing with this question, I return to the subject of the vicarious liability of a superior for the wrongs of a subordinate, which, it will be recalled, I take to be analogous to the imputation of our sins to Christ. Against those who deny that we have any experience of the satisfaction of justice through the punishment of a substitute, I point to cases where, in fact, the punishment of the superior alone is taken as satisfying the demands of justice.

In response, Jansen perceives a shift in my language between retributive justice and plain justice. But the perceived shift is merely verbal; we are still talking about satisfying the demands of positive retributive justice. But Jansen then makes the strange assertion that “Although the practice of vicarious liability may be justifiable in the American justice system, it is not an example of desert-based retributive justice.” I take it that the intent of this statement is to claim that a finding of vicarious liability does not imply that the person involved deserves punishment. This is patently false, since such persons are found guilty by the court and sentenced to punishment. We should not want to impugn our courts by claiming that the guilty receive sentences that they do not deserve. If that were the case, there would be no need for a finding of vicarious liability in the first place! Finally, to repeat, Jansen alleges that we have no legal precedent for vicarious liability for murder, which we have seen to be mistaken.

Finally, I should like to say a word about one of the fascinating examples of penal substitution in our justice system that came to my attention after writing ADC, namely the Law of Indemnity. In law, one party can agree to stand in the place of and legally be, as it were, another party for certain legal purposes, particularly for bearing the legal liabilities for the guilt of another. This is the essence of the indemnity concept. This concept finds expression in an indemnity contract such as a common liability insurance policy or an indemnity covenant collateral to or part of an underlying contract of broader scope (such as a contract to build a building, drill an oil well, or other contractual undertakings).

The indemnity concept works as follows:  party R (the Indemnitor), agrees to “indemnify” or protect and keep harmless party E (the Indemnitee) from the legal consequences of E’s actual or alleged wrongful acts or omissions for which the law prescribes punishment. In the absence of such an arrangement criminal penalties may be imposed on E. Should E be sued and found responsible, R will assume and discharge E’s legal penalty. Thus the law will hold R, who is actually innocent, ultimately responsible for E’s crime. In a very real sense, the indemnified party E looks to the innocent indemnitor R for protection from the punishments that the law would demand of E, were it not for R’s agreement (or covenant) to indemnify E. R by his covenant is E’s “savior.”

According to a simple indemnity covenant, R agrees to indemnify, hold harmless, and defend E from “X,” where “X” is any specified legal situation or risk.  Thus, should E be sued for a claim covered by X, R is obligated to:

• “indemnify” E, that is, to stand in place of E and, should E be found guilty, to pay or otherwise satisfy the legal claims against E, to take E’s legal hit, as it were; and to
• “hold harmless” E, that is, to fully protect and preserve E harmless from all claims (lawsuits, related costs, damages and penalties assessed, etc.) against E; and to
• “defend” E, that is, to personally appear at bar before the Court and opposing lawyers as E’s advocate and counselor and defend E diligently against all claims and claimants.

The theological analogy to Christ’s atoning and intercessory work on our behalf is patent and striking. Indeed, the theologian can appreciate the dramatic definition of “indemnity” given by the classic Black’s Law Dictionary:

A collateral contract of assurance by which one person engages to secure another against an anticipated loss or to prevent him from being damnified by the legal consequences of an act or forbearance on the part of one of the parties or of some third person.

Of course, the offenses for which indemnity under the law is available will be limited. But if God should wish to covenant with man to indemnify him against all offenses, no matter how many or how heinous, who is to deny him?

Cockrell concludes that “Notions of imputed and vicarious liability and punishment are now insinuated into Anglo-American law. The close correspondence between them and PSA is apparent” (personal correspondence, March 2, 2022). It is eloquent testimony to the “silo effect” of academic specialization that theologians and philosophers, in spite of the pervasiveness of vicarious liability and punishment in our justice system, can still assert that there is “nothing in our experience” parallel to the imputation of our sins to Christ and to his substitutionary punishment on our behalf.