Does Presuppositionalism Engage in Question-Begging?

The Gospel Coalition is running a series on methods in apologetics. The latest installment is “Questioning Presuppositionalism” by Dr. Paul Copan, who raises four criticisms of presuppositionalism, one of which is the old canard that presuppositionalists engage in fallacious circular reasoning. (I think all four are misguided in one way or another, but the other three will have to wait for now.) He writes:

First, it engages in question-begging — assuming what one wants to prove. It begins with the assumption that God exists, and then concludes that God exists. Such reasoning would get you an “F” in any logic class worthy of the name!

Dr. Copan is a gentleman and a scholar, so I’m sure he doesn’t realize quite how insulting this sounds to presuppositionalists! (For comparison, imagine someone claiming that evidentialists commit the fallacy of affirming the consequent because they use inductive inferences.) This criticism has been answered many times, so it’s disappointing to find it cropping up yet again (although perhaps presuppositionalists should take comfort from the fact that Dr. Copan doesn’t offer any new criticisms!). Even so, I’ll try to explain one more time why this complaint so badly misses the mark.

In his contributions to the book Four Views on Apologetics, Dr. Copan’s fellow apologist William Lane Craig notes that this is a common criticism of presuppositionalism:

As commonly understood, presuppositionalism is guilty of a logical howler: it commits the informal fallacy of petitio principii, or begging the question, for it advocates presupposing the truth of Christian theism in order to prove Christian theism.1

However, Craig also recognizes that the criticism is fairly superficial:

If this were all presuppositionalism had to offer as an apologetic, it would be so ludicrous than no one would have taken it seriously. But at the heart of presuppositionalism lies an argument, often not clearly understood or articulated, which is very powerful. This is an epistemological transcendental argument.2

Dr. Copan is also well aware of presuppositionalism’s use of transcendental argumentation, as his TGC article illustrates:

Typically, these presuppositionalists (e.g., Bahnsen) avoid traditional cosmological (causal), teleological (design), and moral arguments, but they enthusiastically endorse the transcendental argument for God (TAG) — the argument to show that God is the inevitable ground for all rational thought.

It’s therefore all the more surprising to find him repeating the “begging the question” charge, for once one understands the nature of a transcendental argument it’s clear that no fallacy of petitio principii is being advocated or committed. A transcendental argument typically takes the following form:

(1) If X were not the case, Y would not be possible.
(2) Y is possible.
(3) Therefore, X is the case.

In the presuppositionalist’s argument, X is the existence of God and Y is rational thought. (For simplicity, I’m using Dr. Copan’s own terminology here; I’m also glossing over some technical questions about the formalization of TAG, some of which were addressed in my exchange with David Reiter, but none of them affect my argument here.) So let’s assume that the following is a workable summary of TAG:

(1) If God did not exist, rational thought would not be possible.
(2) Rational thought is possible.
(3) Therefore, God exists.

One common criticism of TAG is that presuppositionalists haven’t adequately defended the first premise. However, that’s not Dr. Copan’s criticism. His charge is that presuppositionalism is guilty of “assuming what one wants to prove.” But how exactly does the argument above assume what it sets out to prove? How does it assume the existence of God in any rationally objectionable fashion?

The problem here is that Dr. Copan, like many critics of presuppositionalism (and even some of its would-be defenders), confuses a presupposition of an argument with a premise of an argument. There’s a significant sense in which the argument above does indeed presuppose the existence of God. For if the first premise is true, the existence of God is a necessary precondition of rational thought, and the possibility of rational thought is a presupposition of all argumentation, including TAG. So in an obvious sense, if TAG is sound then TAG presupposes the existence of God (and so does the cosmological argument, the teleological argument, the moral argument, and every other theistic argument). But this is not at all to imply that the existence of God functions as a premise in the argument. TAG doesn’t look remotely like this:

(P1) God exists.
(P2) …
(P3) …
(C) Therefore, God exists.

Nor does TAG employ any premises that trivially presuppose the existence of God (e.g., “God is all-knowing” or “God has spoken in the Bible”). So it’s hard to see exactly why Dr. Copan thinks that presuppositionalism flunks Logic 101.

The Cartesian Cogito
Perhaps the following analogy will help to set the record straight. Descartes’ famous argument — “I think, therefore I am” — can be seen as a kind of transcendental argument; specifically, a transcendental argument for one’s own existence. I cannot doubt my own existence without thinking, but I can think only if I exist; thus even my doubting presupposes my existence. I can doubt my own existence only on pain of performative inconsistency: the content of my intellectual act (the denial of my existence) is inconsistent with the performance of my intellectual act (the doubting of my existence).

So, does Descartes’ Cogito beg the question? Does it simply assume what it purports to prove? Does it commit a logical howler? Not at all. It’s far more subtle than its crude parodies (e.g., “I stink, therefore I am”!). As I mentioned above, the argument identifies a performative inconsistency in the one who doubts his own existence. (In a sense, all transcendental arguments aim to identify a performative inconsistency in the skeptic’s position.) Does it presuppose its conclusion? Yes, in the sense that the argument can be mentally entertained by a person only if that person exists — but that’s precisely the point. This sort of non-trivial ‘presupposing’ is necessarily involved in all transcendental arguments that purport to identify a necessary precondition of rational thought.

Once you see that Descartes’ argument doesn’t beg the question in any objectionable fashion, it ought to be clear that neither does the presuppositionalist’s argument. For whereas René Descartes argued, in effect, that his existence is a necessary precondition of his intellectual activity, Cornelius Van Til argued that God’s existence is a necessary precondition of his intellectual activity (and of everyone else’s). As someone once quipped, whereas Descartes argued, “I think, therefore I am,” Van Til argued something far more profound: “I think, therefore I AM.” (If you don’t get it, review Exodus 3:14!)

The Van Tilian Cogito
Whatever other criticisms it might invite, Descartes’ anti-skeptical argument would not have garnered the attention it has if it transparently committed the fallacy of petitio principii. Yet the presuppositionalist’s argument can be seen as very similar in its logical form. It simply isn’t true that presuppositionalism commits an elementary logical blunder — and it’s high time its critics laid that tired old charge to rest.

Addendum: Steve Hays has posted some good comments (here, here, and here) in response to Dr. Copan’s article.

  1. Craig, “A Classical Apologist’s Response,” in Steven B. Cowan, ed., Five Views on Apologetics (Zondervan, 2000), p. 232.
  2. Ibid., p. 233.

34 thoughts on “Does Presuppositionalism Engage in Question-Begging?”

  1. If this is the case, then why is the starting point “biblical revelation ?”
    You say that: “TAG doesn’t look remotely like this:
    (P1) God exists.
    (P2) …
    (P3) …
    (C) Therefore, God exists.
    But doesn’t your starting point amount to the same thing?

    1. Roger,

      “Starting point” is ambiguous. What do you mean by that term? There is a sense in which presuppositionalism “starts with” biblical revelation and another sense in which presuppositionalism “starts with” the unbeliever’s own claims and experiences.

      1. If this is the case, then why do you start with “biblical revelation?”
        You say that: “TAG doesn’t look remotely like this:
        (P1) God exists.
        (P2) …
        (P3) …
        (C) Therefore, God exists.
        But doesn’t what you are starting with amount to the same thing?

      2. In what sense does presuppositionalism “start with” the Bible, and in doing so, how is the charge of begging the question avoided?

        1. I can’t speak for all presuppositionalists, of course, but I’d say that presuppositionalism starts with the Bible in at least the following senses:

          1. It seeks to consider first what the Bible has to say about the proper goals and guiding principles for apologetics.
          2. It seeks to develop its apologetic methodology in the context of a biblical epistemology.
          3. It doesn’t appeal only to natural revelation, “bracketing out” special revelation; it appeals to both together, presenting the Christian worldview for consideration “as a unit” over against competing worldviews.
          4. From the outset, it treats the Bible as the Bible treats itself: as divine revelation, not as “generally reliable ancient documents”.

          But in none of this does presuppositionalism require the unbeliever simply to assume without argument that the Bible is the Word of God. Rather, it invites the unbeliever to assume for the sake of argument that the biblical worldview is true and to consider both its inner coherence and its capacity to account for our basic human experiences in a way that no competing worldview can do. The difference between “without argument” and “for the sake of argument” is the difference between viciously circular argumentation and broadly circular argumentation.

      3. Thanks for answering. However, it has always seemed to me (and many others) that presuppositional apologetics makes it appeal to both natural and divine revelation in way that not even God would approve. We are never asked “for the sake of argument” to consider both together as a possible worldview. Rather, God Himself provides unnassailable proof for each individually: natural revelation is proven by the created order, and divine revelation is likewise authenticated and proven through miracles, prophecy, etc.,. Why not use what He has given to us? Why not use the “individual” proofs that He Himself has understood as necessary? Presuppositional apologetics is all very clever, but it would ultimately leave the unbeliever “with excuse.”

  2. I’m not so sure if the Cogito analogy is a correct one. Here’s why:

    If we were to formulate the Cogito using the very same formulation as the “workable summary of TAG” (WSOT from here on) then we would have something like this:

    1) If I did not exist, my thinking would not be possible
    2) My thinking is possible
    3) I exist

    Now, if this is the correct formulation of the WSOT then the glaring problem is premise 1. For it includes “I” and “my” in the very same premise. This is very different from the WSOT that speaks of rational thought in general ( I’m assuming its speaking of human rational thought). So if we were to “reverse” the formulation of Cogito here to God…it would seem that the correct analogy would be:

    1) If God did not exist, God’s thoughts would not be possible
    2) God’s thoughts is possible
    3) Therefore, God exists

    It would seem that they are very different, and that Cogito does in fact beg the question.
    I would really appreciate it if you would correct me where I have faulted. Thanks.

    Sidenote: I noticed that you posted Steve Hays blog, I was wondering if you knew of Choosing Hats, as they are exclusively Van Tillian Apologists. They too have posted something about Dr. Copan’s post

    1. Just because ‘I’ features in both the antecedent and the consequent in the Cogito, it doesn’t follow that we have to substitute ‘God’ in both cases. That would entirely miss the parallel between Descartes’ TA and Van Til’s TA — the parallel I explained in the post. TAG isn’t concerned with the metaphysical preconditions of divine thoughts, but rather the metaphysical preconditions of human thoughts. So your second formulation should read:

      1) If God did not exist, my thoughts would not be possible.
      2) My thoughts are possible.
      3) Therefore, God exists.

      In the schematic transcendental argument I gave, X is the metaphysical precondition of Y, where Y is (typically) some human noetic activity.

      Thanks for the tip about the CH post. Actually, I saw it already. I’m subscribed to their blog. In fact, I should add it to the blogroll.

      What is an “exclusively Van Tillian Apologist”? One that doesn’t date William Lane Craig or Gordon Clark on the side? :)

  3. Pingback: Responses to Paul Copan’s Questioning Presuppositionalism « The Domain for Truth

  4. Pingback: Some thoughts on presuppositional apologetics from The Gospel Coalition :The Aristophrenium

  5. Hi James,
    You’ve pointed out that TAG doesn’t beg the question, whereas Copan claims that presuppositionalism begs the question. Given the importance of TAG for presuppositionalists, I would grant that Copan’s written criticism is hamfisted at best. But I wonder if there’s some charge in the vicinity that would have been correct and that’s consistent with your remarks about TAG. Perhaps he thinks that (i) some presuppositionalists have expressed their position in ways that do give the appearance of employing question-begging argumentation. (The reply to this part of Copan’s article at Choosing Hats distinguished between vicious and non-vicious (“broad”) circularity.) Or perhaps Copan thinks that (ii.) presuppositionalists can’t defend TAG’s first premise (while remaining “presuppositionalist”) without begging the question. My Van Til is rusty, but I can imagine him saying something like this:

    “I can’t defend the claim that, (1) ‘were God to not exist, rational thought wouldn’t be possible,’ on the basis of reason. That would be to treat reason, or my reasoning faculties, as an autonomous or ultimate reference point. So, If I’m going to defend it at all, I have to appeal to divine revelation.”

    If a presuppositionalist thought something like this, then it would seem that (1) either cannot be defended or can only be defended by invoking some premise that “trivially presupposes” God’s existence.

    1. Hi Dan,

      Thanks for the comments.

      I agree with your (i) — which is why I don’t think Copan is entirely to blame for his confusion. (Hence my “would-be defenders” remark.)

      I don’t see anything in Copan’s article to suggest that he had (ii) in mind. Moreover, I disagree that VT would have insisted on appealing to divine revelation to defend TAG’s conditional premise, because that’s not what he actually does in his writings. Rather, he argues (someone obliquely, admittedly) for that premise on philosophical grounds. (See this paper for some discussion.) This wouldn’t be treating reason as an ultimate reference point, merely as a proximate one.

      One qualification, however: VT would insist that the content of the term ‘God’ in (1) must come from divine revelation. But I don’t think that involves any question-begging.

  6. Pingback: Copan's Folly

  7. Pingback: Presuppositionalism and Circularity…Again? « KINGDOMVIEW

  8. Pingback: Dialoging about Covenantal Apologetics – Justin Taylor

  9. seriousactualist

    Setting aside the specific sort of meat of which Copan’s fists consist, I should think that Dan is very nearly correct about the substance of his (Copan’s) charge of question begging. The suggestion (I think) is not that the presuppositionalist argues formally in a circle; i.e., that her argument has the form:


    :. p

    Rather, the idea (very roughly) is that the central premise(s) of her argument(s) — TAG, e.g. — are such that one couldn’t have good reasons for the premise independent of the conclusion. This of course is rough. But in this way, the charge is a logical one, but a matter of informal logic. (I’m inclined to think the distinction Steve Hays points out between logical and epistemological circularity is related but different.)

    A couple of other quick thoughts. First, for the classically Christian theist:

    (1) If God did not exist, rational thought would not be possible

    is a counterpossible, in that the antecedent of that conditional (the proposition that God doesn’t exist) is impossible. On the “standard” counterfactual semantics, both it and:

    (1*) If God did not exist, rational thought would be possible nonetheless

    are vacuously true. Pretty clearly, the presuppositionalist demurs. (I know I do.) It would be fun to say even roughly how we can distinguish between such pairs of counterpossibles in a principled way.

    Second, I think it would be useful to spell out the nature of presupposition (if not here) in a more detail. Pretty clearly, from the point of view of classical Christian theism, that God exists is a necessary condition of any proposition. This isn’t a consequence of TAG, but rather, according to Christian theism, of it’s being a necessary truth that God exists. But surely there’s more to presupposition than that. (I think I just felt David Byron’s ears prick up.)

    Finally, I agree that arguments based on performative inconsistency aren’t (as such) question begging in any relevant sense. But it doesn’t follow that the conclusions of such arguments are true more generally. E.g., I can write that I’m not writing anything and thereby write something true. But it’s perfectly possible outside the scope of that performance (say, five minutes from now) that I not be writing anything. Is the scope of the performance relevant here? Is performance as such even relevant? Isn’t God a condition for the possibility of baseball? (Hopes for sprint spring eternal!) But we needn’t be playing baseball to argue the point.

    Anyway, James, it is a pleasure to read your posts, as always!


    1. seriousactualist


      I can’t write that I’m not writing anything and thereby write something true.


      1. seriousactualist

        Well, for one thing, because James is a Christian theist. As am I. But of course much of what I wrote would presumably hold for other theisms which affirm God’s necessary existence.

    2. Thanks for the comments, Paul. Always stimulating.

      Regarding Copan’s charge, I think you’re being too generous. Note exactly what he said:

      It begins with the assumption that God exists, and then concludes that God exists. Such reasoning would get you an “F” in any logic class worthy of the name!

      I agree that the informal circularity you describe is dialectically problematic. But it often takes some reflection to see the problem when it exists. As you imply, it isn’t even straightforward to precisely specify this kind of circularity, never mind to identify every instance of it. Would such reasoning deserve an “F” in any logic class worthy of the name? (Perhaps you’re a tough grader!)

      Your point about counterpossibles is well taken, but would take us too far afield to discuss here. I agree that the “standard” counterfactual semantics are inadequate to capture these non-trivial truths.

      1. seriousactualist

        Thanks for the reply, James. I did read Copan’s post, and while I do agree that the ‘F’ comment is over the top, I’m trying to be as charitable as I can. Perhaps part of the difficulty is focusing too narrowly on the structure and content of actual presuppositionalist arguments (e.g., TAG). For with argument (as elsewhere), there is an important distinction between act and achievement — here between the act or performance of offering an argument, of advancing reasons in favor of some claim, and the argument in fact offered. And with respect to the act, it matters to whom the argument is given, what else they know (or believe), and so forth. In this connection, I heartily recommend chapter II (Proofs of God’s Existence) of George Mavrodes’ little book, Belief in God: A Study in the Epistemology of Religion (Random House, 1970), in which this is helpfully discussed and proof (for a person P) is characterized in terms of an argument’s being convincing (to P).

        Consider the following argument of the sort Mavrodes’ employs throughout his chapter:

        Either 7 + 5 = 13 or God exists
        It’s not the case that 7 + 5 = 13
        :. God exists

        I think it might be a useful exercise to say exactly why this argument isn’t TAG.

  10. Pingback: What I Read Online – 03/15/2012 (p.m.) | Emeth Aletheia

  11. Pingback: The Gospel Coalition on Presuppositional Apologetics

  12. Pingback: Does Presuppositionalism Engage in Question Begging: James Anderson

  13. I wouldn’t be adverse to saying something like this: presuppositionalism is the claim that everyone begs the question. Van Til, I think, would say that the Triune God, as creator of all things but himself, is the state of affairs which is necessary for the intelligibility of the proposition God exists, or any other proposition, so that even when someone asserts the existence of God, he has done so both directly and indirectly. But then the one who denies the existence of God has done two things simultaneously: he has indirectly affirmed the existence of God (given the claim that the intelligibility of the proposition presupposes the existence of God) and at the same time he’s directly denied that existence. We could also say that the pretended neutrality of reasoning also begs the question in two ways: first, as in the presuppositional claim that rationality is only possible given triune theism, but also in the sense that it claims to reason without reference to triune theism, which is the claim that the denier ends up ‘proving’. So in both those senses the denier, the neutral reasoner, begs the question. I think Van Til might say something like that. So then TAG obviously begs the question, but that’s only because it makes explicit what presuppositionalism claims is implicit in everything.

    1. Nate,

      I think I understand your point, but I question whether it’s the best way to put it. “Begging the question” typically refers to fallacious argumentation. How does it really help presuppositionalism, faced with the charge of question-begging, to say, “Yes, we do indeed beg the question, but so does everyone else!” That sounds like excusing one fallacy (petitio principii) with another fallacy (tu quoque)!

      If by “begging the question” you mean something other than fallacious circularity, why use that terminology, since it only serves to fuel the sort of objections we’ve seen from Dr. Copan? Better to speak of epistemic circularity or (as I prefer) transcendental circularity.

      1. You’re probably right, yes. I still there is room for saying that, as Van Til puts things, the person who presents arguments against the existence of God assumes what he means to prove. But at the end of the day I’m with you: “transcendental circularity” might be a better snapshot of the broader issues involved.


  14. Hezekiah Ahaz

    I’m gonna go with Greg Bahnsen on this one, I did say this the other day but I just wanted to expand, the proper answer should be “So, what if were “begging the question”? Really what’s the alternative? Why should I give up my ground?

    I’m wondering if Dr.Copan really starts with the “falsity” of the bible. I highly doubt it.

    1. Yes, absolutely: so what? I think exerting too much energy worrying about a charge of rhetorical cherry picking is evidence we’ve missed the real strength and depth of presuppositionalism, at least as Van Til was conceiving of it. anyway, if the worst someone can say about presuppositionalism is the narrow and uninteresting charge that it begs the question, I don’t think they’ve read or thought enough about it to warrant bothering with. Sorry to sound so harsh, but really–no one who makes that charge–and so flippantly–understands the first thing about presuppositionalism.

  15. Hezekiah Ahaz


    I agree!!

    ““To admit one’s own presuppositions and to point out the presuppositions of others is therefore to maintain that all reasoning is, in the nature of the case, circular reasoning. The starting-point, the method, and the conclusion are always involved in one another” (DOF 118). – Van Til

    People act like were saying “God exists because God exists”. Anyway, I don’t pay much attention to that stuff.

    Here is a great Article on Van Til if anybody is interested:


  16. Pingback: Are Epistemological Transcendental Arguments for Theism Circular? | Thomas Larsen

Comments are closed.