Can “TAG” Get Us to Epistemic Certainty?

Back in 2011, Dr. James Anderson wrote a blog post arguing that “TAG” does not (maybe: cannot) provide us with “epistemic certainty”.  He provides three reasons (along with prefatory comments) why he holds this view.  I’ll try to briefly interact with each of his three reasons, but first, a question:

He says many of TAG’s advocates have suggested that it *can* provide “epistemic certainty”.  He takes this for granted (and it’s never challenged in the comments), but I wonder whom he has in mind, exactly?  Van Til was not analytically precise in his language and doesn’t use the “epistemic certainty” jargon (as far as I’m aware).  Given how ambiguous his writing is, and given how the presuppositionalist community has evolved (such that Van Til is taken in certain ways, regardless of what he originally wrote), we might do better to look to Dr. Bahnsen for a clearer statement.  Dr. Bahnsen, though, was notorious for blending technical jargon with his preaching, sometimes flowing back and forth between philosophical analysis and helpful metaphors for his lay-audiences.  Unfortunately, he doesn’t use the jargon of “epistemic certainty” either (again, as far as I’m aware) in the same sense that it’s used in contemporary analytical philosophy.

So where does James Anderson get the idea that TAG advocates argue that TAG provides “epistemic certainty”?  Well, the presuppositionalist community has thrived among a lay-audience, many of whom (like the notorious Sye Ten Bruggencate) operate on talking-points and simplified arguments they’ve learned from popularizers like Dr. Bahnsen.  And Dr. Bahnsen *did* insist that arguing presuppositionally leaves us better off somehow than arguing “evidentially” or “classically”.  The following statement is typical and probably most apt to our present considerations:

“I began yesterday with the opening salvo that I believe the existence of God is objectively provable.  Then I took that and started to do an analysis to qualify and explain what I’m getting at.  And we took awhile to talk about the notion of proof.  When I say “objectively provable”, what am I getting at?  By a proof that is “objective”: not interested in the man or source of opinion or how people personally respond or so forth.  We’re talking about proof, not persuasion.  We’re recognizing the difference between metaphysics (what is the case) and epistemology (how people think, what they believe and how they know and things of that nature).

Then after we discussed the notion of proof, we began to look at the kinds of proofs that are available.  The Cosmological Proof was taken as our stalking horse.  I said the Cosmological Proof is bad as a proof.  Really bad philosophically as a proof.  However there’s something to that, this notion of causation, that if you go back and now do a transcendental analysis of it, does turn out to be, I think, a very strong proof for the existence of God.  So now we have the notion of transcendental proof…”¹

Additionally, much presuppositionalist rhetoric suggests there is a moral failing in evidential and classical arguments because these arguments only arrive at probabilistic conclusions instead of definitive truth claims.²

But must the special status often claimed for transcendental arguments, be construed as “epistemic certainty”?  Might we not maintain popular ideas about transcendental arguments (that they provide us with “objective proof” and that arguing otherwise may be immoral) while still bowing knee to secular philosophical truisms (which say no argument can provide “epistemic certainty”?)

I suspect James Anderson could easily do so if he wanted.

But on to his three reasons:

His first reason for rejecting the idea that TAG provides “epistemic certainty” is that TAG must be thought of as a group of arguments instead of a single one.  So before analyzing the claim that TAG provides epistemic certainty, Anderson would need to know “which TAG” is being discussed.

This is readily granted, but I do wonder how diverse Dr. Anderson thinks TAGs can possibly be.  It seems that if the transcendental thrust of the argument is successful, the formal “trappings” (be they modus ponens, modus tollens, or what have you) are secondary considerations.  And I’m not sure there can be different types of transcendental “thrusts” such that were Dr. Anderson to successfully critique one, a presuppositionalist might run to another.

————————————-

His second reason for rejecting the idea that TAG provides “epistemic certainty” is, he’s (supposedly) unaware of any formulation of TAG that proves specifically Christian theism.

Of course, this is a cheeky blow to the lay Presuppositionalist community.  We might easily interpret him here as smacking us around and telling us to wake up from our Bahnsonian delusions.  But in his seminar on Transcendental Arguments, Dr. Bahnsen distinguishes “Van Tillian-styled” transcendental arguments from secular transcendental arguments, by noting that the former are “worldview” in scope, while the latter, argue piece by piece, or proposition by proposition.

Van Tillian TA’s then, says Bahnsen, are distinct precisely because they argue that the entire *Christian* worldview is the necessary precondition of human intelligibility.  Granted, when we demonstrate this, we don’t simultaneously demonstrate that every proposition about the Christian worldview is a necessary precondition of human intelligibility…as if the color of Jesus’ hair, or the number of lions in the den with Daniel were important epistemologically.  But without the entire Christian worldview, all the preconditions of intelligibility provided by Christianity would become “rocks in a bottomless ocean”.

I know Dr. Anderson is aware of this sort of presuppositionalist rhetoric.  To see him dismiss it out of hand, without even an off-the-cuff analysis, is frustrating (to say the least).  I can accept if he thinks it’s wrong to distinguish between Van Tillian TA’s and secular TA’s as Dr. Bahnsen does, or if he thinks Van Tillian TA’s, even when so-construed, fail…but to gloss over the issue as if it’s too unimportant to even mention and to simply conclude that he’s “never” seen a TAG prove *Christian* theism, is … well… again: frustrating.  *ALL* TAGs offered by Van Tillians are meant to prove specifically Christian theism.  But here, Dr. Anderson is implying that despite our best efforts and despite our long nights of apologetics, despite our blood, sweat, and tears, we’re simply failures.

————————————–

His third point is the most important as far as determining if TAG can provide “epistemic certainty”, and so I’ll cite him directly:

Thirdly — and this is the most important point — we need to be clear on what it would take for any argument to deliver its conclusion with epistemic certainty. The strongest type of argument is a deductive argument, in which the conclusion follows necessarily from the premises: if the premises are true then the conclusion must be true. But even in a deductive argument, the conclusion cannot enjoy greater epistemic warrant than the premises (at least in the case where the warrant for the conclusion is taken to derive from the argument itself). What this means is that TAG can deliver epistemic certainty only if all of its premises are epistemically certain.³

He goes on to note that important premises in the TAG argument are not self-evident and thus, the conclusion, even if deductively valid, can never be epistemically certain.

But remember earlier when he suggested there’s no single formulation of TAG?  So it’s not clear if he means, at this point, that *all* formulations of TAG will be guilty of including propositions which aren’t self-evident, or if only some of them do.

Further – it’s at this point we should look to the theologians to help us with our philosophy.  Consider:  Dr. Anderson thinks the following proposition is not self evident:

“God’s existence is a metaphysically necessary condition of some essential feature of human thought or experience”

But often theologians speak of theological truths that, on Christianity, would be self-evident.  The fact that God exists, for example, is often said to be self-evident.  Couldn’t we argue that God’s existence being self-evident also means that certain relationships between God and His creation are also self-evident, even if we can’t immediately articulate (with analytical precision) what the precise nature of this relationship is?

Now, Dr. Anderson might suggest that, even if we have regenerated intuitions about God’s relationship to His creation, we would never be “epistemically certain” that we’ve stated them correctly when formulating a premise for a TAG.  But that simply requires us to learn a little analytical precision … it doesn’t seem to preclude us (especially if we work hard) from articulating a premise that is self evident to every Christian.  So I’m not convinced that God’s being metaphysically necessary for essential features of human thought, is not self-evident, even while granting that it might not always be easy for the average Christian to articulate.

But even supposing it’s self evident, can we say we know it with “epistemic certainty”?  And if not, then Anderson’s point still seems to stand.

But then again (as I pointed out initially), I’m not sure anyone is really concerned with that anyway.⁴  If Van Til (and his disciples after him) relied on Bavinck, then we might consider the following statement from Bavinck’s book “The Certainty of Faith” pg. 74, to be an apt summary of the attitude towards “epistemic certainty” we’d wish to maintain:

“If neither rational argument nor moral experience can explain how the Christian faith comes into being, the question arises whether there isn’t a better way in which man may be lead to trustingly embrace the truth revealed in Christ….This is an unalterable fact.  That the Gospel is made known to us, confronting us with the call to believe and repent does not depend on our will, but on a decree of God. It is He who ordains us to be born of Christian parents, raised in a Christian environment, and without any merit on our part makes us acquainted with the way of salvation in Christ.”

What we want to say then, as presuppositionalists interested in maintaining the Neo Calvinist tradition with Bavinck, is that God can miraculously provide us an epistemically certain Faith (perhaps through some externalist mechanism described by the likes of Plantinga), even while most (I suspect) would be fine with admitting (as James Anderson does) that any argument offered, even one as deductively valid and useful as TAG, can’t get us there.⁵

Faith is an act of God…not of reason.

---

1. This is from the opening comments in lecture 7 of Dr. Bahnsen’s seminar on Transcendental Arguments. Purchase the seminar here: http://www.cmfnow.com/transcendentalarguments.aspx  ↩

2. It’s commonly suggested these apologetic methods are immoral because they, in some sense, “give up God” in order to defend God. For a typical argument, see the last ten minutes of lecture 2 of Michael Butler’s course on Presuppositional Apologetics, available from Sermon Audio.  This aspect of popular presuppositionalist rhetoric doesn’t seem immediately relevant to the topic at hand. ↩

3. In his collection of writings “On Certainty” proposition1, Wittgenstein notes something similar.

When one says that such and such a proposition can’t be proved, of course that does not mean that it can’t be derived from other propositions; any proposition can be derived from other ones. But they may be no more certain than it is itself.↩

4. Critics of Van Til often maliciously note how he would assert one proposition then immediately assert its opposite. For instance: Unbelievers can know things, but unbelievers cannot know anything. In the same way, consider the following citation in light of our discussion:

“…man’s creation in God’s image involves (a) the fact that man’s ideal of knowledge should never be that of the comprehension of God, and (b) the fact that man’s knowledge is nevertheless true.” Intro. to Systematic Theology, pg. 119.

If Dr. Anderson wanted, he might use passages like this from Van Til to correct less-sophisticated Van Tillians when they assert that we might attain “certainty” that can (in principle) rival God’s comprehensive knowledge of a subject. The presuppositionalist community desperately needs leaders to offer these sorts of interpretations of Van Til.↩

5. Interestingly enough, secular philosopher Barry Stroud, (in “Scepticism, Externalism, and the Goal of Epistemology” on page 292 of “Skepticism: A Contemporary Reader” edited by DeRose and Warfield,) suggests that even if a workable externalist account of knowledge is provided, one which, if true, would yield knowledge, the knower would still have an intuitive doubt as to whether his model is actually true.

Ironically, in that same article, Stroud posits a hypothetical “Descartes” who provides a theistic externalist account of knowledge that, to a presuppositionalist, looks exactly like what we’d want to advocate. Stroud arbitrarily rejects it as false, but overlooks the fact that, if this particular externalist model were true, and if it were provided with specifically Christian connotations, it would not be subject to the same sorts of discomforts that his and Ernest Sosa’s non-theistic externalist models would be. A defense of that claim, however, must wait for another blog post.↩