The question arises in the context of the structure of human thought. It bears its own unique dilemma. If we stress too excessively that knowledge of God is univocal we run the risk of lowering the incomprehensible God to the level of the finite and make God as one of us. But if we stress too emphatically knowledge of God per analogiam we may very well deprive God of all likeness to the humanity he has created with the result that all we are left with is a barren, abstraction.
To a considerable extent the argument took fertile shape in the debate of the 1940s between Cornelius Van Til and Gordon H. Clark. At issue was God’s incomprehensibility with both men claiming biblical precedent minus the faux pas of excess. It is a matter of regret that we can only provide an abbreviated review of the controversy. Van Til was jealous to protect the Creator/creature distinction both in reality, or what Frame calls “two-levels of reality,” and in knowledge. “All [man’s] knowledge is analogical of God. God is the original knower and man is the derivative re-knower.” John M. Frame stands with Van Til. Thus “To be a creature is to be limited in thought and knowledge.” And “God’s thoughts are the originals of which ours, at best, are only copies, images.” Earlier, Herman Bavinck said, “There is no knowledge of God as he is in himself. We are human and he is the Lord our God . . . he infinitely transcends our picture of him, our ideas of him, our language concerning him. He is not comparable to any creature.” This led Bavinck to verify a staple of Continental Reformed thought: the archetype/ectype distinction. “Of course, all our knowledge of God is ectypal or derived from Scripture. Only God’s self-knowledge is adequate, underived or archetypal.”
Clark, on the other hand, feared that if man’s knowledge was derivative of God’s then this would lead to skepticism. “It seemed to him that if there was some discrepancy between man’s ‘This is a rose’ and God’s (concerning the same rose). Then the human assertion must somehow fall short of the truth, since the very nature of truth is identity with God’s mind.” Van Til, however, insisted that analogical sapentia though incomprehensive and derivative is nonetheless true knowledge. “We may safely conclude then that if God is what we say he is, namely a being who exists necessarily as a self-complete system of coherence, and we exist at all as self-conscious beings, we must have true knowledge of him . . . All this we express theologically when we say that man is created in God’s image. This makes man like God and assures true knowledge of God. We are known of him and therefore we know him and know that we know him. God is light and therefore we have light.”
In addition to our knowledge being analogical, Van Til held that it is also partial—or as he used to say, “non-exhaustive.” Phrased in a question, “What is the correspondence between God’s self-knowledge and the created reflection known by man?” Strictly speaking, does ectypal theology contain everything that is contained in its archetypal counterpart or do we know only what God wants us to know?” According to Clark, God knows more than we do, but the quality of our knowledge can be the same as his. Van Til argued that both the quality and the quantity of God’s archetypal knowledge differ from our ectypal knowledge.
Finally, both men agree that all knowledge is given by natural and special revelation. But for Clark, that revelation is propositional in the sense that it always conveys univocal truth. Per Van Til, God’s revelation to man is accommodated truth. Accommodated revelation is coherent to man, but true coherence does not afford identity of content between the revelation God gives and what man receives. “Van Til, therefore, insisted that even when God and man were thinking of the same thing (a particular rose, for example), their thoughts about it were never identical—God’s were the thoughts of the Creator, man’s of the creature.” Here we see a precursor to Frame’s covenantal lordship principle. Because man cannot know the essence of God one-to-one, but always and forever as a servant (analogically), even now in the face of Scripture “Man knows in subordination to God; he knows as the covenant-keeper.” Scripture, then, is better thought of a “system of truth.” This is really what Van Til meant by “analogical”—our knowledge must accept God’s revelation as the ultimate standard of truth.
Clark’s seminal idea is that “The intelligibility of the Scriptures presupposes logic.” Clark interprets and paraphrases John 1:1 to mean, “In the beginning was Logic, and Logic was with God, and Logic was God . . . In logic was life and the life was the light of men.” Clark, then, makes the stunning admission that “Logic is God.” Clark does not mean that human logic is God, but that God thinking is logic and that the connection of his logic to ours is univocal. By virtue of this, knowledge of God in se is possible. The source of univocal knowledge is the Bible. Contra empiricism, Clark maintained a very wary view of what we can know through sense experience, arguing instead that knowledge of God was what we learn from Scripture. So then there is a direct correlation between God’s thought and Scripture. “What is said in Scripture is God’s thought” and since our logic is univocal of God’s there exists a direct continuum from God’s thought, to Scripture, to us. Van Til and Frame deny such an epistemological continuum between man and God.
Anticipating a critique of his views based on Isaiah 55:8-9, which stipulates that God’s thoughts are not our thoughts, Clark replies, “If for example, we think that David was King of Israel, and God’s thoughts are not ours, then it follows that God does not think David was King of Israel. David in God’s mind was perchance prime minister of Babylon.”
In an essay, Nathan Pitchford succeeds in countering Robert Reymond’s exact exploitation of Clark’s example. “On the contrary, given the basic legitimacy of the system, it is entirely possible to prove that God’s knowledge is in actuality analogical to what we understand by Nebuchadnezzar’s being King of Babylon; because God himself framed the symbolical/analogical essence of human language as well as the corresponding and uniform human apprehension of that absolute knowledge, he is entirely capable of making the analogical an accurate reflection of the absolute.”
Pitchford charges Clark and Reymond with twisting the Nebuchadnezzar motif. Their point is at base unfair because Van Til nowhere suggests that actual reality and analogical reality are substantially different, as their critique portrays. “On the contrary, Van Til means that every human perception of actual reality (uniform and consistent to the whole race), although not absolute, nevertheless reflects that pure truth in a unique way that corresponds precisely to that real truth, much as the image in a mirror corresponds precisely to the actual thing it reflects, even though it is not that thing . . . Thus, analogical knowledge does not cast one into the morass of ambiguity as Reymond suggests.”
We agree with Pitchford's assessment wholeheartedly and, generally speaking, side with the analogical interpretation of human knowledge as expressed by Van Til.
 “Christians believe in two levels of existence as derived from the level of God’s existence as self-contained and the level of man’s existence as derived from the level of God’s existence. For this reason, Christians must also believe in two levels of knowledge, the level of God’s knowledge which is absolutely comprehensive and self-contained, and the level of man’s knowledge which is not comprehensive but is derivative and re-interpretive” [Van Til, An Introduction to Systematic Theology, vol. 5 In Defense of the Faith, 12].
 Cornelius Van Til, An Introduction to Systematic Theology, vol. 5 In Defense of the Faith, 167.
 Frame, The Doctrine of the Knowledge of God, 21.
 Frame, The Doctrine of the Knowledge of God, 23.
 Herman Bavinck, Reformed Dogmatics, Vol 2. (Grand Rapids: Baker Academic, 2009), 47.
 Ibid. 95. Peter J. Wallace, in a review of Timothy Phillips’ work on Francis Turretin’s (1623-1687) theological method, notes that “Crucial to Turretin’s discussion of theology is his distinction between archetypal theology and ectypal theology; and the further distinction between the ectypal theology of vision, which is the theology of the saints in heaven, and the ectypal theology of the traveler—‘the theology of revelation’—with which we must remain content” [Peter J. Wallace, “The Doctrine of the Covenant in the Elenctic Theology of Francis Turretin,” http://www.peterwallace.org/turretin.txt.The Turretin reference is in the first volume of his Institutes of Elenctic Theology (Phillipsburg, NJ: P&R Publishing, 1992), 4-5.] The distinction was also significant to Amandus Polanus (1561-1610). “Archetypal theology is the exemplar: ectypal theology is the exemplum, which ought to agree with, correspond with, and resemble the exemplar. Thus ectypal theology is, in rational creatures, a part of the image and likeness of God according to which they were created” [Quoted by Richard A. Muller, Post-Reformation Reformed Dogmatics, Vol. 1 (Grand Rapids, MI: Baker Book House, 1993), 135]. In Divine Discourse by Sebastian Rehnman, the author notes the use of the archetypal/ectypal theology distinction in the puritan divine, John Owen (1616-1683), in which God’s knowledge of Himself (in se) is understood to be related to, but different than, our knowledge of God (pro nobis). In his Reformed Dogmatics, Heppe says of the older dogmaticians that “Regularly they discuss the distinction between theologia archetypa (i.e. ‘the knowledge which God has of Himself and in Himself’) and theologia ectypa (i.e. ‘man’s knowledge of God’). The latter they divide into theologia beatorum and theologia viatorum, the second being theology proper” [Heinrich Heppe, Reformed Dogmatics, (1861, repr., Grand Rapids, MI: Baker Book House, 1978), 5].
 Frame, The Doctrine of the Knowledge of God, 22.
 Bavinck puts it like this, “Ectypal knowledge must not be seen as merely symbolic, a product of poetic imagination . . . While our knowledge of him is accommodated and limited, it is no less real, true, and trustworthy” [Herman Bavinck, Reformed Dogmatics, Vol 2. (Grand Rapids: Baker Academic, 2009), 95]. On the fact that accommodated knowledge is nonetheless true knowledge, R. Scott Clark, says, “Ectypal theology, as the adjective suggests, is a reflection of the archetypal theology. It is true, but it is accommodated to human creatures” [R. Scott Clark, foreword to The Free Offer of the Gospel, John Murray].
 Cornelius Van Til, Defense of the Faith (Phillipsburg: P&R Publishing, 1967), 57.
Frame, The Doctrine of the Knowledge of God, 21-22. In an extreme polemic against univocal reasoning, Van Til therefore states, “The distinguishing characteristic between every non-Christian theory of knowledge on the one hand, and the Christian concept of knowledge on the other hand, is, therefore, that in all non-Christian theories men reason univocally, while in Christianity men reason analogically” [Van Til, An Introduction to Systematic Theology, vol. 5 In Defense of the Faith, 11]. Similarly, Augustine gave expression, “We are speaking of God. It is any wonder if you do not comprehend? For if you comprehend, it is not God you comprehend. Let it be a pious confession of ignorance rather than a rash profession of knowledge. To attain some slight knowledge of God is a great blessing; to comprehend him, however, is totally impossible” [Augustine, Lectures on the Gospel of John, tract. 38, NPNF (1), VII, 217-21].
 Van Til, An Introduction to Systematic Theology, vol. 5 In Defense of the Faith, 167.
 Van Til, An Introduction to Systematic Theology, vol. 5 In Defense of the Faith, 173.
 Indeed, Van Til often says that unless knowledge from Scripture is analogical, there is no possibility of knowing the truth; for all truth is a revelation of the incomprehensible God. The man who thinks he comprehends truth as God does, has made himself out to be God, which in turn precludes the possibility of his knowing the truth.
 Gordon H. Clark, An Introduction to Christian Philosophy. 2nd ed. (Jefferson, Maryland: The Trinity Foundation, 1993 ), 64.
 Ibid. 67.
 Ibid. A favorite retort of Clark’s focused on the consequences of the analogical position that verified coherence but not content. So he asked if God has a different arithmetic, in which 2+2=5? [Ibid. 76]. Van Til replied that to know 2+2=clearly, as Clark suggests, is to know it exhaustively, which violates the incomprehensibility of God. But we can know that 2+2=4 on the basis of an “identity of reference point” [Van Til, An Introduction to Systematic Theology, vol. 5 In Defense of the Faith, 167].
 Gordon H. Clark, An Introduction to Christian Philosophy. 2nd ed. 77.
 Followers of Clark are thus given to say that there are no paradoxes or discrepancies in Scripture. If such are thought to exist it is only due to a breakdown in the logical and exegetical study of Scripture. See W. Gary Crampton “Does the Bible Contain Paradox?” The Trinity Foundation (November/December 1990). http://www.trinityfoundation.org/journal.php?isbn=76.
 Clark, An Introduction to Christian Philosophy. 76.
 Nathan Pitchford, “Van Til: His Logic, Epistemology, and Apologetic,” http://www.reformationtheology.com/2006/06/van_til_his_logic_epistemology.php.
 Nathan Pitchford, in a personal email to the author, dated September, 28th, 2010.