Chapter 25
The Coherence of Penal Substitution — A Philosophical Defense

Introduction: Why Philosophy Matters for the Atonement

We have spent much of this book building the biblical and historical case for penal substitutionary atonement. We traced the sacrificial system of the Old Testament (Chapters 4–5), walked through Isaiah 53 (Chapter 6), examined the teaching of Jesus and the apostles (Chapters 7–12), and surveyed how the church has understood the cross from the earliest centuries to the present (Chapters 13–18). In Part V we brought the major atonement models together and argued that penal substitution rightly stands at the center of a multi-faceted understanding of what Christ accomplished on Calvary (Chapter 24).

Now we need to shift gears. We are entering Part VI, the philosophical portion of our study, and the questions here are different in kind from those we have faced so far. Instead of asking, "What does the Bible teach?" or "What did the Church Fathers believe?" we are asking something more fundamental: Does penal substitutionary atonement even make sense? Is it logically coherent? Can one person justly bear the punishment due to another? Or is the whole idea riddled with contradictions that no amount of biblical evidence can rescue?

These are not idle academic puzzles. Critics of penal substitution have pressed precisely these objections for centuries, going all the way back to the Socinian controversies of the sixteenth century. And in our own day, some of the sharpest attacks on PSA come not from exegetical disagreements but from philosophical ones. Scholars like Mark C. Murphy, Eleonore Stump, and others have argued that penal substitution is not merely wrong—it is incoherent. It cannot even get off the ground as a theory, they say, because its central claims violate basic principles of justice and the very definition of punishment.

I take these objections seriously. If penal substitution really were logically incoherent, no amount of proof-texting would save it. A doctrine that contradicts itself cannot be true. So we owe it to ourselves—and to honest critics—to face these challenges head-on and ask whether the doctrine can withstand rigorous philosophical scrutiny.

Unfortunately, most theologians and even most Christian philosophers have not given these challenges the sustained attention they deserve. As Craig observes, the doctrine of penal substitution is typically dismissed by its critics in a single paragraph—sometimes a single sentence—asserting that it would be unjust for God to punish an innocent person for the sins of others, and that is taken to settle the matter.⁴⁴ But far more deserves to be said. The philosophy of law, the theory of punishment, and analytic philosophy of religion have developed sophisticated conceptual tools that bear directly on the questions we are asking. It is time for defenders of PSA to deploy them rigorously.

A word of caution is in order at the outset. When legal theorists and philosophers of law discuss punishment, they almost always have in mind the kind of punishment administered by the state within a system of criminal justice. This framework is analogous to divine justice, but it is not identical to it. There will be features of divine justice that have no counterpart in any human legal system—just as there will be features of human justice that do not apply to God. God is not hampered by limited prison space or budgetary constraints. He does not need to worry about political fallout or the logistics of enforcement. But the fact that human legal systems and divine justice are not identical does not mean we cannot learn from the former. Given how saturated the biblical texts about the atonement are with forensic and judicial language—justification, condemnation, pardon, verdict, sentence—we should expect the tools of legal philosophy to shed genuine light on the doctrine.

I believe it can. In fact, I believe the philosophical case for penal substitution is remarkably strong, far stronger than most critics acknowledge. In this chapter, we will address four major philosophical objections: (1) the alleged incoherence of substitutionary punishment, (2) the injustice of punishing the innocent, (3) the so-called double payment problem, and (4) the forgiveness negation objection. In each case, we will state the objection in its strongest form, then offer a careful and thorough response. We will draw extensively on the philosophical work of William Lane Craig, whose Part III of Atonement and the Death of Christ represents the most rigorous philosophical defense of PSA in contemporary scholarship.¹

The Alleged Incoherence of Substitutionary Punishment

Let us begin with the most fundamental challenge. Some philosophers have argued that penal substitution is not merely unjust but conceptually incoherent—that the very idea of one person bearing another's punishment makes no logical sense. The most sophisticated version of this argument comes from philosopher Mark C. Murphy, and it deserves careful attention.

Murphy's Expressivist Objection

Murphy's argument rests on a particular theory of punishment called expressivism. Made influential by the legal philosopher Joel Feinberg, expressivism holds that punishment is not merely the infliction of harsh treatment on a person. What makes harsh treatment count as punishment, rather than mere suffering or penalty, is that it expresses condemnation or censure of the wrongdoer for the wrong he has done.² A parking ticket, on Feinberg's view, is a mere penalty. But a criminal sentence, accompanied by a verdict of "guilty," expresses society's moral condemnation of the offender. That expressive dimension is what makes it punishment in the fullest sense.

Murphy seizes on this point and argues that if expressivism is correct, then penal substitution is conceptually impossible—not just wrong, but literally unintelligible. Here is the logic: punishment inherently expresses condemnation of the person punished. But God cannot condemn Christ, because Christ did nothing wrong. He is sinless. Therefore, any harsh treatment God inflicts on Christ simply cannot count as punishment, no matter how severe it is. And if Christ cannot be punished, then penal substitution—the idea that Christ bore our punishment—collapses.³

Murphy is careful to distinguish his argument from the more familiar moral objection (that it would be unjust to punish the innocent). His claim is more radical: it is not that punishing Christ would be immoral; it is that "punishing Christ" is a contradiction in terms. You cannot condemn someone you know to be innocent, and you cannot punish without condemnation. So penal substitution, Murphy concludes flatly, "is incoherent."⁴

This is a serious challenge. Let me lay out several responses.

Response 1: Penal Substitution without "Punishment" — The Penalty Distinction

The first response is the simplest, and it may surprise some readers. Not all defenders of penal substitution hold that God punished Christ in the technical sense Murphy has in mind. Some hold that Christ voluntarily took upon himself the suffering that would have been our punishment had it been inflicted on us. On this view, Christ bore the penalty of sin—He endured the harsh treatment that our sins deserved—but He was not "punished" in the narrow, expressivist sense of the word.

Craig makes this point with characteristic clarity. Feinberg himself distinguishes between penalties (like parking fines, sports infractions, and firings at work) and punishments technically so called. If we borrow this distinction, the defender of penal substitution can simply say that God penalized Christ for our sins—Christ paid the penalty for our sins—without claiming that God condemned Him.⁵ The word "penal" in "penal substitution" need not carry the full weight of Feinberg's expressivist definition. It can refer to the judicial consequences of sin, the penalty attached to law-breaking, which Christ bore on our behalf.

Craig observes that this response exposes Murphy's argument as what the great legal philosopher H. L. A. Hart called a "definitional stop"—an attempt to shut down debate by defining the key term so narrowly that the opposing position is ruled out before the conversation even begins.⁶ Hart had made the same critique of an earlier philosopher, Anthony Quinton, who defined punishment in such a way that it became logically impossible to punish the innocent. But as Hart pointed out, innocent people are in fact sometimes convicted and harshly treated for crimes they did not commit. Calling this "not really punishment" is just a semantic maneuver that obscures the real issues.

The philosopher David Lewis responded to this same kind of definitional stop with characteristic wit. He acknowledged the expressivist's narrow definition and simply said: "I trust that the reader will understand: I mean that the volunteer undergoes something that would have constituted punishment if it had happened instead to the guilty offender."⁷ Whether we call it "punishment" or "penalty," the substance remains. Christ endured what was due to us. The debate about justice can then proceed on its real merits.

Response 2: Expressivism Is Not Mandatory

But suppose we are not content merely to sidestep the word "punishment." Suppose we do want to affirm that God punished Christ for our sins, as many in the Reformed tradition have said. Is expressivism so well established that we have no choice but to accept it?

Not at all. Murphy never adequately considers the possibility that a defender of penal substitution might reasonably reject an expressivist theory of punishment altogether. While popular, expressivism is far from unassailable. Craig documents several significant problems with it.⁸

First, the expressivist theory does not "carve the cases properly," as Murphy claims. The line between condemnatory and non-condemnatory harsh treatment does not neatly coincide with the line between punishments and penalties in the law. Many civil penalties—tort awards for assault and battery, fraud, wrongful death—clearly express society's moral disapproval of the wrongdoer. Punitive damages, by their very name, are punitive. Even sports penalties for unsportsmanlike conduct carry an element of censure. Conversely, there are criminal punishments that seem to involve no condemnation at all. Crimes involving mala prohibita (acts wrong only because they are legally prohibited, like certain regulatory violations) are punishable even though the punishment may not carry any moral stigma.⁹

Second, consider cases of strict liability in criminal law—cases where a person is found guilty and punished even though he had no blameworthy mental state (mens rea) whatsoever. These are not rare anomalies. There are thousands of statutory offenses involving strict liability, from possession of controlled substances to selling mislabeled food products. In such cases, the person convicted may be utterly blameless in his intentions and yet is still genuinely punished. If punishment can be imposed without any finding of moral blameworthiness, then the expressivist insistence that punishment inherently expresses condemnation of the person punished starts to look very doubtful.¹⁰

Third—and this is a particularly powerful point—substitutionary punishment was widely recognized and practiced in the ancient world, and far from being condemned, those who voluntarily stepped forward to die as substitutes for others were universally admired as paragons of nobility. As Hugo Grotius documented in the seventeenth century, ancient societies accepted substitutionary punishment as a real and sometimes honorable practice.¹¹ We moderns may regard it as unjust, and we may well be right, but we cannot coherently claim that these societies were not "really" practicing punishment. To do so would be an exercise in cultural imperialism, imposing our narrow definitions on practices that ancient peoples clearly understood as genuine instances of substitutionary punishment.

The legal philosopher Leo Zaibert goes further and argues that Feinberg's expressivist theory is actually dangerous for a democratic society, because it allows the state to inflict clearly punitive treatment while claiming it is not "really" punishment—since it does not carry the expressivist's stamp of formal condemnation. The U.S. Supreme Court's ruling that deportation is merely an "administrative matter" rather than punishment is a troubling example of exactly this kind of abuse.¹²

Response 3: Expressivism without Condemnation of Christ

Here is perhaps the most elegant response of all. Even if we accept expressivism, it does not actually rule out penal substitution. The reason is that expressivism, as typically formulated, does not require that condemnation be directed at the person who is being punished. It requires only that the punishment express condemnation of the wrongful act.

Look carefully at standard expressivist formulations. Feinberg himself says that punishment expresses "the community's strong disapproval of what the criminal did."¹³ Note: what the criminal did—not necessarily condemnation of the specific person bearing the penalty. Similarly, Alec Walen's widely cited characterization in the Stanford Encyclopedia of Philosophy states that punishment must be imposed "as a way of sending a message of condemnation or censure for what is believed to be a wrongful act or omission."¹⁴ Again, the condemnation is aimed at the act, not necessarily at the individual undergoing the harsh treatment.

Craig draws the decisive inference: Murphy's claim that "punishment expresses condemnation of the person punished" simply does not follow from the standard expressivist formulations. It is a non sequitur.¹⁵ God can impose harsh treatment on Christ that expresses His condemnation of sin—His righteous opposition to the wrongful acts of humanity—without that harsh treatment expressing condemnation of Christ personally. In fact, this is precisely what orthodox theology has always claimed: the cross is God's definitive "No!" to sin, His ultimate expression of how seriously He takes human rebellion against His holy law. Christ, the sinless one, bears this expression of divine opposition to sin on our behalf. The condemnation falls on sin itself, not on the Lamb of God who takes it away.

Response 4: Imputation Makes Condemnation Intelligible

There is a fourth response available, one rooted in the Reformation doctrine of imputation that we will develop more fully in Chapter 28. If the sins of humanity are genuinely imputed to Christ—if, in the court of divine justice, Christ is reckoned as bearing the legal guilt of our transgressions—then even on the narrowest formulation of expressivism, God's condemnation would be directed at Christ as the one who is now legally liable for those sins. Christ, though personally sinless, bears the legal guilt of those He represents. Murphy himself concedes that if the doctrine of imputation is granted, his charge of incoherence collapses.¹⁶

Of course, critics will then question the coherence of imputation itself. Can legal guilt really be reckoned to a person who did not commit the offense? We will address that question in detail when we discuss legal fictions, vicarious liability, and federal headship in Chapters 27 and 28. For now, I simply note that the Reformers' doctrine of imputation provides yet another avenue by which the coherence of penal substitution can be maintained. As Craig summarizes, "multiple responses to the coherence objection are available to the penal substitution theorist, depending on how many of the critic's premises one accepts. Therefore, the charge of philosophical incoherence has not been sustained."¹⁷

The Justice of Punishing the Innocent: The Socinian Objection

Even if we grant that penal substitution is coherent—that it is a logically possible idea—a deeper objection remains. This is the charge that has been pressed against penal substitution more than any other, from Faustus Socinus in the sixteenth century down to our own day. The objection is straightforward: It is unjust to punish an innocent person for the crimes of the guilty. Christ was innocent. Therefore, it would be unjust for God to punish Christ for our sins. And since God is perfectly just, He cannot have done so.

Vee Chandler presses this objection forcefully. She cites Ezekiel 18:20—"The one who sins is the one who will die"—and Deuteronomy 24:16—"each will die for their own sin"—to argue that God's own revealed principles of justice preclude the transfer of punishment to an innocent party. "According to what God has revealed about his own justice," Chandler writes, "it would be legally indefensible to transfer to Christ the punishment that belonged to sinners."¹⁸ She adds pointedly that Scripture repeatedly condemns the shedding of innocent blood, making it ironic that penal substitution attributes to God the very thing He has declared to be unjust.¹⁹

This is a powerful objection, and we must take it with full seriousness. Let me offer several lines of response.

Response 1: Christ Is Not a Third Party Dragged In against His Will

The first and most important response concerns the identity and consent of the substitute. The Socinian objection conjures a picture that does not match what Christians actually believe happened at Calvary. It imagines an innocent bystander being dragged, unwilling, into a punishment that has nothing to do with him. But that is precisely what penal substitution does not teach.

Jesus Christ went to the cross voluntarily. "No one takes my life from me," He said, "but I lay it down of my own accord. I have authority to lay it down, and I have authority to take it up again" (John 10:18, ESV). The Son chose to bear the judicial consequences of human sin freely, out of love for those He came to save. This was no coerced sacrifice but a willing self-offering. Paul makes the same point when he writes that Christ "loved me and gave himself for me" (Galatians 2:20, ESV). The voluntariness of Christ's sacrifice is not a minor detail that we tack on to the doctrine; it is essential to its moral logic.

Why does consent matter so much? Because the moral character of an action depends in part on whether the parties involved act freely. A stranger tackling you on the street is assault. A surgeon cutting into your body with your informed consent is medicine. The physical action may look similar, but the consent of the patient fundamentally changes the moral reality. Consent does not make every action permissible, of course. But when someone with the appropriate standing voluntarily chooses to bear a burden on behalf of another, the moral dynamics are categorically different from when suffering is imposed on an unwilling victim.

But it goes even further than consent. As John Stott argued so powerfully, the cross is not the story of an angry Father punishing a reluctant Son. It is the story of God Himself—the Lawgiver, the Judge, the offended party—stepping into the dock and bearing the penalty in His own person. Stott called this "the self-substitution of God," and I believe it is one of the most important theological insights ever articulated about the atonement.²⁰ When we say that Christ bore our punishment, we are saying that God Himself bore it. The Judge took the sentence upon Himself. This is something fundamentally different from the punishment of a random innocent person, and the Socinian objection simply does not reckon with it.

Consider the analogy of a judge who, after passing sentence on a convicted friend, steps down from the bench, removes his robes, and says, "I will serve the sentence myself." We might debate whether such a thing should be allowed in a human courtroom, but no one would call it unjust in the same way that punishing an uninvolved bystander would be unjust. The Judge has every right and every standing to make such a sacrifice. And when the Judge is also the Lawgiver—when He is the one against whom every sin is ultimately committed (Psalm 51:4)—the moral landscape shifts even more dramatically.

This is why the Trinitarian theology of the atonement, which we developed in Chapter 20, is not merely a theological nicety but a philosophical necessity. If the Father and the Son are two separate beings with conflicting interests—one angry, the other victimized—then the Socinian objection carries real weight. But if the Father and the Son are one God, acting together in unified purpose and love, then the cross is not the punishment of an innocent third party but the self-sacrifice of the offended Lawgiver. As Stott put it memorably, "The concept of substitution may be said, then, to lie at the heart of both sin and salvation. For the essence of sin is man substituting himself for God, while the essence of salvation is God substituting himself for man."²¹

Response 2: God Is Not Bound by the Same Rules He Imposes on Us

A second response involves what Craig calls "metaethical contextualization."²² The Protestant proponents of penal substitution—from the Reformers through Turretin—all held to some form of what philosophers call a Divine Command Theory of ethics. On this view, moral duties are constituted by God's commands. God issues commands to human beings, and those commands create moral obligations for us. But God does not issue commands to Himself in the same way, and therefore He does not stand under the same moral obligations we do.

Now, this does not mean God can do anything whatsoever. He always acts in accordance with His nature—He cannot lie, He cannot be cruel, He cannot act contrary to His own perfect character. But He has prerogatives that human beings do not have. He gives life and takes it. He ordains the sacrificial system. He tested Abraham with the extraordinary command to offer Isaac—a command that would have been morally outrageous between human persons but which, coming from the sovereign Lord of all life, served a profound redemptive purpose.

As Hugo Grotius observed in his defense of the atonement, even if God has established a system of justice among human beings that forbids the punishment of the innocent, God Himself is not bound by that prohibition in the same way.²³ He refused Moses' offer to serve as a substitutionary sacrifice (Exodus 32:30–34). He stopped Abraham from sacrificing Isaac. But if God wills to take on human nature in the person of Jesus of Nazareth and give His own life as a sacrificial offering for sin, who is to forbid Him? As Craig puts it, "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?"²⁴

Chandler anticipates this response and argues that "injustice remains injustice even if the one who suffers it is God himself."²⁵ But I think she moves too quickly here. The claim is not that God's identity magically turns injustice into justice. The claim is that the entire moral situation is fundamentally different when the Lawgiver Himself, against whom all sin is committed, freely chooses to bear the penalty in His own person. To treat this as identical to the punishment of an uninvolved innocent party is to ignore the most important features of the situation.

Response 3: Retributive Justice Rightly Understood

The Socinian objection also fails to distinguish carefully between different dimensions of retributive justice. Craig helpfully distinguishes between positive retributivism—the thesis that the guilty deserve punishment—and negative retributivism—the thesis that the innocent should not be punished.²⁶ The essence of retributive justice lies in the positive thesis: the guilty deserve punishment because they have done wrong. God is a positive retributivist who "will by no means clear the guilty" (Exodus 34:7).

But the penal substitution theorist can maintain that God is only a qualified negative retributivist. That is, even if God has prohibited human beings from punishing the innocent (Deuteronomy 24:16), and even if He is too good to punish innocent human persons without reason (Genesis 18:25), He reserves the sovereign prerogative to allow an innocent divine person—namely Christ—to bear the penalty in place of the guilty. This extraordinary exception flows not from a defect in His justice but from the overflow of His goodness and love. It is not a case of God violating His own principles; it is a case of God exercising a unique divine prerogative that goes beyond anything available within the human legal system.²⁷

Response 4: Prima Facie vs. Ultima Facie Justification

Even if we insist that God's justice includes unqualified negative retributivism—that the innocent should never be punished—the objection still does not succeed. Why? Because moral philosophers and legal theorists have long recognized that the prima facie (at first glance) demands of justice can be outweighed in specific cases by weightier moral considerations, making an action justified ultima facie (all things considered).

This distinction between justifying the practice of punishment and justifying a specific act of punishment is well established in legal philosophy. When positive retributivists say "the guilty should be punished," they are stating a general principle about the practice of punishment. But in specific cases, the prima facie demands of retributive justice may be outweighed by other moral concerns—protecting the rights of others, securing cooperation in prosecuting greater crimes, or, in the case of strict liability, protecting public welfare.²⁸

The biblical scholar D. A. Carson makes a striking point: it is precisely the unjust punishment of the Servant in Isaiah 53 that makes salvation possible. "Forgiveness, restoration, salvation, reconciliation—all are possible," Carson writes, "not because sins have somehow been canceled as if they never were, but because another bore them unjustly." He clarifies that "by this adverb 'unjustly' I mean that the person who bore them was just and did not deserve the punishment, not that some moral 'system' that God was administering was thereby" violated.²⁹ The prima facie injustice of punishing an innocent person is overridden, in the case of Christ, by the staggering moral good that results: the salvation of the entire human race.

When we weigh the options—either every human being faces the just consequences of sin, or the sinless Son of God voluntarily bears those consequences so that all might be saved—it becomes very difficult to see how the latter option represents a moral failure on God's part. It looks, instead, like the supreme act of holy love.

Think of it from the opposite direction. Suppose God had refused to allow any exception to the principle of negative retributivism. Suppose He had said, "I will never, under any circumstances, allow an innocent person to bear the penalty for the guilty—not even if the innocent person is My own Son, not even if He volunteers, not even if it would save the entire human race." Would this God be more just? Or would He simply be more rigid? Would we admire such a God—a God who allows every sinner to perish rather than accept the willing self-sacrifice of His own Son? I do not think we would. We would say that such a God had elevated a principle of human justice to an absolute that trumps even His own love, mercy, and creative sovereignty. And that seems like a distortion of justice, not a defense of it.

The biblical picture is far richer than any single principle of retributive justice can capture. The God of the Bible is indeed a God of justice—fierce, unyielding justice that takes sin with utmost seriousness. But He is also a God of staggering creativity, who finds ways to uphold justice and extend mercy simultaneously. The cross is the supreme expression of that divine creativity. It is not a violation of justice but its deepest fulfillment—a fulfillment that no merely human legal system could have imagined.

Response 5: Civil Law, Criminal Law, and the Limits of Analogy

Chandler raises one more objection worth addressing. She argues that defenders of penal substitution improperly blur the distinction between civil law and criminal law. In civil law, one person can pay another's debt—money is fungible, and a creditor does not care whose wallet the payment comes from. But in criminal law, punishment is personal. You cannot serve someone else's prison sentence. Death and corporal punishment are fundamentally non-transferable. Chandler argues that sin falls under criminal, not civil, jurisdiction, and therefore the debt-payment analogy breaks down.³⁰

This is a thoughtful point, and Chandler is right that the analogy between sin and monetary debt has real limitations. But her argument proves too much. As David Lewis has observed, while we do not normally allow someone to serve another person's prison sentence, we do allow someone to pay another person's criminal fine. "Yet this is just as much a case of penal substitution as the others," Lewis notes.³¹ Some criminal fines are every bit as burdensome as prison sentences and carry the same moral stigma. If we were truly single-mindedly opposed to substitutionary punishment, we would have to conclude that fines are an unsatisfactory form of punishment. But we do not.

Moreover, as we shall see in Chapter 27, the law does recognize cases of vicarious liability in criminal law—cases where a blameless employer is held criminally guilty for acts committed by an employee. These cases show that even in criminal law, the idea of one person bearing legal responsibility for another's actions is not as alien as Chandler's sharp civil/criminal distinction suggests.

I want to be clear: human legal analogies can only illuminate the atonement; they cannot define it. The cross of Christ transcends every human legal system. No courtroom in the world has ever seen anything like the sovereign Lord of the universe voluntarily taking on human nature and bearing in His own person the judicial consequences of cosmic rebellion against His own law. The categories of human justice give us windows into what God has done, but the reality is far greater than any analogy can capture.

The Theological Coherence Objection: Stump on Love and Forgiveness

Before we move to the next major objection, I want to address one more sophisticated challenge to the coherence of PSA. Philosopher Eleonore Stump has argued in her important book Atonement that penal substitutionary theories are "incompatible with God's love."³² Since defenders of PSA also affirm God's love, she concludes that the theory is internally incoherent and therefore beyond repair.

Stump's argument runs like this: God is perfectly loving. A perfectly loving God is perfectly forgiving. Perfect forgiveness is unconditional—it has no preconditions. But on penal substitutionary theories, God's forgiveness does have a precondition: the satisfaction of divine justice through Christ's atoning death. Therefore, penal substitution is incompatible with God's love.

This sounds compelling at first. But Craig identifies two fundamental problems with it.³³

First, given Stump's own definitions of love and forgiveness, God as described by penal substitutionary theories actually meets her conditions for being perfectly loving and forgiving. Stump defines love as involving a desire for the good of the beloved and a desire for union with him. She defines forgiveness as maintaining those desires even when the beloved has wronged you. But nothing in penal substitutionary theories denies that God desires the good of every person and desires union with every person, even apart from the satisfaction of divine justice. If Arminian or Molinist defenders of PSA affirm—as they do—that God wills the good of all persons and desires union with all persons, then on Stump's own definitions, God forgives them unconditionally. The satisfaction of justice is not a precondition of God's love; it is the means by which God's love expresses itself in the actual removal of condemnation.

Second, and more fundamentally, Stump's entire approach construes God on the analogy of a private person in a personal relationship rather than as a Judge and Ruler of the moral universe. She frequently compares God to a friend who has been wronged by another friend, and she asks how a loving friend would respond. But as Grotius pointed out centuries ago—and as Jeffrie Murphy has argued in contemporary legal philosophy—there is a critical difference between the private and public spheres.³⁴ A private individual who is wronged can simply choose to let it go. But a judge, a ruler, an administrator of justice, has obligations that go beyond personal feelings. A judge may personally love the defendant standing before him. He may wish with all his heart that he did not have to impose sentence. But his duty as a judge requires him to uphold the law. He can forgive personally and still administer justice publicly.

This is exactly the situation with God. He is not merely a private party in a personal dispute. He is the Lawgiver, Judge, and Ruler of the moral order. As such, He bears a responsibility to uphold the moral integrity of His creation. He can personally love sinners—and He does, passionately—while still requiring that the demands of justice be satisfied before legal pardon is issued. The cross is where these two realities converge: God's infinite love for sinners and His unyielding commitment to justice meet at Calvary, and both are fully satisfied in the self-giving sacrifice of the Son.

The "Double Payment" Objection

We turn now to a different kind of philosophical challenge. Some critics press what is called the "double payment" objection. The logic runs like this: If Christ has already paid the full penalty for sin, then the debt is discharged. The books are balanced. Justice is satisfied. So why must people still exercise faith to be saved? If the penalty has been fully paid, it would be unjust to require anything further. And conversely, if faith is required, then the penalty cannot really have been fully paid.

This objection has a long history. In its strongest form it was pressed by the "hyper-Calvinist" strand of Reformed theology, which reasoned that if Christ truly paid for the sins of the elect, then the elect are already justified from the moment of Christ's death (or even from eternity), and faith is merely the recognition of a salvation that has already been fully accomplished. Some have even argued that if Christ's payment is truly complete, then double jeopardy applies: God cannot demand payment twice—once from Christ and once from the sinner.

I believe this objection, while understandable, rests on a confusion between two distinct things: the objective accomplishment of the atonement and its subjective appropriation.

Accomplishment and Appropriation

When we say that Christ "paid the penalty" for sin, we are saying that He provided the sufficient ground on which God can justly pardon sinners. His death is the basis of forgiveness—it satisfies the demands of divine justice and makes pardon possible. But the Bible consistently teaches that this objective accomplishment must be received by the individual through faith. As Paul puts it, "since we have been justified by faith, we have peace with God through our Lord Jesus Christ" (Romans 5:1, ESV). Justification comes through faith. Faith is not a second payment; it is the means by which the individual receives the benefit of the one payment already made.

The great Puritan theologian J. I. Packer helpfully clarified this point by distinguishing between two very different models of how Christ's work relates to the sinner's benefit. One model, which Packer called the "commercial" or "pecuniary" model, treats Christ's death as a strict financial transaction: the exact amount owed by the sinner is deposited into the divine treasury, the ledger is balanced, and the sinner is automatically freed. On this model, no further action should be required—the debt is cancelled the moment the payment is made. The other model, the properly "penal" model, holds that Christ bore not a specific quantum of debt but the type of punishment due to sin—death and separation from God—and that God, as the sovereign authority, determines how and to whom the benefits of this substitutionary suffering are applied.⁴⁶

This distinction is crucial. On the pecuniary model, the double payment objection has some force. If Christ deposited the exact monetary equivalent of my debt, then demanding further payment would indeed be unjust. But the atonement is not a commercial transaction. It is a judicial act of the sovereign God. And on the properly penal model, God retains the sovereign prerogative to determine the conditions under which the benefits of Christ's death are applied. He has determined that faith is the instrument of application—not because faith earns or adds to what Christ has done, but because God has ordained that the personal reconciliation accomplished objectively at the cross is received through a personal act of trust.

Craig's discussion of divine pardon is illuminating here. Drawing on extensive research into the philosophy of law, he argues that divine forgiveness is best understood on the analogy of a legal pardon issued by a ruler, rather than the mere forgiveness of a private individual. A pardon is an official act by which the executive authority exempts a person from the legal consequences of a crime. But—and here is the crucial point—a pardon must be accepted.³⁵

The landmark U.S. Supreme Court case Burdick v. United States (1915) addressed exactly this issue, and the Court noted that a pardon "carries an imputation of guilt; acceptance a confession of it."³⁶ A pardon is not imposed on someone; it is offered and then accepted. The president may sign the pardon papers, but the convicted person must accept the pardon for it to take effect. Similarly, God has accomplished the objective work of atonement through Christ's death, but the individual must accept this pardon through faith for its benefits to be applied.

The analogy is not perfect, of course. No analogy is. But it captures something important. Christ's atoning death is like a pardon that has been signed, sealed, and published—it is objectively accomplished and universally sufficient. But it does not become personally effective for any individual until it is received by faith. Faith is not a work that adds to what Christ has done; it is the open hand that receives the gift already given.

Payment and Pardon Are Not Incompatible

The double payment objection also assumes that if a debt is paid, no further action is needed—forgiveness becomes automatic. But this confuses two different metaphors. In the strict "commercial transaction" model of the atonement (sometimes called the "pecuniary" model), Christ's death is treated like a monetary payment: once the exact price is deposited, the debt is canceled automatically, and no further action is required. This was the model favored by some in the "strict satisfaction" tradition.

But most careful defenders of penal substitution have recognized that this commercial metaphor, while capturing something true, needs to be supplemented. Stott himself warned against treating the cross as a mere "commercial bargain."³⁷ The atonement is not a bank transaction. It is a personal act of the Triune God, and personal acts require personal responses. God, as sovereign Ruler, determines the conditions under which the benefits of Christ's death are applied to individuals. He has determined that faith is that condition—not because faith adds anything to the sufficiency of the cross, but because God has chosen to apply the benefits of the cross through a personal relationship of trust and dependence.

Think of it this way: a wealthy benefactor might pay the full tuition for every student at a university. The money is deposited. The debt is covered. But each student still needs to enroll—to accept the offer and show up for classes. Enrollment does not add to the payment; it receives the benefit of the payment. Faith functions in a similar way. It is not a second payment but the God-appointed means by which we participate in the benefit of the one payment Christ has already made.

The "Forgiveness Negation" Objection

Closely related to the double payment problem is what I call the "forgiveness negation" objection. It goes like this: If the penalty for sin has been fully paid, then there is nothing left to forgive. Forgiveness means releasing someone from a debt they owe. But if the debt has already been paid in full by a third party, there is no debt left to release. You can have payment or forgiveness, but you cannot have both. So if PSA is true, God does not actually forgive anyone—He merely acknowledges that the account has been settled.

This is a clever objection, and it has real force if we think of the atonement in purely commercial terms. If Christ's death was simply a financial transaction that zeroed out a ledger, then the critic is right: there is nothing left for God to "forgive." But once again, the objection rests on a too-rigid application of the commercial metaphor.

God Determines the Conditions of Satisfaction

The key insight, as Craig develops it, is that God, as the sovereign Lawgiver and Judge, has the prerogative to determine the conditions under which the demands of justice are satisfied.³⁸ He is not like a creditor who is bound by market rules. He is the supreme authority who determines what counts as satisfaction. And He has determined that Christ's death provides the just basis on which forgiveness can be extended. Christ's sacrifice does not replace forgiveness; it makes forgiveness possible without compromising justice.

To see why this matters, think about what the critic is actually claiming. The forgiveness negation objection says: if justice has been satisfied, there is nothing to forgive. But this assumes that forgiveness and the satisfaction of justice are two fundamentally incompatible categories—that you must choose one or the other. Is this true? I do not think so. Consider a human analogy. Suppose a father tells his teenage son: "Because you wrecked the car through reckless driving, I am going to pay for the repairs out of my own savings. But I want you to understand how serious this is. Come sit with me and let us talk about what happened." The father has borne the financial cost himself. The "debt" has been paid. But has the father thereby ceased to forgive the son? Of course not. The act of bearing the cost is itself an expression of forgiveness. Payment and forgiveness are not opposites; in this case, the payment is the very vehicle through which forgiveness is expressed.

The atonement works in an analogous way. God bears the cost of our sin in Christ—that is the satisfaction of justice. And precisely because He has borne that cost, He is now free to extend genuine pardon without any compromise to His moral character. The cross does not eliminate the need for forgiveness; it creates the conditions under which forgiveness can be extended justly. As Paul says in Romans 3:25–26, God put Christ forward as a propitiation "to show his righteousness... so that he might be just and the justifier of the one who has faith in Jesus" (ESV). Both justice and justification are accomplished. Both the satisfaction of the moral order and the merciful pardon of the sinner are real. Neither cancels the other.

Consider again the analogy of pardon. When a governor or president pardons a criminal, the pardon is an act of mercy. But the pardon may be issued precisely because certain conditions have been met—time served, good behavior, or the satisfaction of certain legal requirements. The fact that conditions were met does not negate the pardon's character as an act of mercy. It provides the basis for the pardon while leaving the pardon itself as a genuine act of grace.

In the same way, Christ's death provides the just basis on which God can extend forgiveness to sinners. God's forgiveness is genuine forgiveness—a real act of grace by which He releases us from the liability our sins deserve. But it is not arbitrary forgiveness; it is forgiveness grounded in the satisfaction of justice. As Paul puts it in Romans 3:26, God shows Himself to be "just and the justifier of the one who has faith in Jesus" (ESV). Both justice and mercy are preserved. Neither is sacrificed. The cross is the place where they meet.

Fleming Rutledge captures this beautifully when she writes that the cross is not about choosing between justice and mercy but about the God who refuses to abandon either. The God of the Bible will not simply wave His hand and pretend that sin does not matter—that would be an offense against every victim of evil. But neither will He abandon sinners to the consequences they deserve. The cross is His answer: a justice so thoroughgoing that it takes sin with utter seriousness, and a mercy so profound that it absorbs the cost in His own person.³⁹

It is worth pausing to note how this resolves a tension that many thoughtful Christians feel intuitively. We know, from the whole tenor of Scripture, that God genuinely forgives us. The language of mercy, grace, compassion, and pardon saturates the Bible from Genesis to Revelation. But we also know—from the very same Scriptures—that sin is deadly serious, that God's justice is real, and that the cross was somehow necessary. The forgiveness negation objection tells us we must choose between these two truths. The doctrine of penal substitution tells us we do not. The cross is the place where both truths are fully honored. God's justice is satisfied—really and truly satisfied, not swept under the rug. And God's mercy is extended—really and truly extended, not reduced to a mere legal technicality. This is the glory of the gospel, and it is philosophically coherent.

Engaging Eleonore Stump's Alternative

Before closing this chapter, I want to engage briefly with the alternative atonement theory proposed by Eleonore Stump in her book Atonement. Stump, a distinguished philosopher at Saint Louis University and a committed Catholic, rejects all forms of penal substitution and satisfaction in favor of what she calls a "Thomistic" account focused on the restoration of personal relationship between God and the sinner. On her view, the purpose of Christ's passion and death is not to satisfy divine justice but to address the psychological and relational obstacles that prevent sinners from accepting God's love. Christ's suffering reveals the depth of God's love in a way that can break through human resistance and draw us into union with Him.⁴⁰

There is something genuinely beautiful in Stump's proposal, and I have no desire to dismiss what she gets right. The atonement certainly does involve the restoration of relationship. Christ's death certainly does reveal the love of God in a way that should melt the hardest heart. As we argued in Chapter 22, the moral influence dimension of the cross is a real and important facet of what Christ accomplished.

But I believe Stump's proposal is ultimately insufficient as a complete account of what happened at Calvary, for several reasons.

First, as we noted above, Stump's account treats God exclusively as a private party in a personal relationship. It has no room for God as Judge and Ruler. But the biblical witness is saturated with forensic and judicial language about the cross: justification, condemnation, redemption, propitiation, verdict, pardon. A theory of the atonement that ignores all of this forensic language is a theory that ignores a massive portion of the biblical data.

Second, Stump's account cannot adequately explain why Christ had to die. If the purpose of Christ's mission was merely to reveal God's love and overcome our relational resistance, why was a brutal crucifixion necessary? Could God not have revealed His love just as effectively through Christ's life and teaching—through His miracles, His compassion, His words of grace? A beautiful sunset reveals something of God's goodness. A miraculous healing reveals His power and compassion. Why, then, the agony of Golgotha? Why the cry of dereliction? Why the darkness at noon?

The moral influence theory, standing alone, has never been able to answer this question convincingly. If God's sole purpose was to demonstrate His love in a way that would melt human hearts, there were surely less horrific ways to accomplish this. But penal substitution answers the question directly: Christ died because sin carries a judicial penalty—the penalty of death and separation from God—and that penalty had to be borne. The cross was necessary not merely to display love but to deal with sin. As we argued in Chapter 3, God's love and God's justice are not in tension. They are complementary perfections. And the cross is the place where both are fully expressed: justice in the penalty that is borne, and love in the God who bears it.

Third, as Craig points out, Stump's account has the odd implication that God can both forgive sinners and punish them. Since forgiveness on her view is merely a change of attitude (a desire for the good of the wrongdoer), God can "forgive" sinners in Stump's sense while still sending them to hell—as long as He maintains a benevolent desire for their good even while imposing retributive justice. But this stretches the word "forgiveness" far beyond what any ordinary reader of the Bible would recognize. When Scripture says "If we confess our sins, he is faithful and just to forgive us our sins" (1 John 1:9), it clearly means that God removes the liability to punishment, not merely that He maintains warm feelings toward us while imposing punishment anyway.⁴¹

I believe the better approach—the approach this book has been building—is the one that holds penal substitution at the center while incorporating the genuine insights of other models. The moral influence of the cross is real, and Stump has articulated it beautifully. The relational restoration she describes is real. But these truths are grounded in the penal substitutionary work of Christ, not replacements for it. We love God because He first loved us—and He loved us most definitively by bearing in His own person the penalty our sins deserved.

Engaging Oliver Crisp's Concerns

We should also briefly address concerns raised by Oliver Crisp, a Reformed analytic theologian who has engaged carefully with the doctrine of penal substitution in works like Approaching the Atonement. Crisp is broadly sympathetic to substitutionary atonement but has raised questions about how the imputation of sin and the forensic justification of believers should be understood philosophically.

Crisp characterizes penal substitution as involving a "forensic fiction"—God treats Christ as if he were guilty and treats the believer as if she were sinless, even though neither is the case.⁴² This language understandably worries some defenders of PSA who fear that calling justification a "fiction" undermines its reality.

But Craig argues that this fear is misplaced. The claim is not that penal substitution itself is a fiction. Christ really suffered. Sin was really dealt with. Our forgiveness is real. What is fictitious—in the technical, legal sense—is the imputation of our sinful acts to Christ. God, in His role as supreme Judge, adopts the legal fiction that Christ committed the deeds for which we stand condemned. Legal fictions are a well-established and indispensable feature of human legal systems, and they produce real results. When Lord Mansfield declared that Minorca was "part of London" for jurisdictional purposes, everyone knew it was geographically false—but the legal fiction produced a genuinely just outcome by giving a wronged man access to the courts.⁴³ Similarly, the imputation of our sins to Christ—whether understood as a legal fiction or as a form of vicarious liability—produces the genuinely real outcome of our legal justification before God's bar.

Alternatively, as we will develop in Chapter 28, the imputation of sin may be understood not as a legal fiction at all but as a genuine instance of vicarious liability, where Christ as our representative bears the legal responsibility for the sins of those united to Him. In the criminal law of the Anglo-American tradition, vicarious liability is a well-established practice in which a blameless employer may be found legally guilty for wrongful acts committed by an employee. The guilt of the subordinate is, in a real legal sense, replicated in the superior—not transferred from one to the other, but imputed to the superior as well. Craig argues that this provides a compelling analogy for the imputation of our sins to Christ: our guilt is replicated in Him, so that He stands before the bar of divine justice as one who is legally liable for our wrongdoing, even though He Himself did no wrong.⁴³

The important takeaway from this discussion is that the philosophical resources for defending the coherence of imputation are far richer than critics typically acknowledge. Whether we understand imputation through the lens of legal fiction, vicarious liability, or the covenantal concept of federal headship (which we will explore in Chapter 28), the result is the same: the transfer of legal standing from one person to another is not a philosophical impossibility but a well-attested feature of human legal systems. If our own imperfect justice systems can accommodate such transfers, it would be strange to insist that the sovereign God of the universe cannot.

Conclusion: A Coherent, Defensible, and Beautiful Doctrine

We have covered a great deal of philosophical ground in this chapter, so let me summarize where we have arrived.

We examined four major philosophical objections to penal substitutionary atonement:

First, the coherence objection: the claim that substitutionary punishment is conceptually impossible because punishment inherently expresses condemnation of the person punished. We showed that this objection rests on a contested expressivist theory of punishment, that it can be circumvented by the penalty/punishment distinction, that even within expressivism the standard formulations do not require condemnation of the substitute, and that the doctrine of imputation provides yet another avenue of response.

Second, the Socinian objection: the claim that it is unjust to punish an innocent person for the crimes of the guilty. We showed that Christ is not a random innocent bystander but the divine Lawgiver Himself who voluntarily bears the penalty; that God has unique sovereign prerogatives not available in human legal systems; that the essence of retributive justice lies in positive retributivism (the guilty deserve punishment), not in an unqualified negative retributivism; and that the prima facie demands of justice can be overridden by the incomparable good of human salvation.

Third, the double payment objection: the claim that if Christ paid the full penalty, nothing more should be required. We showed that this confuses the objective accomplishment of the atonement with its subjective appropriation, and that faith is not a second payment but the means by which the individual receives the benefit of the one payment already made—analogous to the acceptance of a pardon.

Fourth, the forgiveness negation objection: the claim that payment and forgiveness are mutually exclusive. We showed that Christ's death provides the just basis for forgiveness rather than replacing it, and that God as sovereign Judge determines the conditions under which the demands of justice are satisfied.

Along the way, we engaged with Eleonore Stump's relational alternative and Oliver Crisp's concerns about forensic fiction, showing that while both raise important questions, neither succeeds in undermining the philosophical coherence of PSA.

What emerges from this analysis is a picture of penal substitution that is far more philosophically robust than its critics typically acknowledge. The common caricature—that PSA is a crude theory dreamed up by people who never thought carefully about justice—is simply false. The doctrine has been refined through centuries of interaction with the most sophisticated philosophical objections available, and at each point, the defender of PSA has not one but multiple avenues of response. This does not mean that every philosophical question has been settled with finality. Philosophy rarely works that way. But it does mean that the charge of "incoherence" is vastly overstated. Penal substitution is a live and defensible philosophical option, not a doctrine that collapses under scrutiny.

I also want to address a concern that some readers may be feeling at this point. "All this philosophy is fine," someone might say, "but doesn't it reduce the cross to a legal abstraction? Doesn't it drain the blood and the tears and the love out of the gospel and replace them with courtroom arguments?" I understand the worry, but I think it gets things exactly backward. The philosophical work we have done in this chapter does not replace the biblical and theological substance of earlier chapters—it defends that substance against attacks that would undermine it. The cross remains what it has always been: the most staggering act of love in the history of the universe. What our philosophical analysis has shown is that this act of love is not irrational or incoherent. It makes sense. It holds together. The God who devised it was not confused about justice or careless about logic. He acted with infinite wisdom and infinite love at the same time.

In fact, I would go further and say that the philosophical coherence of PSA actually enhances our worship. When we understand that the cross was not a crude transaction but a breathtakingly sophisticated act of the sovereign God—one that satisfies every demand of justice while simultaneously expressing the deepest possible love—our hearts should be moved to even greater adoration. The God who saves us is not a God who cuts corners. He is a God who leaves nothing undone, who addresses every dimension of the human problem, who refuses to sacrifice justice for the sake of love or love for the sake of justice. He holds both together at the cross, and that is the glory of the gospel.

I want to close with a personal reflection. Some readers may find all this philosophical argumentation dry or disconnected from the living reality of the gospel. I understand that feeling. But I believe this work is deeply important. When a thoughtful Christian hears a professor or an author claim that penal substitution is "incoherent" or "logically impossible" or "a violation of basic justice," it can shake their confidence in the heart of their faith. I want readers of this book to know that they do not have to choose between intellectual rigor and a robust theology of the cross. The philosophical objections, while serious, are answerable. Penal substitution is not a doctrine that survives only by avoiding hard questions. It is a doctrine that grows stronger the more carefully we think about it.

And at the end of the day, the philosophical coherence of PSA is not just an academic conclusion. It points us back to the breathtaking beauty of what God has done. The sovereign Judge of the universe, acting in perfect love and perfect justice, took on human flesh and bore in His own person the weight of the world's sin—not because He was forced to, not because an angry deity demanded a victim, but because the Triune God, Father, Son, and Holy Spirit, acting in unified love, determined that this was the way to rescue His beloved creatures from the consequences of their rebellion. That is not incoherent. That is not unjust. That is the gospel. And it is the most beautiful story ever told.

In the chapters that follow, we will continue to build the philosophical case. Chapter 26 will examine the nature of divine justice and retribution in more depth. Chapter 27 will address the specific problem of punishment transfer and moral responsibility. And Chapter 28 will explore the concepts of representation, federal headship, and corporate solidarity that make the "transfer" of sin and righteousness between Christ and His people intelligible. Together, these chapters will provide a comprehensive philosophical framework for the doctrine we have defended biblically, historically, and theologically in earlier parts of this book.