Justin Taylor links to another interesting article on the atonement, this time by Don Carson and entitled Why Is the Doctrine of Penal Substitution Again Coming Under Attack? Carson makes some excellent points, but loses me when he simply presupposes the superiority of the penal substitutionary model.
Carson begins the second of his five answers with this:
Some popular slogans that have been deployed to belittle the doctrine of penal substitution betray painful misconceptions of what the Bible says about our Triune God. The best known of these appalling slogans, of course, is that penal substitution is a form of “cosmic child abuse.” This conjures up a wretched picture of a vengeful God taking it out on his Son, who had no choice in the matter. Instead of invoking the Triune God of the Bible, this image implicitly pictures interactions between two separable Gods, the Father and the Son. But this is a painful caricature of what the Bible actually says. In fact, I do not know of any serious treatment of the doctrine of penal substitution, undertaken by orthodox believers, that does not carefully avoid falling into such traps.
Indeed this is a painful caricature. Those who wrote of “cosmic child abuse” were describing this caricature; they do not themselves have “painful misconceptions” but are writing to correct these misconceptions. I am glad that no serious treatment of the doctrine falls into these traps. But it is sadly true that many popular presentations land straight into them, including the one from Mahaney which I have referred to several times.
Carson moves on to discussing different models of the atonement:
In recent years there has been a lot of chatter about various “models” of the atonement that have appeared in the history of the church: the penal substitution model, the Christus Victor model, the exemplary model, and so forth.
Is Packer’s 1973 lecture to be dismissed as “chatter”? But Carson concedes that several different models may be valid, and that if so
choosing only one of them is being unfaithful to Scripture, for it is too limiting. …
One recent work that loves to emphasize the Christus Victor “model” … somewhat begrudgingly concedes that penal substitution is found in a few texts, not least Romans 8:3. But this work expends no effort to show how these two views of the atonement should be integrated. In other words, the work in question denigrates penal substitution as a sort of minor voice, puffs the preferred “model” of Christus Victor, and attempts no integration.
A good point. So one would expect Carson to say the same about some of the many works which emphasize PSA, and grudgingly accept other models as minor voices with no attempt at integration. For it should be clear that a comprehensive treatment of the atonement needs this kind of integration.
But instead, astonishingly, Carson, who is usually so good at spotting exegetical fallacies, falls into just the same fallacy as he has accused this “One recent work” of when he writes:
But I think it can be shown (though it would take a very long chapter to do it) that if one begins with the centrality of penal substitution, which is, as we have seen, grounded on a deep understanding of how sin is an offense against God, it is very easy to see how all the other so-called “models” of the atonement are related to it.
Note that he proposes to begin with, that is presuppose, “the centrality of penal substitution” and construct a framework around this. As a clever theologian, I am sure that in “a very long chapter” he could do this. But if it is really “very easy”, why the “very long chapter”?. He asserts
It is very difficult to establish the coherence if one begins anywhere else.
– but of course the need for “a very long chapter” implies that it is very difficult if one begins here as well! It would indeed be a major task to synthesize the various models of the atonement into a coherent framework in this way. But the only intellectually respectable way to do this is to look at the biblical text with as few presuppositions as possible.
As I said in my comment on the last post, Steve Chalke does make the mistake Carson says he makes. You admit that Carson is right to call it a caricature. Chalke doesn’t seem to want to call it that in this paper. He in fact calls it penal substitutionary atonement and attributes the view to Calvin and Hodge. Then he shoots it down as cosmic child abuse. It is very clearly a caricature of their view.
As for Carson’s more substantive point, I think you’re misunderstanding what he thinks the very long chapter would involve. I don’t think he’s saying that he could assume PSA as a center and then show that the others are subordinate. That would take two lines, not a chapter. Of course they’re subordinate if PSA is central. What he’s saying is that he could write a very long chapter showing that the other “models” of the atonement actually depend to some degree on PSA. They are all true, but it is more fundamental, and the others make little sense without PSA.
For an example of the kind of thing I think he has in mind, see my post on Christus Victor and how it in some ways depends on more fundamental aspects of the atonement because there needs to be something for Christ to be victorious over besides Satan if the war between God and Satan is to have any fundamental meaning. That’s the kind of task that I think he has in mind, although my argument would provide only a small part of the kind of thing he would want to do.
Pingback: Threads from Henry’s Web » Blog Archive » Smartypants Notions Like Cosmic Child Abuse
Jeremy, see my answer to your similar comments about Chalke here. The Chalke article is here.
You wrote:
If this can be done in two lines, why didn’t Carson do it? Perhaps in the absence of such a brief proof from him, you could provide it for me. But no doubt Carson realises that any brief demonstration of this would be full of exegetical fallacies.
Meanwhile I will read your post on Christus Victor.
Chris, to summarise, I accept a carefully stated version of PSA as one valid model of the atonement among others. But that carefully stated version would have to include, as Packer does, an explanation of how Jesus’ death was part of the agreed will of the Trinity, not something decided by the Father alone. For more, see my past posts on the atonement.
Hi Peter –I’ve been thinking carefully about your posts and your interaction with Carson’s thoughts as well as your comments about Mahaney’s summary.
Sorry, I am sure I should have picked up on this previously, though I am relatively new to your blog – – but do you accept PSA as one acceptable model, or do you reject it altogether?