Of all the important differences between the epistemological positions of Clark and Van Til, one of the perhaps less well-publicized elements is their understanding of the structure of knowledge and justification. On the one hand, both certainly held to a coherence theory of truth with respect to God's knowledge. On the other hand, when it came to the human knowledge of justification, both were foundationalists. Of course, God does not quite have a theory of epistemic justification, since He does not require one external to Himself. Our very own human foundationalist epistemology is to be based on His self-revelation, which consists of what He has always known from eternity past, and cannot help but know, according to the incommunicable attribute of His omniscience. But what about the question of what structurally makes something true?
What properties make a proposition true? How is knowledge (as opposed to epistemic justification) structured? Clark and Van Til seem to have differed on this. Van Til seems to have held to a kind of correspondence theory of truth. Indeed, it is precisely the asymmetry between the fact that it is coherence which gives knowledge its "truth" in t he mind of God, whereas it is correspondence with God's thought of it which gives it the property of truth in the human mind, which causes Van Til's epistemology to in some respects bear such a disturbing resemblance to certain idealist philosophies, and which constitutes one of the central manners in which, in Van Til's thought, what gives knowledge the property of "truth" in God's mind must be qualitatively different from what gives knowledge "truth" to the human mind. Perhaps Clark's belief that the human must be a coherentist no less than God was his attempt at consistency and the preservation of the possibility of partial knowledge to which he aspired.
In a way it might be well for us to call our position the Coherence Theory of Truth because we claim to have true coherence. Whether we call our position a correspondence theory or whether we call it a coherence theory, we have in each case to distinguish it sharply from the theories that have historically gone by these names. Accordingly, the determining factor must be a consideration of that which is most fundamental in our theory of correspondence or of coherence. Now this depends upon the question whether we have God’s knowledge in mind first of all, or whether we begin with human knowledge. For God, coherence is the term that comes first. There was coherence in God’s plan before there was any space-time fact to which his knowledge might correspond, or which might correspond to his knowledge. On the other hand, when we think of human knowledge, correspondence is of primary importance. If there is to be true coherence in our knowledge there must be correspondence between our ideas of facts and God’s ideas of these facts. Or rather we should say that our ideas must correspond to God’s ideas. Now since we are dealing with opponents who speak of human knowledge almost exclusively, we can perhaps best bring out the distinctiveness of our position by calling it the Correspondence Theory of Truth. An additional reason for this choice is that at the present time the old correspondence theory has pretty well died down, leaving the coherence theory in control of the field. Hence we have the advantage of a different name from the current name, since we are interested in making it clear that we really have a different theory from the current theory(Van Til, "A Survey of Christian Epistemology").
With respect to human knowledge, Clark was a foundationalist:
Now since it is the conclusion of a demonstration that we are trying to prove, and since it is proved by giving the premises, it follows that the premises of demonstrated knowledge are better known than the conclusion. If we did not know the premises, obviously we could not know the conclusion. The conclusion cannot be more certain than the premises on which it is based. The premises are the cause of the conclusion, and therefore they must be prior to it. And also, in demonstration, although not in every formally valid syllogism, the premises must be true. For demonstration is knowledge, and there can be no known of the non-existent. The premises, therefore, must be statements of what exists or what is so, i.e., they must be true.
Of course, there may be a chain of syllogisms in a demonstration, as there is in geometry. But the chain must have a starting point, and such a starting point must be, not only prior, causal, and true, but in particular primary and indemonstrable. It must be an immediate, basic truth. Nothing can be more certain than these basic truths, for if the least doubt attached to them, doubt would likewise attach to all the conclusions; and this would mean that science would be tottery. But the conviction of pure science must be unshakable.
In the nineteenth century it was commonly believed that science was as unshakable as Aristotle could have wished; but the prevailing mood of the twentieth century is that science is tentative, and that laws stand in need of constant revision. Therefore, the current objection to Aristotle is that the science which he describes is non-existent. The formal validity of syllogisms may possibly be foolproof, but their applications to concrete material, and more especially the premises on which they are based, are not completely beyond all doubt. To Aristotle this would mean that there is no scientific knowledge, as he defined knowledge. There was a similar difference of opinion in his day. Some said there is no knowledge; others said all truths are demonstrable. But Aristotle agreed with neither the one nor the other.
Those who denied the existence of scientific knowledge argued that demonstration is the only method by which something can be known. But demonstrations depend on premises. And if the premises are to be known, they too must be demonstrated. This leads on back in an infinite regress, with the result that the demonstration is never finished, or more accurately, never begins. Accordingly, there is no scientific knowledge. The other group also held that demonstration is the only method by which anything can be known; yet they held that everything can be demonstrated because proof goes around in a circle: Every premise is a conclusion, and there is a finite series in which the end and the beginning are identical. Aristotle replies that a proposition cannot be both prior and posterior as this view requires. Since the exact number of terms is irrelevant, they may be reduced to three and the absurdity becomes apparent. Circular demonstrations would be equivalent to saying that A is B; Why? – because B is C; Why? – because C is A; Why? – because A is B. With circular and infinite demonstration both ruled out, it follows that not everything can be demonstrated and that there must be first, indemonstrable truths.
A philosopher of a different school, Hegel for instance, would no doubt admit that the three-term circle is an absurdity; but he might argue that the exact number of terms is no so irrelevant as Aristotle thought. A bad circle is a little circle; but if a circle can be drawn so as to include everything, it is a beautiful circle. In a rational universe everything is implicated in everything else; and precisely for this reason a three-term circle is absurd: It fails to show the other relationships of A, B, and C. Hegel might even attribute some very small and very bad circles to Aristotle himself: He might ask, Is Aristotle’s reply anything more than a two-term circle, in which demonstration is possible because there are primary truths, and there are primary truths because there must be demonstration?
At any rate, against the two views, Aristotle asserts that not all knowledge is demonstrative. There must be primary basic truths because the regress in demonstration must end in these basic truths, and these are indemonstrable. Therefore, besides the scientific knowledge, which is demonstration, there is its originative source which enables us to recognize the basic indemonstrable propositions. (Thales to Dewey, 2000, pgs. 101-102)
Contrast this foundationalist theory of justification with the coherentist theory of truth which he ascribes to God:
The substantive point needing discussion is whether the law of contradiction is the one and only test of truth.
Ideally or for God this seems to be the case. Since there is nothing independent of God, he does not conform truth to an alleged reality beyond truth and beyond him. Since there is no possibility of “vertical” (to use Carnell’s terminology) coherence, the “horizontal” test, or, better the horizontal characteristic of logical consistency seems the only possible one.
Weaver correctly notes that I do not claim for human beings the ability to apply this test universally. In this sense it is a “negative” or, better, an incomplete test. For this reason it must be supplemented some way or other. (Clark and His Critics, 2009, pg. 287)
Undoubtedly I hold that truth is a consistent system of propositions. Most people would be willing to admit that two truths cannot be contradictories; and I would like to add that the complex of all truths cannot be a mere aggregate of unrelated assertions. Since God is rational, I do not see how any item of his knowledge can be unrelated to the rest. Weaver makes no comment on this fundamental characteristic of divine truth.
Rather, he questions whether this characteristic is of practical value, and whether it must be supplemented in some way. It is most strange that Weaver here says, “I must agree with Carnell,” as if he had convicted me of disagreeing with Carnell by providing no supplementation whatever. Now, I may disagree with the last named gentleman on many points, but since it is abundantly clear that I “supplement” consistency by an appeal to the Scripture for the determination of particular truths, it is most strange that Weaver ignores my supplementation. (Clark and His Critics, 2009, pg. 290)
One who believes in the unity of truth may still believe that the false system entails contradictions; but to prove this is the work of omniscience. (Historiography: Secular and Religious, 1971, pg. 370)
So we have seen that both men were foundationalists in their theory of justification of human knowledge, and coherentists in their theory of divine knowledge. We have also seen that Van Til held to a kind of correspondence theory of truth with respect to human knowledge. For Van Til, what makes something "true" in the mind of God is its coherence with other facts, and what makes something "true" in the mind of man is its correspondence with God's interpretation of it. But what was Clark's theory of truth for human knowledge?
It would appear that for Clark, humans, like God, are bound to a coherence theory of truth. However, Unapologetica argues that, while Clark explicitly advocated a coherentist theory of truth for humans, his actual beliefs about linguistic predication commit him to a correspondence theory of truth. Unapologetica quotes Clark from Christian Philosophy, from the essay Inspiration and Language:
In one sense of the term, a photo corresponds to or looks like its object, but no one supposes that a word corresponds to a thing in this way. Language is not a picture of reality. The letters c-a-t do not look like the purring animal. It is all the more true that words cannot possibly look like spiritual realities, if such there be, for these are not visible entities. But in a non-photographic sense a mathematical formula may be said to correspond to the motion of a freely falling body. Could not this be an absolute correspondence? Or, if the term absolute causes hesitation, could not such a formula be or be understood as a literal assertion? Further, if the sound cat is essentially an arbitrary sign of the animal, what more correspondence could be desired? (pgs. 194-195)
And while there may be some meaning embedded in the language of a man whose ideas are not clear and distinct, the meaning would surely prove to be an hallucination if it could be shown that the words could not be made to correspond to some clear or distinct ideas. Furthermore, how can one construct a parable that relates a known object to something of which we have no concept at all? Meaningful analogies and honest comparisons can be made only if we know something about both terms. Unless a better defense of religious language and thought can be devised, the Logical Positivists, will not be greatly embarrassed. (pg. 203)
Therefore, it would seem that Clark's views about univocal predication commit him to a correspondence theory of truth. Nevertheless, Clark did not seem to have accepted such a theory of truth:
This also seems to me to solve the philosophic problem of truth. The three chief contenders in this field are the correspondence theory, the coherence theory, and the pragmatic theory. For reasons too numerous to include here, I believe pragmatism leads to complete skepticism. The correspondence theory would require us to compare an idea we have in consciousness with some utterly unknown object. This is impossible. The coherence theory remains.
It cannot be charged with skepticism. If it is objected that it requires an impossibility, viz. that a man be omniscient, the reply is that its Hegelian form may involve such an impossibility; but this impossibility does not occur in the Christian system where an omniscient God makes a definite revelation to man. Hence, it is not necessary for a man to know everything before he knows anything. The subjective knowledge of any man depends not on his own complex of thoughts but on God’s system and on the fact of revelation. And contrary to what seems to be Dr. Buswell’s opinion, this is not at all inconsistent with God’s sovereign grace(Clark, "System and Induction").
Clark, by process of elimination, concludes that the coherence theory of truth is the only intelligible option for the Christian. Like Van Til, Clark had to defend his coherentism from the charge of the sort of skepticism associated with the idealists of his day (as coherentism was associated with idealism). He argues that knowledge is nonetheless possible because God makes a definite revelation to man. It is the fact that God thinks in propositions whose content is bound by the law of excluded-middle -- something which idealists such as Bradley emphatically reject -- that partial knowledge is possible for man. But does this not commit the human to a correspondence theory of truth? If a truth in God's mind depends for its meaning on its relations with other propositions, must be not therefore know everything if we are to know anything? Of course, not all forms of coherentism teach this. Rather, propositions are better thought as being logically entailed by other propositions in relation to which they have coherence within a definite set of propositions:
A more plausible version of the coherence theory states that the coherence relation is some form of entailment. Entailment can be understood here as strict logical entailment, or entailment in some looser sense. According to this version, a proposition coheres with a set of propositions if and only if it is entailed by members of the set
Thagard's argument seems to be that if there is a mind-independent world, then our representations are representations of the world. (He says representations “should be” of the world, but the argument is invalid with the addition of the auxiliary verb.) The world existed before humans and our representations, including our propositional representations. (So history and, Thagard would likely say, our best science tells us.) Therefore, representations, including propositional representations, are representations of a mind-independent world. The second sentence of the passage just quoted suggests that the only way that coherentists can reject this argument is to adopt some sort of idealism. That is, they can only reject the minor premiss of the argument as reconstructed. Otherwise they are committed to saying that propositions represent the world and, Thagard seems to suggest, this is to say that propositions have the sort of truth-conditions posited by a correspondence theory. So the coherence theory is false.
In reply to this argument, coherentists can deny that propositions are representations of a mind-independent world. To say that a proposition is true is to say that it is supported by a specified system of propositions. So, the coherentist can say, propositions are representations of systems of beliefs, not representations of a mind-independent world. To assert a proposition is to assert that it is entailed by a system of beliefs. The coherentist holds that even if there is a mind-independent world, it does not follow that the “the point” of representations is to represent this world. If coherentists have been led to their position by an epistemological route, they believe that we cannot “get outside” our system of beliefs. If we cannot get outside of our system of beliefs, then it is hard to see how we can be said to represent a mind-independent reality(Young, 2013).
For Christians, there is no such thing as a mind-independent world. There is indeed an objective world, but nothing is independent of the mind of God. The objective world to which we refer is the mind of God, and the mind of God to whose propositions we refer constitute a part of the sum total of a system of propositions. In this respect, one can consistently ascribe to man, no less than to God, a "coherence" model of truth.
Clark and Van Til would only be committed to a strict correspondence theory of truth if they were non-theists. The reason for this is, because they believed that the objective world is not mind-independent, but instead that all truth is God's truth and therefore emphatically mind-dependent, that is, dependent upon the sum total of God's knowledge, true propositions always necessary refer to the metaphysical absolute of the system of truth in God's mind. Mere correspondence does not a correspondence theory of truth make. Since the metaphysical ground which makes correspondence of the thoughts of the human mind true, is precisely correspondence to truth existing within a coherent set of thoughts in the mind of God, their theory of truth is not necessarily committed to a strict correspondence theory at all. The reason for this is that the metaphysical quality that makes the proposition "true" is its coherence in the mind of God. Furthermore, a proposition represents not merely an isolated belief, but a holistic system of propositions each of which logically entails the other. That we do not know all the propositions of this system, Clark argues, of course, does not mean that we cannot have real partial knowledge.
Indeed, I believe that it is precisely when one's theory of justification (with respect to human knowledge) becomes formally coherentist rather than foundationalist, that the Van Tillian presuppositionalist apologetic strategy of attempting to prove the internal incoherence of the non-Christian through the transcendental argument for the existence of God, risks succumbing to rationalism. I believe that in refusing to acknowledge that there are at least some some non-Christian theistic position against which the transcendental argument cannot be used, the Van Tillian unconsciously lapses from a foundationalist theory of justification to a coherentist theory of knowledge, and that the latter constitutes the ground of its lapse into rationalism.
But why is this not the case when it is used against unbelievers? Because the problem with the unbeliever is that he lacks a definitive foundation upon which to base his epistemic claims. His epistemology is one of radical immanence, in which some standard within creation is taken as his or her epistemic foundation. But this foundation is totally arbitrary, and demonstrably so, for the Van Tillian. The transcendental argument is therefore quite useful in such a case, because it is easily demonstrable to the unbeliever, for example, how the unbeliever cannot derive something like moral norms from a fact within creation.
But suppose the Van Tillian comes up against a Muslim. The Muslim does have a transcendent reference point that is beyond creation; at his in his or her own mind. This transcendent reference point constitutes the ground which conditions all epistemic and moral norms. The Van Tillian therefore cannot criticize the Muslim for adopting an arbitrary, immanent foundation in the same way that they can the unbeliever. Since they cannot use foundationalist arguments in order to problematize the arbitrariness and internal consistency of the unbeliever's worldview, they must adopt a coherentist strategy.
Why is this so? Because for the unbeliever, it is easy for the Van Tillian to demonstrate that even according to their own standards, they cannot account for the sort of phenomena which they desire to have in their worldview. The reason for this is a lack of a foundation which would provide the preconditions for such beliefs. But the Muslim does have such a transcendent conditioner which provides such transcendental preconditions. The argument therefore does not work for the Van Tillian. It is for this reason that they must switch from a foundationalist polemics to a coherentist one.
While they can easily and effectively critique the unbeliever by their own standards of rationality, the fact that the Muslim, unlike the unbeliever, has a transcendent ground which can provide the preconditions of rationality, intelligibility and consistency for the sort of phenomena they want to be able to account for, they have definite standards of such rationality, intelligibility and consistency, and unlike the unbeliever, are able to reject the antithetical arguments of the Van Tillian as foreign to their own transcendence-provided epistemic norms. It is at this point that the Van Tillian switches from his foundationalist critique of the unbelievers lack of norms even according to his own standards, to a coherentist critique of the Muslim's internally inconsistent norms according to Christian standards which the Muslim is committed, according to his own transcendence-provided epistemic norms, to reject.
Young, James O., "The Coherence Theory of Truth", The Stanford Encyclopedia of Philosophy (Summer 2013 Edition), Edward N. Zalta (ed.), URL = <plato.stanford.edu/archives/sum2013/entries/truth-coherence/>.