Note: John Johnson published an article in Evangelical Quarterly 75:3 (2003) entitled 'Is Cornelius Van Til's Apologetic Method Christian, or Merely Theistic?'. A rejoinder by John Frame and Steve Hays was published in EQ 76:3 (2004). Recently a direct response from Johnson appeared in the online Global Journal of Classical Theology (edited by John Warwick Montgomery). Since I had been following the exchange with interest, I emailed some comments to Steve, which he subsequently encouraged me to post online. So here they are (edited slightly for readability). Steve has also posted his own comments on his world-famous blog.
1. Johnson's article is shot through with a confusion between arguments for presuppositional apologetics (i.e., as a distinctive methodology) and presuppositional apologetic arguments (i.e., the actual arguments employed by presuppositionalists). In his introduction, Johnson states that his goal is to show that Muslims could use Van Tilian presuppositional arguments just as well as Christians: "My point is not that Islamic apologists are using the Van Tillian system to promote Islam. Rather, I wish to show that they easily could if they so desired."
However, the entire first section of the article is irrelevant to showing this. Johnson argues that Muslims could agree with Van Tilian claims about the noetic effects of sin and the suppression of knowledge of God. But unless I've missed something, Van Til nowhere appeals to these doctrines as part of an apologetic argument for Christian theism, i.e., as grounds for thinking that Christianity is true. The relevance of these doctrines for Van Til is their support for a presuppositionalist methodology. In other words, Van Til is appealing to Christians to recognize that acceptance of these doctrines (among others) requires one to take a presuppositionalist approach. The issue here is an intra-mural debate over apologetic methodology, not an extra-mural debate over the truth of Christianity.
So at most Johnson shows that some Muslims beliefs provide support for a presuppositionalist methodology (i.e., a Muslim could consistently advocate a presuppositionalist form of Islamic apologetics). But that's strictly irrelevant to his declared thesis. One might as well argue (against a Christian evidentialist, e.g., Gary Habermas) that since Muslim apologists also appeal to evidences, evidential arguments for Islam must be on a par with evidential arguments for Christianity!
2. Van Til considered the Christian doctrine of the ontological Trinity to be at the centre of his apologetic. And if any component of his presuppositionalist argument supports a distinctively Christian theism, this is it. But while Johnson apparently recognizes this point, he barely addresses it.
I certainly think that Van Tilians need to do more work on this issue; to spell out the argument in more rigorous detail. Yet with criticisms as superficial and vacuous as Johnson's, one hardly feels under pressure to do so! The closest he comes to a critique is to quote Geisler. But the quotation only offers us Geisler's conclusions. There's simply no argument here! We want to know why Van Til "has not shown that is it necessary to posit a triune God." Are we just to take it on Geisler's authority? Johnson takes the debate no further forward.
By the way, I find it somewhat amusing that in the footnote providing the Geisler reference (36), Johnson acknowledges Van Til's emphasis on the ontological Trinity as the solution to the problem of the One and the Many, and then recommends Frame's analysis in Cornelius Van Til: An Analysis of His Thought — as if that somehow supported Johnson's case! Why didn't Johnson engage with that?
3. I'm amazed that Johnson didn't understand Frame's claim that Judaism is essentially a Christian heresy. Surely it's obvious that Frame was referring to modern Judaism. Or does Johnson think that modern Judaism is contiguous with Old Testament Judaism?
4. Johnson wants to have his cake and eat it. In the first section, he appeals to alleged theological agreement between Christianity and Islam so as to undermine the claim that presuppositionalism is unique to Christianity. But in the second section, he appeals to theological disagreement between Christianity and Islam so as to undermine Frame's characterization of Islam as a Christian heresy. So which is it? Does Islam have much in common with Christianity or does it not?
5. Johnson doesn't recognize that the capriciousness of Allah undermines the rationality of induction. It's not enough that Allah does, as a matter of contingent fact, maintain order in the universe. We also need sufficient epistemic warrant to believe that Allah will continue to do so; indeed, that he has pledged to do so (and will not go back on his word).
For the Christian, the reliability of induction is grounded in God's covenant faithfulness (Gen. 8:22; Jer. 33:19-26). Allah has no comparable attribute. Indeed, it would be beneath Allah to bind himself to men with a covenantal promise!
So, contrary to Johnson's claims, Islam cannot provide the epistemological preconditions of induction, even if we grant that Allah is responsible for the order in the universe. After all, the same theology that deems Allah to be "the force that binds the universe together" can just as easily support the view that with the same force Allah could at any time, in an act of sheer unfettered will, throw the universe into utter disarray! Who can say one way or the other?
6. Johnson also doesn't appreciate the epistemological instability in Muslim claims about divine revelation. Islamic tradition claims both (i) that the Torah, Psalms, and Gospels were indeed divinely inspired scriptures (thus revelatory at least at the time of inscripturation) and (ii) that all of the scriptures prior to the Koran have been hopelessly corrupted. But surely this furnishes the epistemologically self-conscious Muslim with a defeater for his belief in the reliability of the Koran. If Allah permitted the earlier scriptures to become corrupted, to the degree that Jews and Christians became lost in falsehood, what assurance does the Muslim have that Koran has not also become corrupted? What's sauce for the goose is sauce for the gander. For all he knows, the 21st-century Muslim is in no better position than the 6th-century Jew or Christian.
Obviously the Muslim cannot block this defeater by appealing to the testimony of the Koran, for that would be epistemically circular. In the absence of any other defeater-defeater (and what else is there to do the job?), radical theological skepticism beckons.
All this does not bode well for the claim that Islam is on a par with Christianity with respect to the preconditions of human knowledge. Once you take Islamic theism "as a unit" (cf. Van Til's "Christian theism as a unit") — rather than treating it as just one particular flavour of a generic theism — Johnson's parity thesis just doesn't hold water.
7. In the third section, Johnson points out that the Muslim will appeal to the authority of the Koran to 'trump' the Christian's appeal to manuscript evidences in defence of the reliability of the New Testament. And this is true enough; Johnson rightly recognizes that the Koran is deemed to be an ultimate self-attesting authority within the Muslim system just as the Bible is deemed to be an ultimate self-attesting authority within the Christian system.
But once again, Johnson wants to have his cake and eat it. For his closing shot is to recommend an appeal to "the resurrection of Jesus as . . . strong, objective evidence for Christianity" to break the alleged 'stalemate' between the Muslim and the Christian. Yet this obviously won't do; for if an appeal to the Koran trumps any evidence for the reliability of the New Testament, it will equally trump any evidence for the historicity of the resurrection! Johnson can't have it both ways. If he wants to maintain that there is a presuppositional stalemate between the two worldviews, on account of their competing revelational authorities, then he's pulled the carpet from under any evidentialist apologetic. A tenacious Muslim, reasoning consistently in accordance with his "Koranic presuppositions", will no more be moved by supposed evidence for the resurrection than supposed evidence for the integrity of the New Testament.
Van Til saw this point clearly. And that's one of the reasons he insisted that a presuppositionalist apologetic cannot be satisfied with proving a merely generic theism.
There are a number of other problems with Johnson's article, although they're less serious that those I've mentioned above.
Bottom line: Johnson's critique gives Van Tilians little cause for concern. But you knew that already!