Chapter 25
The Coherence of Penal Substitution — A Philosophical Defense

Introduction: Can Penal Substitution Survive Philosophical Scrutiny?

We have spent a good deal of this book examining the biblical evidence for penal substitutionary atonement, tracing its historical development through the centuries, and exploring how it fits together with other atonement models like Christus Victor, recapitulation, and moral influence. But there is another kind of challenge we need to face — a philosophical one. Critics of penal substitution have argued that the doctrine is not just wrong but actually incoherent. That is, they claim it does not even make logical sense. In their view, penal substitution involves contradictions so deep that no amount of biblical evidence can save it, because a doctrine that contradicts itself cannot be true no matter how many proof-texts you line up in its favor.

These are serious charges, and we need to take them seriously. In this chapter, I want to show that penal substitutionary atonement is a philosophically coherent doctrine that can withstand the most rigorous objections raised against it. The common charges of logical incoherence, moral impossibility, and injustice can all be answered with careful reasoning. We will not be leaving the Bible behind — far from it. But we will be putting on our thinking caps and engaging with the kind of philosophical arguments that critics have leveled at the doctrine, especially in recent decades.

This chapter draws heavily from William Lane Craig's outstanding philosophical treatment in Part III of his Atonement and the Death of Christ, particularly his chapters on the coherence and justification of penal substitution.¹ Craig is arguably the most important philosophical defender of penal substitutionary atonement writing today, and his work has broken significant new ground. We will also interact with the objections raised by Mark Murphy, Eleonore Stump, Oliver Crisp, and others, as well as the criticisms presented by William Hess in Crushing the Great Serpent.² And we will draw on the responses of John Stott, Steve Jeffery, Michael Ovey, and Andrew Sach, who have provided important pastoral and theological responses to these same objections.³

I should say at the outset that this chapter is primarily philosophical in nature. Some readers may find this kind of argument a bit more demanding than previous chapters. But I believe the effort is well worth it. If penal substitution is truly incoherent, then we have a major problem — not just for one atonement model, but for the entire gospel message as the New Testament presents it. On the other hand, if the objections can be answered (and I believe they can), then we emerge with a stronger and more confident understanding of what Christ accomplished on the cross.

Let me preview where we are headed. We will address four major philosophical objections to penal substitution: (1) the claim that substitutionary punishment is logically incoherent — that it is conceptually impossible for one person to bear another's punishment; (2) the claim that punishing an innocent person is always unjust, and therefore God could not justly punish Christ; (3) the so-called "double payment" objection — if Christ already paid the penalty, why must people still exercise faith?; and (4) the "forgiveness negation" objection — if the penalty is fully paid, then there is nothing left to forgive, and we cannot have both payment and forgiveness. I believe each of these objections can be answered, and that when we work through them carefully, the doctrine of penal substitution emerges not weaker but stronger.

Setting the Stage: What Exactly Are We Defending?

Before we dive into the objections, we need to be clear about what we are defending. This matters a great deal, because many philosophical critiques of penal substitution are actually aimed at caricatures of the doctrine rather than at its best formulations. As we argued in Chapter 20, the Trinitarian framework is absolutely essential. We are not defending the idea that an angry Father sadistically tortured an unwilling Son. We are defending the claim that the Triune God, acting in unified love, bore in the person of the Son the judicial consequences that were due to humanity because of sin.

Craig offers a helpful definition that serves as our starting point. He defines penal substitution as "the doctrine that God inflicted upon Christ the suffering that we deserved as the punishment for our sins, as a result of which we no longer deserve punishment."⁴ Notice something important about this definition. It says Christ endured the suffering that we deserved as the punishment for our sins. It leaves open the question of whether Christ himself was, technically speaking, "punished." This is not a dodge — it is a careful philosophical distinction that turns out to be very important, as we will see.

Some defenders of penal substitution, including John Stott, have cautioned against saying that God "punished" Christ. Stott wrote that "we must never make Christ the object of God's punishment."⁵ What Stott meant is that we should not think of the Father and the Son as pitted against each other. Rather, as Stott so memorably argued, the cross is the "self-substitution of God" — God Himself, in the person of His Son, bore the penalty that God Himself, as holy Judge, had rightly declared against sin.⁶ The Father did not inflict something on an unwilling victim. The Son willingly accepted what was due to us, and the Father was present with Him in love throughout the ordeal. As we showed in Chapter 20, the entire Trinity acted in unified, self-giving love at the cross.

Similarly, Jeffery, Ovey, and Sach define penal substitution carefully: "The doctrine of penal substitution states that God gave himself in the person of his Son to suffer instead of us the death, punishment, and curse due to fallen humanity as the penalty for sin."⁷ Notice again: God gave Himself. This is not a third party being dragged in against his will. It is God acting sacrificially on our behalf.

With this clear understanding of what we are defending, let us turn to the objections.

Objection 1: The Alleged Incoherence of Substitutionary Punishment

The first objection does not claim that penal substitution is immoral but that it is logically impossible. In other words, the charge is that one person simply cannot bear another person's punishment — not that it would be wrong to do so, but that the very idea is a contradiction in terms.

The most influential version of this argument comes from the philosopher Mark Murphy. Murphy draws on what is called the "expressivist" theory of punishment — the idea, made popular by legal philosopher Joel Feinberg, that punishment must include an expression of condemnation or censure toward the person being punished.⁸ On this theory, harsh treatment only counts as "punishment" if it carries a message of disapproval directed at the person receiving it.

Murphy's argument goes like this: If punishment necessarily involves condemning the person punished, then God cannot punish Christ, because Christ did nothing wrong. God could not condemn Christ, because Christ is sinless. And if God cannot condemn Christ, then any harsh treatment God inflicts on Christ would not count as punishment. Therefore, penal substitution is not just wrong — it is conceptually impossible. Murphy states flatly: "The classic penal substitution view of the Atonement is incoherent."⁹

This is a clever argument, but I believe it fails for several reasons. Let me walk through the responses.

Response 1: Penal Substitution without "Punishment" in the Technical Sense

The first response is surprisingly straightforward. As Craig points out, a defender of penal substitution who holds that God did not technically "punish" Christ — in the narrow expressivist sense of that word — is completely unfazed by Murphy's argument.¹⁰ If Murphy wants to define "punishment" so narrowly that it must include condemnation of the person being punished, fine. We can simply use a different word. Feinberg himself distinguished between punishments (which express condemnation) and penalties (which do not necessarily carry condemnation — like parking tickets, sports penalties, or workplace firings). Borrowing this distinction, the defender of penal substitution can say that God penalized Christ for our sins — that Christ paid the penalty for our sins — without any conceptual problem.

Think about it this way. If someone pays a fine on your behalf, that person is bearing a penalty that was due to you, even though no one is "condemning" the person who pays. The penalty is real. The payment is real. The release from obligation is real. The word "punishment" in the narrow expressivist sense may not apply, but the substance of what penal substitution claims — that Christ bore the consequences that were due to us — remains perfectly intact.

Craig rightly identifies what Murphy is doing as a "definitional stop" — a maneuver that the eminent legal theorist H. L. A. Hart warned about long ago.¹¹ Murphy has simply defined "punishment" in such a narrow way that substitutionary punishment is excluded by definition. But this is just a semantic trick. It does not tell us anything about whether Christ actually bore the consequences of our sin. The philosopher David Lewis put it well: even if we grant the expressivist's definition, we can still speak of Christ undergoing "something that would have constituted punishment if it had happened instead to the guilty offender."¹² The discussion of whether such treatment is just may then proceed on its own terms.

Response 2: Punishment without Expressivism

But suppose we do not want to concede the expressivist definition. Suppose we want to maintain that God genuinely punished Christ for our sins. Is there a way to do this?

Yes. Hart himself argued that we should recognize "secondary" forms of punishment — including punishment of persons who are not themselves offenders.¹³ On Hart's broader view, substitutionary punishment is a legitimate kind of punishment, and the expression of condemnation is not a necessary condition for something to count as punishment.

Craig shows that the expressivist theory actually does not "carve the cases properly," as Murphy claims it does.¹⁴ In civil law, penalties for torts like assault, fraud, and wrongful death often do carry strong disapproval and censure — yet expressivists classify them as mere "penalties" rather than "punishments." Conversely, there are actual crimes — so-called mala prohibita (things that are wrong only because the law says so, not because they are inherently evil) — where punishment does not seem to carry moral condemnation at all. Think of a minor regulatory violation. The person is punished, but no one is expressing deep moral censure.

Even more telling are cases of "strict liability" in criminal law. In these cases, a person can be found guilty and punished even though they had no guilty mind — no mens rea — at all. Craig cites the case of a pharmacist who sold drugs based on what turned out to be a forged prescription. The pharmacist acted in good faith, with no knowledge of the forgery, and was not negligent in any way. Yet he was convicted and punished under strict liability laws.¹⁵ These are not exotic edge cases. There are thousands of statutory offenses involving strict liability — crimes involving possession of controlled substances, firearms, mislabeled food, and more. In all these cases, punishment happens without any expression of moral condemnation of the person punished.

If our own legal system recognizes punishment without condemnation, then the expressivist theory is simply too narrow. And if it is too narrow, then Murphy's argument — which depends entirely on expressivism — collapses.

Response 3: Expressivism with Condemnation Directed at Sin, Not at Christ

There is a third possibility as well. Even if we accept the expressivist view that punishment must express condemnation, we can argue that God's treatment of Christ on the cross does express condemnation — but the condemnation is directed at sin, not at Christ personally. God's harsh treatment of Christ expresses His absolute opposition to human sin and rebellion. The condemnation falls on the sin that Christ bears, not on the person of Christ Himself.

This may sound like a fine distinction, but it makes perfect sense within the doctrine of imputation that we will discuss more fully in Chapter 28. When our sins are imputed to Christ — that is, when they are legally reckoned as His — the condemnation that falls on those sins falls on Him as their bearer, without implying that Christ is personally sinful. The Bible captures this with remarkable precision: "For our sake he made him to be sin who knew no sin, so that in him we might become the righteousness of God" (2 Corinthians 5:21, ESV). Christ is "made sin" in a legal or representational sense, while remaining personally sinless. The condemnation expressed in the punishment is directed at the sin He carries, and in that sense, even expressivism is satisfied.

Summary: The Incoherence Objection Fails

We have now seen three distinct responses to the incoherence objection, any one of which is sufficient to defeat it. First, the penal substitution theorist can simply distinguish between "punishment" and "penalty" and note that Christ bore the penalty for our sin. Second, the expressivist theory of punishment is too narrow and does not accurately describe how punishment actually works in our legal systems. Third, even on expressivist terms, the condemnation expressed in Christ's suffering is directed at the sin He bears, not at His person. Murphy's objection, for all its philosophical sophistication, does not succeed in showing that penal substitution is incoherent.¹⁶

Objection 2: The Injustice of Punishing the Innocent

We now turn to the objection that critics consider the most devastating. Even if substitutionary punishment is logically coherent, is it just? This objection goes back at least to Faustus Socinus in the sixteenth century, and it remains the most common philosophical complaint against penal substitution today.

The argument is straightforward. It is an axiom of retributive justice — the idea that people should be punished because they deserve it — that it is unjust to punish an innocent person. Christ was an innocent person. Therefore, God could not justly punish Christ. And if God could not justly punish Christ, penal substitution is false.

Craig lays out the argument in formal terms:

(1) God is perfectly just.
(2) If God is perfectly just, He cannot punish an innocent person.
(3) Therefore, God cannot punish an innocent person.
(4) Christ was an innocent person.
(5) Therefore, God cannot punish Christ.
(6) If God cannot punish Christ, penal substitution is false.¹⁷

Tom Smail pressed the same objection. He insisted that "punishing an innocent man — even a willing victim — is fundamentally unjust."¹⁸ Stuart Murray Williams agreed, arguing that guilt and punishment are inseparably attached to the person who incurred them and simply cannot be transferred to another.¹⁹ William Hess, in Crushing the Great Serpent, raises the same concern when he emphasizes that on the PSA model, "the holy, righteous, and perfect judge chose to punish the innocent and guiltless incarnate Son" and asks how this can be called justice.²⁰

These are important questions. Let me respond with several lines of argument.

Response 1: The Version of PSA That Does Not Require "Punishing" Christ

First, as Craig points out, this objection has "no purchase" against versions of penal substitution that hold God did not technically punish Christ but rather that Christ voluntarily took upon himself the suffering that would have been our punishment.²¹ On this view, Christ chose to pay a penalty on our behalf — much as our legal system allows one person to pay another's fine without any moral objection. If Christ was not "punished" in the strict sense, then the argument that it is unjust to punish an innocent person simply does not apply.

This is not a retreat from penal substitution. It is a clarification of what the doctrine actually claims. Christ bore the consequences of our sin. He endured the suffering that justice demanded. He satisfied the demands of divine justice on our behalf. Whether we call this "punishment" or "penalty-bearing" or "satisfaction," the substance is the same.

Response 2: God's Unique Prerogatives as Lawgiver, Judge, and Ruler

But suppose we do affirm that God punished Christ. Is this unjust? The answer depends on how we understand God's relationship to moral law and justice. Craig makes a powerful point here that deserves careful attention.

The Protestant defenders of penal substitution were advocates of Divine Command Theory (or something like it) — the view that moral duties are constituted by God's commands. On this view, God does not issue commands to Himself. He has no moral duties in the way that we do, though He always acts in accordance with His perfect nature. Since there is no external law hanging over God that He must obey, He has unique prerogatives that we do not have. He is free to act in any way consistent with His nature — and what could be more consistent with His nature of holy love than to bear in Himself, in the person of His Son, the cost of human sin?²²

Hugo Grotius made this point centuries ago. Even if God has established a system of justice among human beings that forbids the punishment of the innocent, He Himself is not so forbidden. God refused Moses' offer of himself as a substitutionary sacrifice (Exodus 32:30–34). But if God wills to take on human nature and give His own life as a sacrifice for sin, who is to forbid Him? He is the Lawgiver. He is the Judge. He is the Ruler. In our Western systems of government, the legislative, judicial, and executive functions are separated. But God embodies all three in one — and He is therefore in a unique position to determine how justice is satisfied.²³

This is not the same as saying "might makes right." It is saying that God, as the source and standard of all justice, has prerogatives that no creature has. When the Lawgiver Himself steps in to bear the penalty that His own law demands — voluntarily, out of love — the objection from injustice simply does not have the same force it would if a human judge tried to punish some random innocent bystander.

Response 3: Prima Facie vs. Ultima Facie Justice

Craig makes another important philosophical distinction that helps enormously. He distinguishes between what philosophers call prima facie obligations and ultima facie obligations.²⁴ Let me explain these terms in plain language.

A prima facie obligation (Latin for "at first glance") is something that is generally required but that can be overridden by weightier moral considerations in specific circumstances. An ultima facie obligation (literally, "at the final face") is what you are truly, all-things-considered, obligated to do.

Here is an everyday example. It is generally wrong to break a promise. That is a prima facie obligation. But if you promised to meet a friend for coffee and you come upon a car accident on the way where someone desperately needs help, you are morally right to break your promise and help the injured person. Keeping your promise was prima facie required, but helping the injured person was the ultima facie right thing to do. The weightier obligation overrode the lesser one.

Now apply this to penal substitution. Even if God's essential justice includes what philosophers call "negative retributivism" — the principle that the innocent should not be punished — this is a prima facie demand that can be overridden by weightier moral considerations. And what considerations could be weightier than the salvation of the entire human race? Even the staunchest contemporary retributivist, Michael Moore, acknowledges that retributive justice makes prima facie demands that can be overridden in extreme cases. Moore calls himself a "threshold deontologist" — someone who abides by the rules of retributive justice until doing so would produce consequences so catastrophic that the rules must yield.²⁵

If Moore, who is not a theologian and has no stake in the atonement debate, recognizes that the demands of retributive justice can be overridden in extreme circumstances, then the penal substitution theorist can make the same claim with even greater force. God, by waiving the prima facie demands of negative retributive justice and bearing in Christ the suffering that we deserved, has mercifully saved the world from destruction. That is a morally sufficient reason — indeed, an overwhelmingly compelling reason — for overriding the general principle that the innocent should not be punished.

Response 4: The Doctrine of Imputation

But there is an even stronger response available. The Protestant Reformers who formulated penal substitution did not simply say that God punished an innocent person. They taught the doctrine of imputation — the idea that our sins were legally reckoned to Christ's account, so that He became, in a legal sense, guilty before God.

Now, I want to be very careful here. This does not mean that Christ became personally sinful. He remained, as always, perfectly virtuous and holy. As 2 Corinthians 5:21 says, He "knew no sin." But our sins were imputed to Him — credited to His account, as it were — so that He bore the legal guilt of those sins. Because of this imputation, the punishment that fell on Christ was not punishment of an innocent person at all. It was punishment of one who, in the legal and covenantal sense, bore the guilt of others.²⁶

Craig points out that Murphy objects to the doctrine of imputation on the grounds that "we have no experience" of the transfer of moral responsibility or guilt from one person to another.²⁷ But why should we expect to find analogies in human experience for what is a uniquely divine act? Imputation is a divine prerogative. God, who is Lawgiver, Judge, and Ruler all in one, has the authority to impute guilt in a way that no human court does. So the absence of an exact human analogy is hardly surprising.

And as it turns out, we are not as "utterly bereft" of analogies as Murphy suggests. Craig identifies two important legal parallels: legal fictions and vicarious liability. Let us consider each briefly.

Legal fictions. A legal fiction is something that a court consciously knows to be false but treats as if it were true for the sake of achieving a just outcome.²⁸ Legal fictions are a long-established, widespread, and indispensable feature of our legal systems. When we say that our sins were "imputed" to Christ, we are saying that God, as supreme Judge, adopted for the purposes of our redemption the legal fiction that Christ had done our deeds. This does not make our redemption "fictional" — Christ's suffering was real, His satisfaction of justice was real, our forgiveness is real. What is fictional is merely the attribution of our specific wrongful acts to Christ. And legal fictions, as scholars of jurisprudence have long recognized, produce profoundly real results in the world.

Vicarious liability. In civil and criminal law, there is a well-established principle called respondeat superior ("the master is answerable"), by which the liability of a subordinate can be imputed to a superior.²⁹ An employer can be held legally liable — even criminally liable — for acts done by an employee, even though the employer did not personally commit those acts and may have been entirely blameless. This is not the transfer of guilt but its replication — the guilt of the employee is replicated in the employer by virtue of their relationship. Similarly, Craig argues, our guilt is replicated in Christ by virtue of His relationship with us as our covenant representative — not transferred from us (we remain guilty until the penalty is borne) but replicated in Him so that He bears it as our representative head.

As Jeffery, Ovey, and Sach argue, the doctrine of union with Christ (explored in depth in Chapter 28) is the key to understanding why imputation is not unjust. Christ is not a random third party hauled in against his will. He is united to His people by a real, spiritual bond. Our sins are imputed to Him not as a mere legal trick but because of a genuine covenantal and spiritual union between Christ and those He represents. The imputation flows from the union — and the union is real.³⁰

Response 5: The Devastating Consequences of Denying Imputation

Jeffery, Ovey, and Sach make a further point that is worth pausing over. If we deny that our guilt can be imputed to Christ, we face devastating theological consequences.³¹

First, if our guilt was not imputed to Christ, then the very fact that He died at all becomes an appalling injustice. The Bible teaches that death is the penalty for sin (Romans 6:23). Christ had no sin of His own. If He bore the guilt of no one else, then He should not have died. And since God is sovereign over all of history — including the death of Christ — we would be forced to hold God responsible for this injustice. The only way to explain how Christ could have died without God being unjust is to say that our sin and guilt were imputed to Him.

Second, if our guilt has not been imputed to Christ, it remains on us. We are still guilty. God would then have to either condemn all of humanity without exception or simply ignore our sin in order to save some of us. But both options are unbiblical. The first contradicts the Bible's clear teaching that some will be saved. The second violates God's justice and holiness — He cannot simply sweep sin under the rug.

Third — and this is the clincher — any objection to the imputation of guilt applies equally to the imputation of righteousness. If our sin cannot be imputed to Christ, then by the same logic, His righteousness cannot be imputed to us. But this is the doctrine of justification by faith — the teaching that we are counted righteous before God on the basis of Christ's righteousness, reckoned to our account. If you pull the thread on imputation, the whole fabric of salvation unravels. As Paul declares: "There is therefore now no condemnation for those who are in Christ Jesus" (Romans 8:1, ESV). But this promise holds only because of the great exchange: He took our sin; we received His righteousness.

Response 6: The Witness of Stott on Self-Substitution

I want to add one final line of response that I find personally very compelling. John Stott argued in The Cross of Christ that the key to understanding the justice of penal substitution is the concept of the self-substitution of God.³² When we speak of God "punishing Christ," we are not speaking of one party inflicting suffering on another. We are speaking of God Himself — the one against whom all sin is ultimately committed (Psalm 51:4) — stepping in to bear the consequences of that sin in His own person.

Stott makes the point powerfully. He notes that P. T. Forsyth wrote: "He must either inflict punishment or assume it. And he chose the latter course, as honouring the law while saving the guilty. He took his own judgment."³³ This is the central insight. The penalty was not inflicted by one party on a different, unwilling party. It was assumed by the Lawgiver Himself. God bore His own judgment. The one who declared the sentence is the one who served it — in the person of His Son, who is God from God, light from light, true God from true God, as the Nicene Creed confesses.

When we understand the cross in this way, the objection from injustice dissolves. It would be unjust for a human judge to grab an innocent bystander off the street and punish them for someone else's crime. But it is an entirely different thing for the Judge Himself — who is also the Lawgiver and the offended party — to step down from the bench and bear the sentence in His own person. That is not injustice. That is the most astonishing act of justice and love the world has ever seen.

Objection 3: The "Double Payment" Problem

We come now to a different kind of objection — one that accepts the basic framework of penal substitution but argues that it leads to absurd consequences. The "double payment" objection goes like this:

If Christ paid the full penalty for sin, then the debt is paid. Justice is satisfied. But if the debt is already paid, how can God still require anything further from sinners — like faith, or repentance? You do not owe a debt that has already been paid. If Christ's payment was truly sufficient, then everyone should be automatically saved, with no conditions whatsoever. And if not everyone is saved, then either Christ's payment was insufficient or God is demanding double payment — punishing Christ and then also punishing those who reject Christ — which would be grossly unjust.

This objection is clever and has troubled many thoughtful Christians. Jeffery, Ovey, and Sach note that even an "unjust judge could not demand payment for the same sins twice."³⁴ So how do we respond?

The key is to distinguish between the objective accomplishment of the atonement and its subjective appropriation. Think of it this way. Christ's death on the cross accomplished everything that is necessary for anyone and everyone to be saved. The penalty has been paid. The price has been rendered. The satisfaction of divine justice has been achieved. In that sense, the work is complete and sufficient for the entire human race — which is why we affirmed in Chapter 30 that Christ died for all people without exception.

But the benefits of this objective accomplishment must be received. They do not apply automatically. A pardon may be issued, but it must be accepted. A gift may be offered, but it must be received. Faith is the means by which the individual lays hold of what Christ has accomplished and makes it his or her own. This is not a "second payment." It is the channel through which the one sufficient payment flows to the individual sinner.

Consider an analogy from everyday life. Suppose a wealthy benefactor pays off the entire student loan debt of every student at a particular university. The payment is made, in full, to the lending institution. The debt is objectively canceled. But each student still needs to accept the gift — to sign the paperwork, to acknowledge the benefactor's payment. If a student refuses to accept the gift, his debt is not paid a second time. The payment has been made, but the student's refusal means he never personally benefits from it. The benefactor is not demanding "double payment." He is simply recognizing that a gift, by its nature, must be received.

Similarly, Christ has paid the price for the sins of the whole world (1 John 2:2). But God, in His wisdom, has ordained that this payment is appropriated — received, applied, made effective for the individual — through the instrument of faith. Faith does not add anything to Christ's work. It simply receives what Christ has already accomplished. As Paul says, we are "justified by his grace as a gift, through the redemption that is in Christ Jesus, whom God put forward as a propitiation by his blood, to be received by faith" (Romans 3:24–25, ESV). Notice: the propitiation is "by his blood" — that is the objective accomplishment. It is "received by faith" — that is the subjective appropriation. Both are necessary, but they are distinct.

The double payment objection, then, rests on a confusion between two different things: the sufficiency of Christ's payment (which is complete) and the application of that payment to individuals (which requires faith). There is no injustice in requiring faith as the instrument of reception. No one is being "charged twice." Those who perish do so not because Christ's payment was insufficient but because they refused to receive it.

Objection 4: The "Forgiveness Negation" Problem

Closely related to the double payment objection is what we might call the "forgiveness negation" objection. It goes like this:

If the penalty for sin has been fully paid by Christ, then there is nothing left to forgive. You can have payment or forgiveness, but not both. If I owe you a thousand dollars and a third party pays my debt in full, you have not "forgiven" me anything — you have simply been paid. So if God required full payment for sin (through Christ's death), then God has not actually forgiven anyone. He has simply collected what was owed. But the Bible clearly teaches that God forgives our sins. Therefore, either the payment model is wrong, or forgiveness is an illusion.

Hess raises a version of this concern in Crushing the Great Serpent when he argues that if God must receive satisfaction before He can forgive, then the forgiveness is not truly free. As Hess puts it, Anselm and the tradition that followed him claimed that "it is wrong for God to just freely forgive without some form of recompense." But, Hess objects, "the whole point of being freely forgiven is so one does not inherit a punishment."³⁵

This is a genuinely interesting philosophical puzzle. Let me offer several lines of response.

Response 1: Distinguish Between Forgiveness as an Attitude and Forgiveness as a Legal Act

Craig's response to this objection is philosophically rigorous and very helpful.³⁶ The key is to recognize that the word "forgiveness" is being used in two different senses, and the objection works only if we confuse them.

In one sense, forgiveness is an attitude — a willingness to let go of resentment, a desire for reconciliation, a posture of grace toward the offender. In this sense, God forgives freely and unconditionally. God's love and desire for reconciliation with sinners does not depend on Christ's death. God loved us "while we were still sinners" (Romans 5:8). He has always willed our good and desired union with us. This is forgiveness in the personal or attitudinal sense — and it requires no prior payment.

But in another sense, forgiveness is a legal act — a pardon, a declaration that the penalty has been dealt with and the offender is released from liability. This is forgiveness in the judicial sense. And this kind of forgiveness does require a basis. A judge cannot simply wave away a convicted criminal's sentence because he feels warm and fuzzy toward the person. A judge has a responsibility to uphold the rule of law. If the sentence is to be set aside, there must be a legal basis for doing so — a completed sentence, a legitimate pardon, or (in our case) the satisfaction of justice through substitution.

God is not merely a private party in a personal dispute. As Craig emphasizes, drawing on Grotius, God is Judge and Ruler of the moral universe.³⁷ In His personal love, He forgives freely. But in His role as Judge, He must have a just basis for releasing sinners from their liability. Christ's death provides that basis. The penalty has been satisfied — not because God needed to be "paid off" like a greedy creditor but because the moral order of the universe requires that sin be dealt with seriously. God's justice and God's love are both satisfied at the cross.

This is why Paul can write with such confidence: "And you, who were dead in your trespasses ... God made alive together with him, having forgiven us all our trespasses, by canceling the record of debt that stood against us with its legal demands. This he set aside, nailing it to the cross" (Colossians 2:13–14, ESV). The "record of debt" with its "legal demands" has been cancelled. That is the judicial dimension. God has "forgiven" us. That is the personal and relational dimension. Both are real. Both are accomplished at the cross. They are not contradictory — they are complementary.

Response 2: God as Judge vs. God as Private Party

Craig identifies the fundamental error in Eleonore Stump's version of this objection: "Her entire approach to the doctrine of the atonement is based on construing God on the analogy of a private party involved in various personal relationships rather than as a Judge and Ruler."³⁸ Stump frequently compares God to a friend dealing with a personal offense. But this comparison, while illuminating in some respects, breaks down fatally when it comes to the atonement. God is not merely an offended friend. He is the sovereign Lawgiver and Judge before whom the whole world stands accountable (Romans 3:19).

A private individual can choose to forgive an offense without any satisfaction, because no principle of public justice is at stake. If my friend insults me, I can forgive freely without any "payment." But a judge cannot acquit a convicted criminal merely because the judge likes the person. The judge has a duty to uphold the law. Mercy, for a judge, must be exercised within the framework of justice, not against it.

God, as Stott so helpfully explains, embodies both roles. He is the offended party who loves us and desires reconciliation (the personal dimension). And He is the Judge who must uphold the moral order of creation (the judicial dimension). The cross is where both roles converge. In His personal love, God freely forgives. In His judicial capacity, God provides the satisfaction that His own justice demands. "He himself bore the penalty that he himself inflicted," as Stott puts it.³⁹ The result is that mercy and justice "kiss" (Psalm 85:10) — they are reconciled, not set against each other.

Response 3: Answering Hess on Free Forgiveness

Hess's version of this objection deserves a specific response. Hess argues that the idea that God must be satisfied before He can forgive is fundamentally contrary to the teaching of Jesus Himself, who "rebuked anyone who demanded payment for forgiveness."⁴⁰ Hess points to the parable of the unforgiving servant (Matthew 18:23–35) as evidence that God forgives freely and expects us to do the same.

Hess raises a legitimate concern about certain caricatures of penal substitution. If we picture God as a petty tyrant who cannot forgive unless someone is made to suffer, we have badly distorted the doctrine. And I agree with Hess that some popular presentations of PSA have made precisely this mistake. But the best formulations of penal substitution do not say that God is unable to forgive. They say that God, in His wisdom and in accordance with His nature, chose to provide satisfaction for sin through the cross — not because He was incapable of doing otherwise but because doing so most fully expressed His character of holy love.

There is a significant difference between saying "God cannot forgive without payment" and saying "God, as the sovereign Ruler of the moral universe, chose to deal with sin in a way that fully honors both His justice and His love." The first makes God sound like a reluctant bureaucrat. The second makes God sound like the wise and loving King He is.

As for the parable of the unforgiving servant — Hess is right that the parable teaches the radical generosity of divine forgiveness. But notice: the parable also shows that the forgiven debt is real. The king absorbs the loss. When the king forgives the servant's enormous debt, the king does not pretend the debt never existed. He bears the cost himself. This is precisely the logic of penal substitution: God forgives our debt by absorbing the cost in Himself, through the cross. The parable, far from undermining penal substitution, actually illustrates its inner logic.

Engaging Eleonore Stump's Alternative

Before we conclude, I want to engage briefly with one of the most sophisticated contemporary alternatives to penal substitution — the proposal offered by Eleonore Stump in her book Atonement.⁴¹ Stump, a distinguished philosopher at Saint Louis University, rejects penal substitution in favor of what she calls a "Franciscan" model of the atonement, rooted in the theology of Thomas Aquinas as interpreted through the lens of personal relationship. On her view, the atonement is not about the satisfaction of justice or the payment of a penalty. It is about the restoration of a broken relationship between God and human beings — a restoration accomplished through Christ's sharing in human suffering, which enables us to overcome our shame and resistance and be reconciled to God.

There is much in Stump's proposal that is valuable. She is right that the atonement is deeply personal and relational. She is right that shame and alienation are important aspects of the human predicament. And she is right that the cross powerfully demonstrates God's solidarity with human suffering. These are genuine dimensions of the atonement that complement the penal substitutionary model, as we argued in Chapters 22 and 24.

But Stump's rejection of penal substitution is, I believe, based on a fundamental mistake that Craig identifies: she treats God as a private party rather than as Judge and Ruler.⁴² On her model, the problem that the atonement solves is purely relational — like two friends who need to be reconciled after a falling-out. But the Bible presents the problem as also legal and judicial. We are not just estranged friends of God. We are condemned criminals who stand guilty before His bar of justice (Romans 3:19). A model that addresses only the relational dimension while ignoring the judicial dimension is incomplete.

Moreover, Stump's definition of forgiveness is thin enough that, as Craig points out, God could both forgive sinners and punish them for their sins — since "forgiveness" on her view simply means willing the good of the offender and desiring union with them.⁴³ But this is deeply counterintuitive. Biblically, people whose sins are forgiven are not punished for those sins. Forgiveness means that the penalty has been dealt with. It is precisely here that penal substitution provides what Stump's model lacks — an account of how God's forgiveness is not just attitudinal but legally effective, removing our actual liability to punishment.

Engaging Oliver Crisp and Contemporary Alternatives

Oliver Crisp, in Approaching the Atonement, offers another important philosophical engagement with penal substitution.⁴⁴ Crisp is sympathetic to the Reformed tradition but carefully probes the internal logic of various atonement theories, including penal substitution, satisfaction, and what he calls "penal non-substitution" (the view that Christ bore the consequences of sin without being a strict substitute). Crisp raises helpful questions about the mechanism of imputation and about whether traditional penal substitution can be distinguished from what he calls a "strong" version (where Christ literally bore the exact punishment owed to each individual sinner) and a "weak" version (where Christ bore suffering that was sufficient to satisfy divine justice without being numerically identical to what each sinner would have endured).

I find Crisp's probing questions illuminating rather than threatening to penal substitution. The "weak" version of penal substitution — where Christ's suffering is sufficient to satisfy divine justice without being numerically identical to what every individual sinner would have endured — is, I believe, both philosophically coherent and theologically preferable. Christ's suffering was not an infinite quantity of pain measured out to match some exact ledger of human sin. It was the voluntary, God-man-endured bearing of the consequences of human rebellion — and because it was the suffering of a divine-human person, it was of infinite worth and sufficient to satisfy the demands of justice for every human being who has ever lived or ever will live. The question is not "Did Christ suffer the exact same amount that I would have suffered in hell?" The question is "Was Christ's sacrifice sufficient to satisfy the demands of divine justice?" And the answer, given the infinite dignity of the person who suffered, is an emphatic yes.

The Kantian Objection: Moral Guilt as Non-Transferable

Behind many of the objections we have considered lies a deeper philosophical assumption that goes back to the great Enlightenment philosopher Immanuel Kant. Kant argued that moral guilt is "the most personal thing" about a human being — more personal than one's property, one's reputation, or even one's body. Guilt is so intimately bound up with the individual moral agent that it simply cannot be transferred to someone else. As Kant put it, moral guilt is not like a financial debt that can be paid by a third party. It is an expression of one's own moral character, and moral character is inalienably one's own.⁵⁷

This Kantian conviction runs deep in modern Western culture. We tend to assume — often without even realizing it — that morality is entirely a matter of individual autonomy and individual responsibility. Each person is a self-contained moral unit. What you do is on you. What I do is on me. No one else can take my place in the moral arena.

There is something right about this, of course. The Bible itself insists on individual moral responsibility: "The soul who sins shall die" (Ezekiel 18:20, ESV). Human courts are rightly forbidden from punishing children for their parents' crimes (Deuteronomy 24:16). And we instinctively recoil from the idea of punishing an innocent person for another's wrongdoing.

But the Kantian picture, taken to its extreme, actually conflicts with important aspects of human experience — and with clear biblical teaching. As Jeffery, Ovey, and Sach observe, the radical individualism of modern Western culture may actually be what makes the transferability of guilt seem so strange to us.⁵⁸ In the biblical world — and in many cultures throughout history — corporate identity and corporate responsibility were deeply embedded in how people understood themselves. The actions of a king affected the whole nation. The sin of Achan brought judgment on all Israel (Joshua 7). The faith of Abraham brought blessing to his descendants. The disobedience of Adam brought death to the entire human race (Romans 5:12). And the obedience of Christ brings life to all who are "in Him" (Romans 5:18–19).

The Adam-Christ parallel of Romans 5 is especially important here, and it receives full treatment in Chapter 28. But even a brief consideration makes the point. Paul's entire argument in Romans 5:12–21 presupposes that the actions of one person can have legal and moral consequences for others who are united to that person. Adam's sin brought condemnation to all. Christ's righteousness brings justification to all who believe. If the Kantian view were correct — if moral guilt and moral merit are absolutely non-transferable — then Paul's argument would be nonsensical. But Paul clearly does not think it is nonsensical. He presents it as the very heart of the gospel.

We should also note that even in modern legal systems, the Kantian principle does not hold absolutely. As we discussed above, vicarious liability allows the guilt of one person to be imputed to another in well-established areas of both civil and criminal law. Corporate liability means that an entire organization can be found guilty for the acts of its employees. And the entire concept of legal representation — where an attorney acts on behalf of a client and the client is bound by the attorney's actions — presupposes that one person's decisions can have binding legal consequences for another. These are not exotic exceptions. They are pervasive features of every functioning legal system.

The Kantian objection, then, proves too much. If taken seriously, it would undermine not only penal substitution but also the doctrine of original sin, the doctrine of justification by faith, and the entire biblical framework of covenantal representation. It would also create problems for well-established features of human law and society. The better conclusion is that the Kantian picture of radical moral individualism, whatever its merits in certain contexts, does not capture the full reality of how moral responsibility works — either in the biblical world or in our own.

The Sufficiency of Christ's Sacrifice: A Philosophical Note

One further philosophical question deserves brief attention before we move on. Some critics have asked: even if penal substitution is coherent in principle, is it coherent in this particular case? How can the finite suffering of one man, even a divine-human person, be equivalent to the infinite punishment that all of humanity supposedly deserves? Is there not a massive quantitative mismatch between what Christ endured (a few hours of suffering on a cross) and what billions of human beings would have endured (eternal condemnation)?

This is a question that theologians have wrestled with for centuries. Hess, drawing on Charles Hodge, notes that some Reformed theologians argued that Christ must have literally "experienced hell" during His crucifixion — suffering not just physical death but the full spiritual agony of eternal separation from God — in order for the payment to be adequate.⁵⁹ I believe this approach, while understandable, pushes things in the wrong direction. It creates serious problems for the doctrine of the Trinity (how can the Son be separated from the Father if they share one divine essence?) and for the coherence of the incarnation itself.

A better approach, and one that I believe is both philosophically sound and theologically rich, focuses not on the quantity of Christ's suffering but on the quality and dignity of the person who suffers. The value of Christ's sacrifice is not measured by counting up units of pain and matching them against units of human guilt, as though atonement were an accounting exercise. The value comes from who is offering the sacrifice. Because Christ is a divine-human person — fully God and fully man, as the Council of Chalcedon confessed — His offering has infinite worth. The eternal Son of God, who upholds the universe by the word of His power (Hebrews 1:3), chose to enter into the depths of human suffering and death. The sheer magnitude of that condescension, the infinite dignity of the one who suffers, is what gives the sacrifice its sufficiency.

Anselm grasped this insight even if his framework was different from ours. He argued that the value of Christ's offering is determined by the worth of the one who makes it, not by the duration or intensity of the suffering per se. Because the person who offers is divine, the offering is of immeasurable worth — more than sufficient to satisfy the demands of justice for any number of human sinners.⁶⁰ Craig, building on this tradition, argues that we need not claim that Christ endured the exact same suffering that each individual sinner would have endured. We need only claim that His sacrifice was sufficient — accepted by God as adequate to satisfy the demands of divine justice. And given the infinite dignity of the person who suffers, this claim is eminently reasonable.

This also explains why we can affirm, as we do throughout this book, that Christ died for all people without exception — not just for the elect. If the value of Christ's sacrifice were determined by summing up the exact punishments owed to a specific number of individuals, then debates about the "extent" of the atonement might reduce to arithmetic. But because the value comes from the infinite worth of the person who offers, the sacrifice is inherently unlimited in its sufficiency. "He is the propitiation for our sins, and not for ours only but also for the sins of the whole world" (1 John 2:2, ESV). There is no philosophical problem with affirming both that Christ's sacrifice was specifically substitutionary (He died "in our place") and that it was universally sufficient (He died "for all").

A Word About Hess's Broader Critique

We have engaged with specific arguments from William Hess throughout this chapter, but I want to address his broader philosophical posture toward penal substitution before we conclude. Hess describes PSA as teaching that "God is primarily characterized as an angry deity, offended by sin" who "demands retribution before He can love" sinners.⁴⁵ He argues that the entire framework of satisfaction and retribution is foreign to the biblical witness and closer to paganism than to the God revealed in Jesus Christ.

I have a good deal of sympathy with Hess's concerns about certain presentations of penal substitution. When popular preachers describe the cross as God "pouring out His wrath" on an unwilling victim, or when they suggest that God "hated" Jesus on the cross, they are presenting a caricature that rightly troubles thoughtful Christians. As I argued in Chapter 20, any formulation of penal substitution that pits the Father against the Son must be rejected as Trinitarian heresy.

But Hess goes too far in rejecting the penal dimension altogether. The problem is that the biblical data demands it. When Isaiah 53:5 says "he was crushed for our iniquities; upon him was the chastisement that brought us peace," the language is unmistakably penal. When Paul writes that God "condemned sin in the flesh" by sending His Son (Romans 8:3), the judicial dimension cannot be removed without violence to the text. When Peter declares that Christ "bore our sins in his body on the tree" (1 Peter 2:24), the substitutionary logic is inescapable. As argued at length in Chapters 6 through 12, the penal and substitutionary dimensions are woven deeply into the fabric of both the Old and New Testaments — too deeply to be dismissed as a Reformation invention. The question is not whether there is a penal dimension to the atonement, but how best to understand it within a Trinitarian, multi-faceted framework.

And that is precisely what we have sought to do throughout this book. Penal substitution, rightly understood, does not depict an angry deity who "cannot" love until someone is made to suffer. It depicts a Triune God who, out of overflowing love, steps into the consequences of human sin in order to absorb them in Himself. The Father does not inflict suffering on an unwilling Son. The Son voluntarily offers Himself. The Spirit sustains Him through the ordeal (Hebrews 9:14). The whole Trinity acts in unified, self-giving love. This is penal substitution at its best — and it is philosophically coherent, morally compelling, and deeply faithful to the biblical witness.

Conclusion: A Doctrine Tested and Vindicated

Philosophical scrutiny is not the enemy of faith. When a doctrine is true, it can withstand rigorous examination — and it often emerges from that examination stronger and clearer than before. I believe that is exactly what has happened with penal substitutionary atonement in this chapter.

We began with four serious objections — objections that, if successful, would have shown that penal substitution is fundamentally flawed at its philosophical foundations. But none of them succeeded. The incoherence objection turned out to rest on an overly narrow definition of "punishment." The injustice objection dissolved once we recognized God's unique prerogatives as Lawgiver and Judge, the doctrine of imputation, and the concept of the self-substitution of God. The double payment objection was resolved by distinguishing between the objective accomplishment and the subjective appropriation of the atonement. And the forgiveness negation objection was answered by distinguishing between God's personal, attitudinal forgiveness and His judicial act of pardon.

At every turn, the doctrine of penal substitution proved to be not only defensible but remarkably rich. Far from being a simplistic or "primitive" theory, it engages with some of the deepest questions in philosophy of law, ethics, and theology — questions about the nature of justice, the relationship between law and grace, the role of representation and imputation, and the character of God as both perfectly just and perfectly loving.

Craig's philosophical work has been particularly groundbreaking in demonstrating that penal substitution is not the intellectually embarrassing doctrine that its critics have often portrayed it as. When we engage with the best thinking in legal philosophy, ethics, and metaphysics, we find that the conceptual resources are available to mount a powerful and persuasive defense. The doctrine does not ask us to check our brains at the door. It invites us to think deeply — and the deeper we think, the more we appreciate the wisdom and beauty of what God accomplished at the cross.

In the chapters that follow, we will continue to explore the philosophical dimensions of the atonement. Chapter 26 will examine divine justice and the moral government of God in greater depth. Chapter 27 will take up the problem of punishment transfer and moral responsibility. Chapter 28 will explore the concepts of representation, federal headship, and corporate solidarity that ground the doctrine of imputation. And Chapter 29 will consider how the atonement is appropriated through faith. Together with the present chapter, these discussions will provide a comprehensive philosophical framework that supports and illuminates the biblical and theological case we have already made.

For now, we can say with confidence: penal substitutionary atonement is not only biblically grounded, historically rooted, and theologically coherent — it is also philosophically defensible. The cross stands firm under scrutiny. And the gospel it proclaims — that God Himself, in the person of His Son, bore the penalty that we deserved, so that we might be forgiven and reconciled to Him — remains the most extraordinary message the world has ever heard.