Last December, an article by Daniel Linford entitled "Do Atheists Reject the Wrong Kind of God? Not Likely" appeared at Scientia Salon. Certain recent "popular books,” according to Linford, have advanced a "mystical" notion of God, arguing that contemporary atheists have directed their disbelief only toward "smaller" conceptions of the divine. Three contemporaries are singled out: Karen Armstrong, John Haught, and David Bentley Hart.

On what Linford denominates the "mystical" view, God is radically transcendent, not a being within the cosmic order, and cannot be circumscribed by human language. Many of the atheist assaults are directed against a God who is, more or less, a being among beings, and a person much like us. Linford believes that sophisticated thinkers have sought to outflank such arguments by moving to higher ground, positing a God who escapes our language and ways of thought.

But the higher ground strategy fails, according to Linford, for three reasons. In the first place, it is unclear (to Linford) why God must transcend our language and our concepts, rather than having some other sort of transcendence. Instead of expanding on this objection, Linford merely directs the reader to arguments made elsewhere by Alvin Plantinga.

The second reason the retreat to mysticism fails, according to Linford, is that God’s revelations to us cannot be trusted if we cannot really know God’s character for truthfulness. After all, if we must deny that God is good or true in any sense that we can recognize, we can't very well claim that our defective notion of God's goodness or truthfulness precludes divine trickery (for leading one to err is precisely what a defective understanding of something does).

As a third ground for rejecting the notion of a radically transcendent God, Linford claims that God's absolute unknowability deprives us of any criteria with which to determine what phenomena would count as evidence for God. In the case of a more anthromorphic God—a God who is good like us, compassionate like us, thinks like us, and so on—we could know what would count as evidence for his activity. If we know what God is like, we can know what God is likely to do. But if we cannot know God's nature, then we cannot know what God would do. We could not determine in advance what effects such a God would render in the world, and therefore, could not know what to look for.

I suspect some readers will find these arguments puzzling on their face. I do. Who, for instance, thinks that assertions of God's transcendence of space and time, the web of language, or finite minds is limited to mystical theology? In fact, the vast majority of the great theistic traditions—Christianity, Islam, Judaism, many forms of Hinduism, certain forms of Buddhism, the classical philosophical tradition, and so on—regard God as transcendent in precisely such a sense. The misapprehension that belief in God's transcendence is a recent concoction from university educated theologians who have modified theology to their more sophisticated tastes—or to escape atheistic criticism—is almost as widespread among popular atheistic tracts as the notion divine transcendence is in the theology, spirituality, scriptures, etc. of the great theistic traditions. Aside from the Mormons who doggedly visit my home with some regularity, I know of no religious believers who believe God to be a finite spirit within the imminent order—with the probable exception of the aberrant (historically speaking) tradition of theistic personalism.

But these are mere quibbles compared to the real difficulty with Linford's case. Linford confuses the claim that what God is, God's essence, cannot be captured in language with the claim that we can have no knowledge of God whatsoever. Put in a more technical vocabulary, Linford confuses the claim that earthly minds cannot have quidditive knowledge of God’s essence with the claim that we cannot have any knowledge of God.

Over and over again, one finds Linford supposing that God is powerful or loving "in no way that we can understand ...," or that God's goodness "cannot be understood by finite humans," or that if "the mystical theologians wish to say that God is truthful and trustworthy ... this would involve knowing things about God's goodness which the mystical theologian maintains we cannot know."

But do mystical theologians claim we know nothing about God? Let's consider the example of St. Thomas Aquinas, apparently regarded by Linford to be a mystic. Linford claims that Aquinas believes that "we do not even know what it means to say that God exists," and he cites the Summa Theologica Ia, Q. 3, prologue as proof. Such a claim seems, to anyone who has read Aquinas, misleading at best. The prologue Linford cites runs as follows:

"Once we have ascertained that a given thing exists, we then have to inquire into its mode of being in order to come to know its real definition (quid est). However, in the case of God we cannot know His real definition, but can know only what He is not; and so we are unable to examine God’s mode of being, but instead can examine only what His mode of being is not….
 
By excluding from God certain things that do not befit Him, e.g., composition, change, and other things of this sort, it is possible to show what His mode of being is not. So, first of all, we will inquire into His simplicity, by which composition is excluded from Him (question 3). And because among corporeal things the simple ones are imperfect and mere parts, we will inquire, second, into His perfection (questions 4-6); third, into His infinity (questions 7-8); fourth, into His immutability (questions 9-10); and fifth, into His oneness (question 11)." - ST Ia, Q. 3, Prologue (Freddoso Trans.)

The very text that Linford cites as evidence that "we do now even know what it means to say that God exists" says that we can know that God exists, is metaphysically simple, that he is perfect, infinite, and one. Linford's unwary readers might, I think, feel misled. If we know nothing of God, how can we know God is simple or infinite? The answer is that Aquinas hardly believes we don’t know what it means to say that God exists, but rather that we don’t comprehend God’s essence. This sort of misleading conflation is foundational for Linford's whole argument.

Such logic fails even for our finite, worldly knowledge. No one would say, for instance, that the pre-moderns knew nothing of water, even though they did understand its essence (H₂0). Nor did the ancients grasp the essence of the stars, but they could nevertheless predict celestial movements. One can lack knowledge of the essence of a thing while still observing many of its properties, characteristics, effects on other things, and so on. One can know quite a bit about something without grasping the thing’s essence.

Linford's mistake seems to come from a misunderstanding of how negation works in theology (mystical or otherwise). He recognizes the alternation between apophatic and cataphatic movements briefly, but misses what is going in the dialectic. One can affirm, for example, that God is like the sun (in that he brings life) but then deny that God does so as a physical entity without going back to square one. The original affirmation is preserved; its limitations denied. Or again, to affirm that God is good, but not good as we are (as relatively fragile beings that must achieve perfection or beatitude from a certain poverty in our being) does not simply negate the whole of the original sense of God's goodness. It only negates the limitations in the original sense, while preserving the affirmation.

Perhaps this sounds more of the soft strains of poetry than the more substantial power of reason. Linford's primary target is Karen Armstrong—though he cites others, such as David Hart and Denys Turner—and Armstrong does sometimes overemphasize the apophatic way beyond what is metaphysically reasonable. However, the central insight that God is radically transcendent, not an entity in the cosmic order, etc. is not only the cornerstone of mainstream theism, it is eminently susceptible rigorous metaphysical accounts. By way of illustration, let us reconsider Linford's three objections.

Response to Linford's First Objection

Linford's first objection to the transcendence of God is that it is unclear why "God has this sort of transcendence—the sort where we do not possess words adequate to describe God—and not some other." Yet, it is immediately and abundantly clear why God—at least the God of classical theism—transcends the sense of ordinary words. Our language is adapted for finite entities and their relations. When we say X is Y, for example, we generally mean that some determinate thing, a this or that, has some relatively determinate property. Gold is yellow, rather than some other color, for example. Ordinary language relies on the finite determinations of finite things.

We use language, then, in light of the finite mode of things, their properties, and their relations. But God—what the great theistic traditions mean by God—is not finite. God is qualitatively infinite, meaning that his nature is not limited or qualified in any way. If there is a God, then, it is perfectly clear why he would transcend our ordinary linguistic habits. The inadequacy of language to express God’s nature follows straightforwardly from the ontological difference between finite and infinite being.

Response to Linford's Second Objection

Linford's second objection is that the restrictions of our knowledge of God on which the "mystical" theologians insist render any revelation unreliable. If we cannot understand the truth or goodness of God, how can we appeal to God's goodness or truthfulness to secure the authority of revelation?

It will be immediately evident to the reader that this again trades on the conflation of essential knowledge, or perhaps perfect knowledge, and non-essential, or perhaps imperfect knowledge. I know hardly anything about tapirs, and certainly don't grasp precisely what it is that makes tapirs tapirs, but I know they do not read Shakespeare. Or, to use a previous example, though premoderns did not understand water in its essence, they knew that it existed and what it could be expected to do. Similarly, we may not fully grasp God's goodness, but this hardly means we have no idea of God's goodness or what it means for us. Our notion of God’s nature can be inadequate as essential knowledge, but nevertheless more than enough to ensure the trustworthiness of revelation.

Response to Linford's Third Objection

Linford's third objection is only slightly more substantial than the first two. Since we do not know what God is in himself, then—so Linford argues—we cannot know what would count as evidence for God. Were "made by God" coded into the genetic code of living beings, we could not use this as evidence of God's existence unless we know that God is the type of deity who would do such a thing. This objection, like the second, conflates essential knowledge with accidental knowledge, and fails for the same reason.

This objection is interesting, though, because it provides an opportunity to illustrate how we can reach negative knowledge of God. Consider the following argument from contingency by way of illustration. Note that by “contingent being,” I mean any being whose essence does not explain its existence.

Premise 1: only contingent beings have a restricted mode of being (i.e., have a determinate, finite, way of existing, e.g., existing this way rather than that).

Premise 2: the existence of a contingent reality caused by a set (whether finite or infinite) of solely contingent realities is inexplicable.

Premise 3: nothing exists inexplicably.

Premise 4: a non-contingent cause must be posited to explain the existence of a given contingent thing (from premises 2 and 3).

Premise 5: there can be no more than one being with an unrestricted mode of being.

Conclusion: there exists one infinite (i.e., unrestricted) being who causes contingent things to be (from 1 and 4).¹

I hasten to add that this argument won't be compelling without auxiliary arguments to establish the premises. The point is to see how an argument for God can move from the existence of contingent beings to the existence of a qualitatively infinite, absolute being (or beings) negatively. If the premises are defensible, one can infer the existence of an infinite being without ever needing to conceive of the essence of that infinite being, simply from the inadequacy of finite being's ability to account for itself. If it is evident from a consideration of the finitude of contingent beings that contingent beings cannot account for their existence, then we can infer (by negation) the need for non-finite, non-contingent being.

On such an account, it would be clear what would count as evidence (indeed, conclusive evidence) for God’s existence: the existence of any contingent, finite thing. And this without the need to grasp God's essence, except apophatically—understanding God as neither finite nor contingent. And, indeed, this "negative way" is often how God's simplicity, infinity, perfection, and absoluteness are traditionally established.

Conclusion

Well, I have complained that Linford has made elementary mistakes about divine transcendence. What constructive suggestions have to offer? How might the interested reader get a foothold in the notion of divine transcendence?

Gregory of Nyssa's Life of Moses is a wonderful account of the ascent to God by apophatic spiritual practices, using the story of Moses as a metaphor. Readers interested in learning about the more Eastern, mystical notion of God would do well to start there. For those seeking a more philosophical account, Pseudo-Dionysius' Divine Names and Mystical Theology are a wonderful statement of the metaphysics of the East.

Those who prefer a more familiar philosophical style could consult W. Norris Clarke's The One and the Many for a lucid argument for the existence of God using the Thomist categories of essence and existence. Clarke both argues for the existence of God and offers a clear account of the relation of the finite to the infinite. For the more advanced reader, Erich Przywara's recently translated Analogia Entis is perhaps the best statement of God's transcendence. Przywara employs Thomist categories but engages with philosophers such as Husserl and Heidegger, who still loom large on the philosophical scene.
 
 
(Image credit: Telegraph)

Dr. Peter Kreeft once noted that, “When Saint Thomas Aquinas wrote his great Summa Theologica, he could find only two objections to the existence of God, even though he tried to list at least three objections to every one of the thousands of theses he tried to prove in that great work” (Fundamentals of the Faith, p. 54). The two objections that Aquinas had in mind were the problem of evil and the apparent ability of science to explain everything without God. During my doctoral studies, however, I came across a series of articles devoted to disproving the existence of God by showing that the very attributes of divinity are incoherent and thus impossible to actualize in reality. These are known as “ontological disproofs.”

Ontological disproofs are logical arguments against of the existence of a thing based on what it would be if it existed. These arguments are very important because they do not simply purport to prove that God does not exist (like, say, the Easter Bunny or martians), but that God cannot exist (like a square circle or a married bachelor). This fairly recent form of argumentation has gained quite a bit of popularity among philosophical atheists (for example, Michael Martin devoted 23 out of 33 articles in his book The Impossibility of God to ontological disproofs).

The basic form of these arguments is something like this:

1. If God exists, he must be like ‘X’.  [Here ‘X’ = some attribute(s) of God, e.g.,  he must be good, loving, omnipotent, etc.].
2. ‘X’ is actually impossible.
3. Therefore, God cannot exist.

Unlike the simplistic rants of some atheists online, these are sophisticated arguments that demand equally sophisticated responses. The argument forms are valid, so to prove them unsound, some premises must be shown to be false.

Another reason for my interest is that I found myself unhappy with the responses I saw from some Christian apologists. It seemed like they were allowing the atheists to set up the rules in such a way that they could not lose, and the apologists were playing into their hands. Further, it seemed that some of the solutions proposed by these apologists would lead to theological heresy.

I saw the overall problem as one of how God can be spoken of correctly without creating these dilemmas that atheists use to argue against his existence. I believe the answer can be found in the classical "doctrine of analogy" – a way of understanding God-talk that does not open itself up (as easily, at least) to ontological disproofs, and also safeguards orthodox theology.

That pretty much summarizes my dissertation. For those interested in the topic, a more robust discussion follows.

Constructing Ontological Disproofs

Constructing a basic ontological disproof is fairly straightforward:

1. Assert that God must have some attribute (‘X’).
2. Define ‘X’.
3. Offer counterexample(s) to the possible actualization of ‘X’.
4. Redefine ‘X’ to avoid counterexample(s) problem.
5. Repeat 3 and 4 until procedure shows that ‘X’ cannot be actualized,.
6. Conclude that because ‘X’ cannot be actualized, God cannot exist.

Examples:

• Omnipotence: The ability to do all things must include the ability to not do certain things, for some abilities preclude others by definition.
• Omniscience: No other being can know propositions with certain indexicals (time, location, subject-object, etc).
• Supernaturality: There must be at least one fundamental law of creation which is not a result of God’s will – that of God’s will being effective. Therefore God’s ability to will is also natural.

A variant of this kind of argument is to pit two attributes against one another. If it can be demonstrated that God must have two attributes that cannot actually coexist, then God cannot exist:

1. Assert that God has attribute ‘X’.
2. Assert that God has attribute ‘Y’.
3. Show that A follows from ‘X’.
4. Show that ~A follows from ‘Y’.
5. Conclude that because “‘A’ and ‘~A'” cannot be true, God cannot exist.

Examples:

• Perfection vs. Creator: A perfect being needs nothing, but without a need God would not create.
• Creator vs. Immutability: An unchanging being cannot intend to create at one time and not at another.
• Immutability vs. Omniscience: An unchanging being cannot know changing truths (which differ from one time to another).
• Transcendence vs. Omnipresence: A transcendent being cannot be present as well.
• Justice vs. Mercy: No being can both give what one deserves and not give what one deserves.

Answering Ontological Disproofs

Modern Philosophy

Although many responses to these kinds of arguments have been offered, most follow the same basic method. Namely, they accept that the only way out of these logical conundrums is to change the definition of God’s attributes until they no longer suffer from the atheist’s attack, and proceed to do so until they have reached that goal. A good example of the procedure is found in two chapters of William Lane Craig and J. P. Moreland’s Philosophical Foundations for a Christian Worldview (reproduced online here). In his response to ontological disproofs, Craig makes two interesting remarks:

1. Scripture gives philosophers a wide latitude with regard to doctrinal formulations.
2. Anti-theistic critiques can be helpful in forming more adequate conceptions of doctrine.

He goes on to say that “two controls have tended to guide this inquiry into the divine nature: Scripture and Perfect Being theology.” He goes so far as to conclude that “Theists thus found that antitheistic critiques of certain conceptions of God could actually be quite helpful in framing a more adequate conception.” An example of this procedure is found in his response to problems of God’s immutability (changelessness):

"Rejection of radical immutability leaves it open for us to affirm nonetheless that God is immutable in the biblical sense of being constant and unchangeable in His character. Moreover, He is immutable in His existence (necessity, aseity, eternity) and His being omnipresent, omniscient, and omnipotent. These essential attributes are enough to safeguard God’s perfection without having Him frozen into immobility."

While “framing a more adequate conception” in this way might get around some skeptical arguments, this kind of thinking also leads Craig to deny that God is eternal (atemporal):

"A second powerful argument for divine temporality is based on God’s being all-knowing. In order to know the truth of propositions expressed by tensed sentences like “Christ is risen from the dead” God must exist temporally. For such knowledge locates the knower relative to the present."

Similar thinking has also led to Craig’s affirmation of admittedly heretical positions concerning God’s will and the the Son’s procession from the Father. By playing by the atheist’s rules, Craig and others may be handing over Christian orthodoxy in order to save (some kind of) theism.

I think the problem with most of these incoherency arguments, and many modern apologetic attempts to answer them by rethinking God’s attributes, is that they assume we can define God’s attributes in a univocal (same) manner according to how they are found in creatures. Definitions from finite reality are simply “blow them up” to the “size” of God (i.e., infinite).  For example, God’s omniscience (all-knowing) is defined as “knowing all true propositions” – as if making man’s limited knowledge of true propositions unlimited is all that is necessary to describe God’s knowledge. The problem is that such a method actually defines God’s attributes as “infinite finitude” or “limitless limits.” It is easy to see how contradictions will arise when an attribute of God is defined as an unlimited version of a necessarily limited concept. In theistic ontological disproofs, the atheist is simply noting this outcome.

Medieval Theology

Admittedly, in speaking about God, our language is using limited modes of expression. It must, for we have no other means of communication. All the words we have at our disposal are labels for things we experience in reality, and everything we experience is limited. It might seem, then, that univocal God-talk is all that is available to us if we are to say anything true about him. But because God transcends his limited creation, no concepts are attributable to God in the same way. On the other hand, we can say true things about God. So there must be a third way. That way is analogy.

In analogical talk, words mean similar things but are not univocal. We can speak of a  good knife and a good shoe, or say food is healthy and people are healthy. The common term in these pairs of statements have the same logical meaning, but do not pick out the same ontological reality. So while we can define “good” as “that which increases a thing’s ability to perform its function” we do not thereby say that “sharpness” in a good knife is the same as “comfort” in a shoe. Thus, in analogy we must know a thing’s nature – what it is, or what it is for – before we can know the meaning of the words used concerning it. For example, if I say, “My wife is a rose,” one immediately perceives that I am saying she is beautiful. Why not that she grows in the ground or photosynthesizes light? Because in knowing both the nature of a woman and the nature of a flower we can pick out logical similarities between them (e.g., beauty, softness, etc.) exclude dissimilarities (e.g., thorns, photosynthesis), then apply those logical similarities to the ontological realities.

With God we must continually purify our creaturely language of everything in it that is limited, image-bound, or imperfect. If we do not, we can easily be misled into thinking that the thoughts in our minds evoked by the terms we use in God-talk are really what they are in God. As Michael Martin himself notes:

“ordinary men tend to understand God in ways that are familiar to them despite the protests of theologians and intellectual ministers. As a result, God tends to be conceived of in the image of a man – a man much more powerful, moral, knowledgeable, and so on than ordinary men.”

Craig seems to agree with this assessment when he bemoans the problem of analogy:

"While we can say what God is not like, we cannot say what He is like, except in an  analogical sense–which must in the end fail, since there is no univocal element in the predicates we assign to God–, leaving us in a state of genuine agnosticism about the nature of God. Indeed, on this view God really has no nature; He is simply the inconceivable act of being. Why should we adopt so extraordinary a doctrine?"

I would say the reason we should accept such an “extraordinary doctrine” is that God is truly extra-ordinary! If God transcends all we know from finite reality, then we do not know his infinite nature directly. If we do not know God’s nature directly, then our words (which communicate only the finite reality in our minds) will fail to communicate the ontological essence of God – even when they are true statements. But we can know the truth of propositions about God without knowing God’s essence. For example we can know much about a person by looking at a painting of her, but paintings do not give us the true essence of a thing (one is paint on canvas, another is a human being). Another example: a child can know what it is to love pizza, love his toys, and love his parents – but when he considers that his parents love each other he does not really know what that love is. However, he can make true statements about their love and recognize their kind of love when he grows older and experiences it himself.

No creaturely words are used univocally (i.e., with the same meaning) of God and us; nor are they simply used equivocally (i.e., with two completely different meanings). Rather, God talk is analogical (i.e., in a way partly the same and partly different). So we must not confuse what we are thinking when we say something of God with what is literally true of him. Our talk of God is analogical – partly alike (univocal) and partly unlike (equivocal). We can thus speak truthfully of God in several ways:

• Negation of the Limited (e.g., infinite, eternal, aseity, impassibility).
• Super-Eminent Affirmation of the Unlimited (e.g., omnipotent, omnipresent, omniscient).*
• Metaphor (e.g., God’s “arm,” God’s “repentance,” “fire,” “rock”).

To say this leads to “genuine agnosticism about the nature of God” or the view that “God really has no nature” is to confuse what we can know about God with knowing God. Knowing God is not simply affirming true propositions about him – it is having his very essence in our minds. As finite creatures, we cannot know infinite reality in this way. But we can know true statements about God. Univocity is not found in the ontological referents of words, but in their abstract definitions. Knowing that good is “comfortable” when said of a shoe (because a shoe’s purpose is comfort) and that good is “sharp” when said of a knife (because a knife’s purpose is to cut) does not help me know what good is in anything else unless I know what that thing is. But I can abstract from its usage that “good” means “attains to its purpose.” Thus I can affirm statements about another thing’s goodness when I know what it is. Since I do not fully know the ontological reality being picked out by the word when it is used of God, I may not know what it is (= have it existing in my mind) – but I can affirm that the statement “God is good” is true.

Conclusion

Good theology begins with proper metaphysics – not by simply denying finitude to finite concepts. This is why theological definitions based on classical metaphysics of God’s perfections often sound so obtuse:

If the doctrine of analogy is correct, the linguistic precision with which these atheists (and their apologist correlatives) approach ontological disproof arguments is simply not available for use. Finite univocal concepts are necessarily incoherent when applied to an infinite being, and so their use will naturally result in contradictions. While perhaps too unwieldy for contemporary analytic philosophers, these more precise definitions do not open themselves to the same kinds of critiques when understood analogically.
 
 
(Image credit: Graphicality)