CHAPTER 7

ATONEMENT AND THE CONCEPT OF PUNISHMENT

DANIEL J. HILL AND JOSEPH JEDWAB

images/himg-21-1.jpg

INTRODUCTION

In a broad sense, atonement refers to the state or condition of being at one or reconciled. In a narrower sense, it refers to the means by which such reconciliation is achieved. In Christian theology, this narrower sense of atonement refers to Christ’s sacrificial act by which such reconciliation between God and humans is achieved. Our sin creates an obstacle to right relationship with God. Christ’s sacrifice is (at least) part of the means whereby God removes this obstacle to such right relationship. There are, however, different theories as to how Christ’s sacrifice does this.

The penal-substitutionary theory of the atonement (PSA1) says that we deserve punishment for our sin, but instead of punishing us, God2 imposes a punishment on Christ for our sin. PSA finds its basis in a number of texts from the Bible. One is Isaiah 53:4 – 5:

Surely he took up our pain and bore our suffering, yet we considered him punished by God, stricken by him, and afflicted. But he was pierced for our transgressions, he was crushed for our iniquities; the punishment that brought us peace was on him, and by his wounds we are healed (NIV).

This text is picked up in the New Testament and explicitly applied to Christ in 1 Peter 2:24 (NIV):

“He himself bore our sins” in his body on the cross, so that we might die to sins and live for righteousness; “by his wounds you have been healed.”

Defenders of PSA have typically said that these texts teach that Christ’s sufferings and death on the cross were laid or imposed on him by God. God may have imposed them on Christ for many different reasons, but, according to the defender of PSA, one reason is penal in nature.

There are many different possible versions of PSA. One line of difference concerns whether Christ substitutes for a group (the church) collectively or for that group distributively, i.e., for each and every member of that group. Another line of difference concerns whether he substitutes (collectively or distributively) for a definite number of individuals, the view known as “definite atonement,” or whether he substitutes (again, collectively or distributively) for an indefinite number of individuals, the view known as “general atonement.” A further line of difference concerns whether Christ takes all the punishment due to those for whom he substitutes or only some of the punishment due to them. Yet another line of difference concerns whether those for whom Christ substitutes are said to be punished “in” Christ (or “in” his punishment), or not.

We shall not comment further on these differences, but now turn our attention to one final line of difference, whether Christ is punished by God or not. Here we find three views: what we shall call “the strong version,” that God did indeed punish Christ; “the weak version,” that God judicially imposed suffering (but in lieu of punishment, rather than as punishment) on Christ;3 and “the intermediate version,” that God imposed punishment on Christ, even though God did not punish Christ. We shall list some of the historical sources for these views in an appendix at the end.

In this paper we propose to defend the strong version of PSA against the conceptual argument as deployed in Mark Murphy’s article “Not Penal Substitution but Vicarious Punishment.”4 The conceptual argument concludes that PSA is conceptually confused, and it derives this conclusion from the concept of punishment itself. While the moral argument may tell equally against all versions of PSA, the conceptual argument does not tell at all against the weak version. And if our defence of the strong version is successful, it will also be a successful defence of the intermediate version.

MURPHYS ARGUMENT FOR THE INCOHERENCE OF STRONG PSA

Murphy starts from the premise that an essential feature of punishing is the “condemning of the agent who failed to live up to the standard, the violation of which justifies the punishment.”5 He also states that “punishment expresses condemnation of the wrongdoer, of the wrongdoer as performer of the wrong.”6 He thinks that it follows from this that “punishment will be non-transferrable”7 and that “penal substitution is unintelligible.”8 But note that there is a jump here: Murphy assumes that one cannot condemn the wrongdoer by punishing someone else.9 And the defender of PSA can deny this assumption.

So the defender of PSA can either (1) deny that it is necessary for punishment that the offender, rather than the offence, be condemned, or (2) deny that it is necessary for punishment that the offender be condemned by way of being punished. It seems to us that (1) is less plausible than (2). It might be responded, however, that a consideration of the larger issue of the nature of punishment might resolve the matter in favour of Murphy. So we shall now turn to his treatment of the nature of punishment in his earlier monograph Philosophy of Law.10

MURPHY ON THE NATURE OF PUNISHMENT

Murphy’s analysis runs as follows, to extract the necessary conditions from his continuous text:

First, it is essential to punishment that, in itself, it is an evil of some sort. . . . Second, a punishment is imposed for the failure to measure up to some binding standard. . . . Third, a punishment is imposed by a personal agency who is authoritative. . . . [Fourth,] whenever punishment is found, it expresses judgments of disapproval and attitudes of indignation and resentment. . . . The object of condemnation must be the person . . . as performer of this act.11

We shall comment on each of Murphy’s conditions in turn.

(A) “An evil of some sort.” Murphy explains that “[p]unishment involves the deprivation of goods or the imposition of bads,”12 so it need not be a positive evil, so to speak, and the evil could be mental or physical. We add that it does not have to be felt as an evil by the victim: the imposition of pain could still be a punishment for someone (like a masochist) who would not think it such. It will follow from the fourth condition that for something to be a punishment, the imposer must think it an evil: it would not be a punishment if someone administered a whipping in the belief that a whipping were a jolly good thing with no bad consequences.

(B) “Punishment is imposed for the [supposed] failure to measure up to some binding standard.” We have added the word supposed here to make clear what Murphy accepts,13 that it is possible to impose punishment mistakenly or falsely — there does not need to be an actual failure on account of which the punishment is imposed. So it is possible to impose punishment on someone for stealing even if in fact no theft occurred, provided that, for example, the one imposing the punishment mistakenly believed the supposed offender was a thief or falsely asserted this.

Next we comment on the word for. This means that if A is imposing the punishment, then A is doing so on account of, or for the reason of, the supposed failure. It would not be a case of punishment if B’s failure to measure up to some binding standard somehow caused the evil to be imposed without B’s failure’s constituting a reason in A’s mind. Suppose that a schoolmaster, A, can punish a pupil at boarding school for going outside after hours by sentencing him or her to stay indoors for a week. Now suppose that a pupil, B, goes outside after hours and gets a chill, and A says that B must stay in the sanatorium for a week in order to recover. In the case described there is no punishment, because the staying inside is not for the breach of norms in going outside after hours, even though there is a causal relation between the two.

Third, Murphy is clear (rightly, we suggest) that “binding standard” applies more widely than just to breaches of morality or breaches of the law. For example, it also applies to a child’s breaking house rules, or, as in our previous example, to a pupil’s breaking school rules.

(C) “Imposed by a personal agency who is authoritative.” Murphy offers an explanation for each of the two main parts of this condition. First, he explains “personal agency”: “The imposition of a punishment is the work of beings who can think and judge; it is not, that is, simply a natural process.”14 We agree with this, and note that the word simply allows for the possibility of God’s punishing people through natural processes. We also think that the condition does, rightly, allow for courts and states to impose punishment too.

Next, Murphy explains “who is authoritative”: “Punishment is an activity by an authority, whether that authority is genuine or de facto only . . .: a private person can carry out vigilante justice, but he or she cannot punish.”15 The first thing to note here is that something can be a punishment even if it is imposed by a self-appointed authority like a Mafia boss, whose authority is not recognized beyond his or her immediate sphere of influence. That said, we also agree with Murphy that vigilante justice is not the same as punishment, and that an injured party merely executing revenge for the failure to measure up to some binding standard is not imposing punishment.

Moreover, this point holds even if the injured party executing revenge is in fact authoritative: a judge is authoritative, but that does not license him or her to impose evils outside the courtroom. So it is not sufficient that the person imposing the evil be authoritative; the person has to be acting authoritatively in imposing the evil.

(D) “Expresses judgments of disapproval and attitudes of indignation and resentment.” Murphy gives three examples in which, he says, the first three conditions are met without there being punishment. The point is that this, if correct, shows that a fourth condition is needed. The clearest example is probably that of tort remedies, such as the award of damages. Damages are evils, and they are authoritatively imposed by personal agencies (e.g., judges) for the (supposed) failure to measure up to some binding standard. Nevertheless, awards of damages and other tort remedies are, intuitively, not punishments; punishments belong to criminal courts, not civil ones. And we agree with Murphy that the reason why they are not punishments is that they do not in and of themselves express judgments of disapproval.16

We agree with Murphy that, in the light of this and other cases, the notion of punishment needs a fourth condition. We are, however, less happy with his precise formulation of the fourth condition. In particular, it does not seem to us that indignation and resentment are necessary features of punishment at all. Nevertheless, it does seem to us that disapproval or condemnation of the wrongdoer as performer of the wrong is indeed a necessary ingredient of punishment, as Murphy says.

MURPHYS ANALYSIS APPLIED TO CHRISTS SUFFERINGS AND DEATH

Could Christ’s sufferings and death on the cross qualify, on Murphy’s definition, as a punishment imposed by God? Yes. The defender of PSA can say that Christ’s sufferings and death on the cross constituted an evil imposed by God for the failure to measure up to some binding standard, and that they expressed a judgment of disapproval and condemnation. But, it will be objected, Christ did not fail to measure up to any binding standard of God’s, and, further, God did not condemn him. Indeed not, but it is not a feature of Murphy’s definition that the one punished has to be the one who failed to measure up to the binding standard, nor does his definition specify that the one punished has to be the one condemned. Even Murphy’s gloss, “The object of condemnation must be the person . . . as performer of this act,”17 does not specify that the one punished is the object of condemnation. Of course, this could be added as a fifth condition, but in the absence of argument, this would simply beg the question.

At this point we may be asked whether the defender of PSA has any non-question-begging argument of his or her own for the contention that Christ could have undergone our punishment. It seems to us that there is a relatively common practice to which the defender of PSA could appeal: the practice of imposing suffering on every member of a group, or a representative of a group, for an offence committed by only some of the group. This is a form of punishment. (Whether or not it is just is not to the present point.) For example, sometimes schoolteachers impose suffering (e.g., a detention) on every member of a class for an offence committed by only some of the class.

This practice is sometimes, but not always, motivated by ignorance as to the real culprits. Nor is it the case that the reason for the whole class’s being detained is that their not preventing the culprits from committing the offence is considered itself an offence worthy of punishment. No, the idea is often that the solidarity of the class is such that it can be treated as a unit for purposes of punishment and reward, even if only some members performed the action punished or rewarded. The practice is perhaps more frequent with rewards — e.g., the granting of a half-day holiday to a school if one pupil distinguishes himself or herself — than it is with punishments, but, as Murphy remarks,18 the principle is the same. For a similar example, a teacher might accept the offer from the class president (form captain) to accept responsibility for the class by taking the punishment due to the whole class.

A second kind of example can be found in the armed forces. Here again it is common for a sergeant to impose suffering on every soldier in the squad because of the actions of only some of its members, and — rather more rarely — to give out good treatment to every soldier in the squad because of the actions of only some of its members. For a similar example, a sergeant might accept the team leader’s offer to take the punishment due to the whole fire team. It seems to us that there is good reason to count these as punishments or rewards (again, whether just or not is not to the present point).

These examples seem to us to show that it is not a necessary truth that if A imposes punishment on B, then the punishment thereby condemns B for a supposed failure on B’s part to measure up to a binding standard. And if this is not a necessary truth, then it is hard to see on what grounds PSA could be ruled conceptually incoherent.19

WHAT IS IT TO PUNISH?

The foregoing analysis, however, has defended only the thesis that it is possible that Christ suffered a punishment for our sin, as asserted by the intermediate version of PSA. It might be objected that this does not show it is possible that God punishes Christ for our sin, as the strong version of PSA asserts. In what follows, we shall build on Murphy’s analysis of punishment, as discussed above.

It seems to us that to punish it is necessary that one either impose punishment (as judges do when they sentence offenders) or implement punishment (as executioners do when they carry out a sentence). It might at first seem as though the broader notion underlying this disjunction was simply the notion of causation, such that the correct definition of punish would boil down to this:

(D0) A punishes B if and only if A causes B to suffer a punishment.

In fact, we do not accept (D0) since the concept of causation seems to cast the net too widely: By the transitivity of causation, anyone who caused the judge to punish the offender (say, by forcing the judge to adopt this course of action) would count as punishing the offender, and that seems to us wrong.

If it is not simply a matter of causation, then what is it to impose a punishment, and what is it to implement a punishment? It seems to us that for A to impose a punishment on B is for A authoritatively to mandate that B suffer a punishment, and for A to implement a punishment on B is for A authoritatively to implement the mandate that B suffer a punishment.20 It is to be noted that we think these notions of imposition and implementation in and of themselves imply that A is a personal agent acting authoritatively; if this is right, then Murphy’s third condition in his definition of punishment will in fact also be covered by this part of the definition of punish.

At first sight it might seem that the disjunction of imposition and implementation is not only necessary for punishing but also sufficient, so that the correct definition of punish is:

(D1) A punishes B if and only if A imposes a punishment on B or A implements a punishment on B.

It seems to us, however, that the first disjunct, “A imposes a punishment on B,” does not suffice for punishing B. This is because, intuitively, one might impose a punishment on B without the mandate’s ever being implemented. For example, B might die before the punishment is executed. Or another might undergo the punishment in B’s stead. Or the court or law enforcement agency might suspend, or simply forget to enforce, the punishment. In these cases it seems clear to us that if the punishment is not implemented on B, then B is not actually punished. (The corresponding point does not hold for “A implements a punishment on B,” since it is in the nature of implementing a punishment that the punishment implemented has been imposed.)

A third attempt at the correct definition of punish might be this:

(D2) A punishes B if and only if A imposes on B a punishment whose mandate is authoritatively implemented or A implements on B a punishment.21

This, however, raises another concern. Suppose A imposes on B the punishment of being hanged in five days’ time, and, because of a delay, B is hanged in six days’ time. It is obvious that B is still punished by A in this situation. Indeed, if B is executed by lethal injection instead of being hanged as A directed, it still seems to us that B is punished by A. The potential problem here is that it might seem that B does not suffer the exact punishment that A imposes. To avoid this problem, we shall add a clause yielding:

(D3) A punishes B if and only if (1) A imposes on B a punishment, x, whose mandate is authoritatively implemented, or A implements on B a punishment, x; and (2) B suffers x or something close enough to x.

Whether a given evil suffered, y, will be close enough to x will depend on the context. It is certainly conceivable that in some contexts, if A sentences B to a horrible mode of death but the executioner actually executes B humanely, then A will not have punished B, since the evil suffered will not be deemed close enough to the evil imposed as sentence.

It might seem as though we need to add a clause specifying that for it to be the case that A punishes B, it must be the case that A intends to impose or implement a punishment on B. Punishing is not something that can be done by accident. We are of the opinion, however, that this is already implicit in the notions of imposition and implementation, such that stating it is unnecessary: one cannot impose or implement a punishment by accident either, since one cannot mandate something by accident or implement a mandate by accident.22 Suppose, for example, that A is an executioner preparing to implement an imposed death penalty on B, but that A accidentally discharges the firearm while preparing it for B’s execution and accidentally kills B. We say that in this situation, A neither punishes B nor implements any punishment on B, since the accidental killing of B does not implement the death penalty.

Of course, it is still possible to impose or implement a punishment on B mistakenly, e.g.. in the wrong belief that one is imposing or implementing it on C (as when the defendant’s identical twin takes his or her place in the dock for sentencing or on the gallows for hanging). But by the same token, it is possible to punish B in the wrong belief that one is punishing C.

The main question under our consideration here is whether it follows from our definition, (D3), together with the previously established definition of punishment drawn from Murphy’s work, that it is necessary, in order for A to punish B, that the punishment condemn B as the one thought to have failed to have measured up to the binding standard. (We have seen already that it is not necessary that B or anyone have actually failed to measure up to a binding standard.) It does not follow. It is compatible with our definition, (D3), that A punish B intending thereby to condemn C (or C’s offence).

Does there, then, have to be any connection at all in A’s mind between the one that failed to measure up to the binding standard, C, and B in order for A to punish B? Suppose A thought that the “breach in the moral fabric of the universe could be shown up, condemned, and repaired” if he or she caused a random passerby to suffer an evil. Could that action count as (unjust) punishing? It seems to us that it could. To take a clearer example: suppose that A believed that anybody genetically related to the doer of the evil were fair game for punishing. We contend that this could indeed be a case of (unjust) punishing. Similarly, if the person causing B to suffer the evil believed that anybody spatially or temporally connected with the doer of the evil were fair game for punishing; then again we contend that this could indeed be a case of (unjust) punishing. And the examples we gave earlier in connection with the imposition of punishments also work for punishing, we think: it is possible for a school teacher to punish all the pupils in the class for a deed that he or she knows was performed by only some of them, and it is possible for a sergeant to punish all the soldiers in a squad for a deed that he or she knows was performed by only some of them. (Whether in these examples the punisher justly punishes is not to the present point.)

Now, is this definition met in the case of God and Christ? It seems to us that defenders of the strong version of PSA above will say yes: on the cross God punishes Christ on account of a failure (by the people collectively or distributively, for whom he substitutes) to measure up to moral standards. Presumably, God both imposes and implements a punishment on Christ. God authoritatively mandates that Christ suffer a punishment, and God authoritatively implements this mandate. And in God’s punishing Christ, the punishment expresses condemnation of such people as wrongdoers. And whether or not there must be any connection in God’s mind between such people and Christ in order for it to be conceptually possible that God punish Christ, God knows that Christ is connected in an appropriately intimate way (by union) with those whose sins are atoned for on the cross.

WHAT ARE THE NECESSARY CONDITIONS FOR PENAL SUBSTITUTION?

So for C to be B’s substitute in being punished by A, it is necessary that A punish C, the conditions for which have been explained above. A particular implication of this is that a penal substitute must be capable of being punished. So if B is due to receive forty lashes and substitutes for himself or herself a waxwork that looks like B, then there is no penal substitution since the waxwork is not capable of being punished. Similarly, if B thinks that he or she will be sentenced to pay a huge fine and substitutes for him- or herself a bankrupt, C, in the dock before the judge, A, then while A may impose a penalty on C, C will not suffer it, and hence not be punished by A, since C is unable to bear it. So there is no penal substitution in this case either, even though it is attempted.23 To apply this point to PSA, it seems to us that Christ is capable of being punished.

A further condition is that somebody has to intend that C substitute for B. Is it necessary, then, that C intend to be a substitute? We think not. If another party, D, subtly removes B from the dock and replaces B with C, then D has substituted C for B, and in consequence, C is B’s substitute, whether C intends it or not. Moreover, it is not even necessary that C know that he or she is a substitute or that he or she is going to be, or is, being punished. Nor is it necessary that B know or intend that he or she is being substituted. Nor is it necessary that the one punishing, A, know or intend that a substitution occur. It is necessary, however, that for there to be a substitution, the person bringing about the swap intend that there be a substitution. If there is a genuine mixup, and A punishes C under the wrong impression that C is B, without anybody intending this outcome, then there is no substitution — just a mistake. Finally, to apply this point to PSA, it seems to us that (at least) God substitutes Christ for us and intends such a substitution.

We have not addressed in this paper the question whether penal substitution is ever just, and, if so, under what circumstances. The defender of PSA will, of course, go further than we have been able to do in this paper, and insist that Christ’s penal substitution for us on the cross shows that, no matter how serious the offence, it can be just for one person to be substituted for another. But the defender of PSA may also insist that the case of the cross is unique, since Christ was unique, and that features of this case cannot necessarily be generalized to merely human punishments. In consequence, the defender of PSA may think that arguments by analogy with the injustice of penal substitution in human judicial systems are not persuasive. There is no space here, however, to consider all the many arguments that have, in fact, been levelled against the justice of PSA.24

HISTORICAL APPENDIX

It has been alleged that PSA has never historically included the assertion that God punished Christ.25 In response to this, we now set out a representative sample of Christian thinkers who have held this view, the strong version:26

(1) Gregory the Great:

But we must consider how He is righteous and ordereth all things righteously, if He condemns Him that deserveth not to be punished. For our Mediator deserved not to be punished for Himself, because He never was guilty of any defilement of sin. But if He had not Himself undertaken a death not due to Him, He would never have freed us from one that was justly due to us. And so whereas “The Father is righteous” in punishing a righteous man, “He ordereth all things righteously,” in that by these means He justifies all things, viz. that for the sake of sinners He condemns Him Who is without sin.27

(2) John Wycliffe: “Just as God justly and righteously punished Christ for the offence of his brothers.”28

(3) Martin Luther: “God himself struck and punished Christ.”29

(4) Johan Piscator: “But he punished us in Christ, or, what comes to the same thing, he punished Christ for us and in our place.”30

(5) Samuel Rutherford: “The Lord punished Christ for us to declare the glory of his Justice in punishing sin in his own Son. . . . He punished Christ, who was not inherently, but only by imputation the sinner, with no hatred at all, but with anger and desire of shewing and exercising revenging justice, but still loving him dearly, as his only Son.”31

(6) John Owen: “The person punishing is Jehovah, the person punished called ‘him,’ — that is, he who is spoken of throughout the whole prophecy, the Messiah, Jesus Christ.”32

(7) Francis Turretin: “Christ is justly punished by God the judge both in body and soul in order to make expiation for sinners.”33

(8) Solomon Stoddard: “God punished Christ with death for sin.”34

(9) Christopher Love: “It may well consist that God punished Christ for our sins.”35

(10) John Downame: “It would not indeede haue stoode with Gods iustice, to haue punished Christ as he was innocent and righteous, nor to haue acquitted and absolued vs who were vnrighteous and wicked; but he punished Christ in respect that he had taken vpon him the sins of all the faithfull.”36

(11) Edward Philips: “Christ and we are ioyned together, for otherwise it had not stood with Gods iustice to haue punished Christ in our flesh, nor to haue accepted our obedience in Christs person, if wee had not beene in him and he in vs.”37

(12) Charles Wesley: “For what you have done, his blood must atone; The Father hath punished for you his dear Son.”38

(13) A. A. Hodge: “Christ is the one satisfied as well as the one satisfying, the one punishing as well as the one punished; but he loves us enough to punish himself in our place.”39

(14) Herman Bavinck: “God condemned sin in his flesh [Rom. 8:3] and punished him with the accursed death on the cross.”40

The clearest historical example of the weak view, that God judicially implemented suffering on Christ but did not punish Christ, is that of Archbishop William Magee of Dublin:

I will not contend, that this should be called suffering the punishment of those sins, because the idea of punishment cannot be abstracted from that of guilt. . . . But it is evident, that it is notwithstanding a judicial infliction; and it may perhaps figuratively be denominated punishment, if thereby be implied a reference to the actual transgressor, and be understood that suffering which was due to the offender himself; and which, if inflicted on him, would then take the name of punishment. In no other sense, can the suffering inflicted on one, on account of the transgressions of another, be called a punishment; and, in this light, the bearing the punishment of another’s sins, is to be understood as bearing that, which in relation to the sins, and to the sinner, admits the name of punishment, but with respect to the individual on whom it is actually inflicted, abstractedly considered, can be viewed but in the light of suffering.41

It is much harder to find a text explicitly and unambiguously declaring the intermediate view that Christ suffered a punishment imposed as a punishment by God, but was not punished by God. This reading would, however, be the face-value interpretation of a number of texts; for example:

He bore the punishment of our transgressions, the punishment needed to bring us peace. . . . Yet we must be very careful in stating the truth of vicarious punishment not to go beyond what is written. Expressions like “God punished Christ” and still more “God was angry with Christ” should not be used.42

1. We shall use “PSA” as an abbreviation for “the doctrine of penal-substitutionary atonement” rather than simply for “penal-substitutionary atonement”

2. When we write “God,” we shall omit the qualification “(or God the Father).”

3. NB that it is necessary that the suffering be judicially inflicted in order for this theory to be a version of PSA.

4. Mark C. Murphy, “Not Penal Substitution but Vicarious Punishment,” Faith and Philosophy 26, no. 3 (July 2009): 253 – 73.

5. Ibid., 257.

6. Ibid., 256.

7. Ibid.

8. Ibid., 257.

9. Murphy also states that “[p]unishing presupposes that the object of punishment has failed in particular respects,” but he provides no argument for the move from “agent who failed to live up to the standard” to “object of punishment.” Ibid.

10. Mark C. Murphy, Philosophy of Law (Oxford: Blackwell, 2007).

11. Ibid., 113 – 16.

12. Ibid., 113.

13. Ibid., 114.

14. Ibid.

15. Ibid.

16. We concede that sometimes disapproval of the action is expressed along with the award of damages or other tort remedy, but we maintain that this disapproval is not part and parcel of the award itself or the damages themselves.

17. Ibid., 116.

18. Murphy, “Not Penal Substitution,” 256 – 57.

19. Of course, this does not address the objection that penal substitution is unjust.

20. It is necessary that the mandate be logically prior to the implementation but not necessary that the mandate be temporally prior to the implementation. After all, it is possible simultaneously to impose and implement a punishment.

21. The phrase “on B” here rules out B’s being punished if, say, B pays the fine imposed on another.

22. NB Implementing a mandate is different from merely acting in such a way that the mandate’s conditions are fulfilled.

23. Somebody might define penal substitution more broadly than we have done, so that if C substitutes for B as the subject of A’s imposition of a punishment, then C penally substitutes for B. As we are using the phrase, however, C penally substitutes for B only if C is punished in B’s stead.

24. We’d like to thank Gert van den Brink, Stephen Clark, Tony Garrood, Simon Gathercole, Crawford Gribben, James Heather, Paul Helm, Sarah Hill, Howard Marshall, Richard Muller, Mark Murphy, Alexander Pruss, Stephen Rees, Sebastian Rehnman, John Ross, Eleonore Stump, Peter J. Williams, and Caleb Woodbridge. We’d also like to thank Oliver Crisp and Fred Sanders for hosting LATC 2015. Of course, none of them is to be held responsible for what we say.

25. “It is not a case of God punishing Christ but of God in Christ taking on himself the sin and its penalty. Indeed, at some point the challenge needs to be issued: where are these evangelicals who say that God punished Christ? Name them!” Howard Marshall, “The Theology of the Atonement,” in The Atonement Debate: Papers from the London Symposium on the Theology of Atonement, eds Derek Tidball, David Hilborn, and Justin Thacker (Grand Rapids: Zondervan, 2008), 63.

26. Translations not listed in the footnotes as coming from a translation were done by Daniel . Hill.

27. “Qui Deus Christum innocentem poenis affecit. — Sed pensandum est quomodo justus sit, et omnia juste disponat, si eum, qui non debet puniri condemnat. Mediator etenim noster puniri pro semetipso non debuit, quia nullum culpae contagium perpetravit. Sed si ipse indebitam non susciperet, nunquam nos a debita morte liberaret. Pater ergo cum justus sit, justum puniens, omnia juste disponit; quia per hoc cuncta justificat, quod eum qui sine peccato est, pro peccatoribus damnat.” Gregory the Great, Morals on The Book of Job, tr. Anonymous (Oxford: John Henry Parker, 1844 [original around 591]), bk. 3, ch. 14, §27, 148 – 49.

28. “Ideo cum Deus iuste et debite punivit Christum pro delicto suorum fratrum.” John Wycliffe, Sermones, Volume 1: Super Evangelia Dominicalia, ed. Johann Loserth (London: Wyclif Society, 1887 [original 1383/4]), sermon 42, §284.

29. “Ipse Deus percussit et punivit Christum.” Martin Luther, Werke: Kritische Gesamtausgabe: Volume 40, Part III (Weimar Ausgabe), ed. J. K. F. Knaake (Weimar: Böhlaus, 1930 [original 1543/44]), 715.

30. “At punivit nos in Christo, seu quod idem valet punivit Christum pro nobis, et loco nostri.” Johan Piscator, Analysis logica epistolae Pauli ad Romanos (Herborn: Corvinus, 1608 [original 1595], 4th ed.), 106, comment on Romans 5:16. We do not ourselves agree that these come to the same thing, though. Particular thanks are due to Richard Muller for the reference.

31. Samuel Rutherford, The Covenant of Life Opened (Edinburgh: Robert Broun, 1655), pt. 1, chap. 7, §32; pt. 2, ch. 7, §251, accessed December 30, 2014, http://reformedlayman.com/CovenantOfLifeOpened/THE%20COVENANT%20OF%20LIFE%20OPENED.htm.

32. John Owen, Works: Volume 9, ed. Thomas Russell (London: Baynes, 1826 [first edition 1655]), 443, accessed December 30, 2014, http://www.ccel.org/ccel/owen/vindicevang.i.xxxi.html.

33. “At Christus a Deo Judice ex Justitia punitur et in corpore et in anima ad peccatorum expiationem.” Francis Turretin, De Satisfactione Christi Disputationes (Geneva: De Tournes, 1666), Disputatio VI, Quae Quarta est de Veritate Satisfactionis, pt. 4, ch. 20, §163.

34. Solomon Stoddard, The Safety of Appearing at the Day of Judgment, In the Righteousness of Christ: Opened and Applied (Boston: Samuel Phillips, Stoddard, S. 1687), ch. 3, §51.

35. Christopher Love, The souls cordiall in two treatises: Volume 3 (London: Nathaniel Brooke, 1652), 165.

36. John Downame, The Christian Warfare (London: Kyngston, 1604 [first edition 1603]), ch. 16, §196.

37. Edward Philips, sermon on Romans 8:1 in Certain godly and learned sermons (London: Arn. Hatfield, 1607), 395.

38. Charles Wesley, “All Ye That Pass By” (Hymn 707), in A Collection of Hymns, ed. John Wesley (London: Wesleyan-Methodist Book-Room, 1889 [first edition 1779]), accessed January 1, 2015, http://www.hymntime.com/tch/htm/a/y/e/ayetpaby.htm. “Punished” was changed to “bruised” in George Booth, The Primitive Methodist Hymnal (London: Edward Dalton, 1889), 71.

39. Archibald Alexander Hodge, The Atonement: Its Nature, Design, and Application (Grand Rapids: Eerdmans, 1953 [first edition 1867]), 308.

40. Herman Bavinck, Reformed Dogmatics: Volume 3, ed. John Bolt and trans. John Vriend (Grand Rapids: Baker, 2006 [original: fourth edition, 1928]), 398.

41. William Magee, Discourses and Dissertations on the Scriptural Doctrines of Atonement and Sacrifice (London: Henry Bohn, 1859 [first edition 1816]), 118, italics original.

42. H. E. Guillebaud, Why the Cross, (London: InterVarsity, 1946, 2nd ed.), 145.