When a theologian undertakes a law degree in order to better apply legal theories to their understanding of the atonement, it demonstrates them to be aiming a very high level of care and precision in that which they produce. Those characteristics will already be recognisable to those familiar with Oliver Crisp’s previous work (particularly on the person of Christ).
This small book is just a foretaste of Crisp’s forthcoming major constructive account of the atonement (Partication and Atonement, 2021). It mostly covers existing models of the atonement (though to term them models is in itself controversial, and an issue which Crisp takes up), assessing their distinctive elements and commenting on their strengths and weaknesses. Crisp, again characteristically, also offers potential solutions to those weaknesses in the form of responses or proposed modifications to the views under discussion.
Refreshingly, Crisp is clear from the outset that offering a full account of the atonement requires some explanation of the mechanism by which it is achieved, the how question. On this criteria, the once much-vaunted Christus Victor model does not count as an account of the atonement at all, for while victory over evil may be the result of the atonement, the Christus Victor accounts themselves do not explain (or at least not usually – Crisp offers some suggestions how they might) how this comes about. Crisp also declares his intention of considering the nature but not the scope of the atonement, given that decisions regarding scope are not intrinsic to the issue of how the atonement works.
What follows is not a blow-by-blow trip through the book but some selected comments on aspects I found most interesting. I may add to it over time.
Patristic accounts and Athanasius
The chapter on Patristic atonement accounts is indebted to the work of Benjamin Myers on the subject (in the volume of essays Locating Atonement, edited by Crisp himself). Crisp is keen to claim that Athanasius in particular should not be seen as offering a version of penal substitutionary atonement (PSA), despite a clear sacrificial and substitutionary element to his thinking. For Athanasius, according to Crisp, doesn’t view Christ’s death as punishment. The debt paid by Christ’s death is paid to death itself, a “shadowy power that holds sway over fallen humanity” (41). Crisp makes much of the observation of Myers that in the locus classicus of De Incarnatione used by subsequent writers to find PSA in Athanasius, Christ offers his sacrifice to death, not to God.
It seems to me there are a few problems here. For one thing, Athanasius also speaks of this offering being to the Father:
And thus taking from our bodies one of like nature, because all were under penalty of the corruption of death He gave it over to death in the stead of all, and offered it to the Father — doing this, moreover, of His loving-kindness, to the end that, firstly, all being held to have died in Him, the law involving the ruin of men might be undone (inasmuch as its power was fully spent in the Lord’s body, and had no longer holding-ground against men, his peers), and that, secondly, whereas men had turned toward corruption, He might turn them again toward incorruption, and quicken them from death by the appropriation of His body and by the grace of the Resurrection, banishing death from them like straw from the fire.
(https://www.newadvent.org/fathers/2802.htm)
In this light the “offering to death” language doesn’t seem particularly important – here, after all, Athanasius says both that he gave his body over to death and offered it to the Father.
Further more, this passage also makes a clear connection between being under penalty and Christ’s death in our stead. That may fall short of an explicit penal substitutionary mechanism, and to note this is present is not to claim that it is the central concept that Athanasius works with. But I am unconvinced that either Myers or Crisp have proven there are no penal substitutionary themes in Athanasius.
(There is a legal theme to Athanasius’ argument (do with “what was owed”, not just a shadowy power) that also doesn’t seem to be adequately drawn out in Crisp or Myers’ treatments. But this latter claim is one that would require a closer reading of the whole of the work to substantiate.)
[Edit: having just re-read Anselm’s Cur Deus Homo, I think Crisp gives an accurate and very user-friendly summary of how his doctrine of atonement works in Ch 4. He also helpfully show how it is similar to (resting on shared foundations) but differs from penal substitution.]
Penal substitution
Crisp has an interesting relationship to PSA. My guess would be that he thinks there is much truth in it, but that it is held too easily and too crudely by the majority of evangelicals. He therefore terms his own account “participatory” atonement, even though the mechanism within is very similar to PSA.
(Arguably Crisp’s own account doesn’t do much more than address the legal fiction objection to PSA by means of union with Christ, as well as make more prominent some of the patristic themes and language around atonement. Contrary to what Crisp seems to say, this isn’t very novel, given that a number of Reformed writers ground the basis for the legal union of Christ with his people in their real union – as Edwards puts it, as Crisp even quotes. Bavinck does something similar.)
Crisp rightly points out that to have a substitutionary element in an account of atonement is not sufficient for it to be a species of PSA. For PSA, the substitution needs to be a particular kind: the substitution of person undergoing penalty.
Interestingly, Crisp holds to a particular definition of punishment by virtue of which it is not possible for an innocent person to be punished. Punishment, according to Crisp, is inherently linked to the guilt of the person punished. If an innocent is wrongly convinced and sentenced to prison, they suffer, but are not punished for they have not done the wrong. According to this definition, Christ simply cannot be punished for the sin of his people as in (at least some articulations of) PSA, for he is not guilty of it. Christ can bear the penal consequences of sin, as if being punished, and Crisp proposes this might be a way of reformulating the idea of PSA to make sense (and it’s what he opts for in his own account).
I’ve not yet done the work to comment on whether this is a venerable definition of punishment, but for the moment it seems to me to be confused. After all, if a person is wrongly convicted and sentenced to prison, we would seem to normally say they are being unjustly punished, not that they are not being punished at all. A suffering is being inflicted on account of a wrong done, although upon the wrong person. Crisp’s definition seems to exclude the possibility of unjust punishment in terms of punishing the wrong person (though not punishments unjust in degree, for example). Consider also the punishment of a whole class of pupils for the misdemeanour of one. We might feel it unjust (most of us can still remember such a feeling!), but we would not think we were not being punished, because of the connection between the suffering and the wrong committed.
The real problem is whether Christ bearing punishment (or bearing penal consequences if you follow Crisp) is just, because he is not the guilty party. Crisp makes much of the legal fiction involved in accounts of imputation at this point. And it is true that this is the major place at which PSA accounts have to do work to justify themselves. However, it seems to me that Crisp’s own participatory defence could be appropriated by a more traditional account of PSA. The connection between Christ and his people is the issue, and in Reformed thinking Spirit union with Christ is portrayed as an impressive reality rather than mere legal fiction.
Redemptive violence
A criticism that affects many of the approaches to atonement Crisp surveys is that of atoning violence. The argument goes that violent acts are morally reprehensible, and so cannot form part of the design of the atonement (even for the sake of a good end).
Crisp gives short shrift to the divine child abuse objection, which he thinks is impossible to level against any account of atonement that operates with an orthodox view of the Trinity. (For my part I think this now so thoroughly established that opponents of PSA would be well-advised to leave it alone.)
Crisp thinks that often attempts to rebut this argument (esp. by defenders of PSA) operate merely by “rhetorical flourish.” That is, they argue that as crucifixion is inherently violent, and as the cross is central to the atonement, therefore violence is part and parcel of the atonement. For Crisp, this is question-begging; the conclusion is smuggled into the premise. This is so in two ways: firstly because such a defence assumes that because crucifixion is violent the atonement is an act of violence, and secondly by conflating crucifixion with atonement.
I don’t think Crisp is quite as clear on this as he normally is. What he means I think becomes clear only once his own preferred approach to the issue is laid out. Crisp argues that the issue of intent is inherent to the definition of violence: a hard shove to move you out of the way of a speeding car is not therefore an act of violence. Clearly with such a definition, it could be possible to say that God’s work of atonement, although involving violent acts on the part of the human actors involved, was not itself an act of violence. The crucifixion is a part of the atoning work, but the act of crucifixion itself is not to be treated in this regard extracted from the design of the whole. With such alternatives in place, we can see why Crisp would argue that it is question-begging simply to assert the atonement is violent because it involves the violence of crucifixion.
Having said that, if one is to advance an atonement theory that actually achieves atonement via the crucifixion (possibly among other means, but in which the crucifixion is necessary), then violent acts are at least involved. Crisp may think he can solve this problem by insulating God from the violence by means of differentiating God’s intent and the intent of the human actors, but not all will agree with him. The root issue is really that of the sovereignty of God, the relationship between God’s ordination and the acts of human agents, as we will see.
Crisp considers two solutions to redemptive violence. The first simply accepts that some violence can be good, like the pain that warns your hand not to remain on the hot plate. However, Crisp rejects this because of his contention that a violent act cannot be virtuous, for if the intent is virtuous then the act should no longer be described as violent. This is really a definitional issue. The acts that make up the atonement are the same, Christ is still pierced; the question is whether this should be termed violence (especially divine violent). It seems to me that Crisp is right that we should not frame the atonement as divine violence (though I suspect it is only the opponents of traditional acts of the atonement who have done so), in that the divine intent is reconciliation, though being clear that violent acts are necessarily involved. I do question however, how Crisp would characterise the divine act of eternal punishment. Is this violence? Perhaps not, on his definition, in that what is intended and accomplished is not evil. But it certainly is divinely-intended suffering. Perhaps Crisp would say it is divinely-intended justice, experienced as suffering, although I think few would be much convinced by the distinction.
The option Crisp finds more promising is that of a double-effect response, in which it is make clear that God ordains the acts of the soldiers crucifying Christ, but does not intend the violence himself. This is easier, Crisp implies, if one holds a view of God’s sovereignty in which he foresees but does not determine human free action. If one is committed to a more characteristic Reformed view in which all things are ordained by God (though Crisp would want to insist this is but one view within the Reformed tradition), then there may still be room to say that God intends the same actions for good that the human actors intend as evil (and therefore the act is not violent on God’s part, though it is on the part of the soldiers). As such, in the crucifixion God intends to bring about injury to Christ without intending to harm Christ.
It seems to me this is all useful, and the distinction in intent is a right one. Popular depictions of the atonement that claim God is angry with Christ seem to be wide of the Scriptural language. That his wrath is visited on Christ is not quite saying the same, and is preferable.
But this is not a discussion particular to the atonement. It is the application of just the same categories as standardly discussed when considering God’s relation to evil in the world. Genesis 50:20, a key text in such discussions, would seem very apt to Crisp’s discussion: “you intended it to harm me, but God intended it for good.” Whatever means of parsing out how God wills the free act of human agents is used (such as a doctrine of concursus), such distinctions will be in play.
At this level, I think the “rhetorical flourish” response, for all it’s bluntness, still has a point. Scripture makes plain that the crucifixion, in all its human violence, was intended by God (Acts 4:28 being the tip of the iceberg of this theme) to achieve atonement. Whether or not this was divine violence is a vital question, as Crisp shows, but one suspects that even to distinguish between divine and human intent will not satisfy many of those who raise the issue of redemptive violence. Only a theory of the atonement that makes the cross into something not intended by God at all will absolve him of the charge of violence. And as Crisp shows so well, such a theory is unlikely to form a very satisfactory account of atonement at all. There is, I would maintain against Crisp, a certain burden of proof on this type of objector to show how a satisfactory account of atonement can be given that completely absolves God of involvement in violence and yet retains the cross as acheiving atonement in some meaningful way.