A Van Tilian Non Sequitur

There was a time when I would have considered myself to be a full-blooded Van Tilian presuppositionalist. I was introduced to Van Til's writings in my sophomore year of college and found myself dazzled by what I perceived to be a the most consistent and penetrating apologetical system I had ever encountered. I was convinced by the end of my junior year that Van Til's apologetic was the only legitimate apologetical option and that all there was left to do was to translate his insights into more contemporary Anglo-American philosophical idiom and perhaps clarify some of his more recondite arguments.

But as I went along and tried to formulate Van Til's apologetic more rigorously I found that at crucial junctures the conclusions Van Til drew simply did not follow from his premises. I can't tell you how frustrating and disappointing this was for me.

Allow me to give an example. Van Til argues that it is improper to offer probabilistic arguments for Christianity thus, "How could one ever argue that there is greater probability for the truth of Christianity than for the truth of its opposite if the very meaning of the word probability rests upon the idea of Chance? On this basis nature and history would be no more than a series of pointer readings pointing into the blank….He is obviously thinking of such a God as could comfortably live in the realm of Chance. But the God of Scripture cannot live in the realm of Chance."[1]

Now, this criticism of probabilistic arguments for Christianity seems to me to be a simply mistaken. It is perfectly sensible to use the language of "probability" to describe the epistemic status a proposition has for you without implying that the truth of that proposition is contingent upon sheer "Chance." Take Goldbach's conjecture that every even number greater than 2 is the sum of two primes. So 4 is 3 and 1, 6 is 3 and 3 or 5 and 1, 8 is 7 and 1 and so on. Goldbach's conjecture has not been proven, but it must be either true or false. What's more, because it's a proposition about mathematical realities, it must be either necessarily true or necessarily false. The truth or falsehood of Goldbach's conjecture is not contingent and therefore cannot be determined by or contingent upon "Chance." But, like I said, it hasn't been proven (and Gödel has given us reason to suspect that it may not be provable at all).

Now, let's say someone gives me a particularly impressive argument for Goldbach's conjecture. It's not a "proof" according to the most rigorous definitions of "proof," but it's a good argument. Upon considering the argument in light of all else that I know, it seems more probable than not that Goldbach's conjecture is true. 'Given the premises of the argument, it is probable that Goldbach's conjecture is true.' To say this sort of thing does not commit me to any particular view of the ontic or modal status of Goldbach's conjecture because, like I said, Goldbach's conjecture is either necessarily true or necessarily false. It's just a description of how I perceive the strength of the arguments available to me for the proposition or, put another way, a description of how strongly I am inclined to believe the proposition given what I know.[3]
If this be the case, it doesn't follow that if you argue that it is more probable than not that Christianity is true, that therefore you have conceded that God's existence is contingent upon "Chance" or that God "lives in a realm of Chance" or anything like that. Arguing thus doesn't really commit you to any particular stance on the modal status of God's existence any more than arguing thus commits you to a particular stance on the modal status of Goldbach's conjecture. One can believe that if God exists, He must exist necessarily and coherently speak of God's existence in terms of greater and lesser epistemic probabilities with respect to the evidence one has been given. This point is important because the line of argument that I'm criticizing is mainly used by Van Tilians to bash other Christian apologists for their usage of apologetical argument forms that deliver epistemic probability rather than absolute certainty.[2] In my estimation, Van Til's criticism of probabilistic arguments is a non sequitur and Van Tilians would do well to abandon it.

[3] This Goldbach’s conjecture example was inspired by the first lecture or two of Saul Kripke’s Naming and Necessity.
[2] I don’t really think Van Til’s so-called “transcendental argument” really gets you absolute (Cartesian) certainty either. At the end of the day it would seem Van Til's argument is not formally distinct from other apologetical arguments and gives only epistemic probability.

[1] Van Til, as quoted in Bahnsen, Van Til’s Apologetic, p. 584