Original Argument: The Ontological Argument Against God

Maz’s Critique: A Reply to C.B. Robertson

Overview

First, I want to thank Maz for his thoughtful and thorough response. I was recently reading Martin Luther’s Bondage of the Will, and it occurred to me that some of the best writings in history come out of these sorts of back-and-forth in writing. But I especially want to thank Maz for allowing me to clarify a few points that, thanks to his response, I now see were not entirely clear in my initial argument 一 or, at least, were not as clear as I had intended them to be.

Broadly speaking, Maz is treating my argument as an argument against God’s nature, based upon an assumption of God’s nature from classical theology. This is an entirely understandable misreading, both for my invocation of how God is described by classical theologians, and for calling it an “ontological argument” against God, and what I had hoped to clarify by the substance of my argument has remained obscured by those words. So allow me to clarify here that my Ontological argument against God 一 though it certainly does concern God’s nature 一 does not begin with God’s nature, but rather with the language that makes even the conceptualization of such things as the attributes ascribed to God possible, as I attempted to convey in my original paper:

Similarly, revelation of the sort received by Moses and Mohammed is necessarily interpreted in light of a concept of a universal God of foundation that precedes the revelation itself. In other words, our knowledge of a hidden God is, itself, transcendentally contingent upon an idea of God. This paper is concerned with the origins of this idea. 

My argument gets to “ontology” by way of ascribing these very attributes to language itself, and then makes the claim that God’s nature 一 as described by theologians 一 is therefore linguistic, and not something external to the human mind. But the beginning of this argument is not exactly God’s nature, as something assumed a priori to precede our descriptions of it, but as it is described.

Divine Simplicity

A byproduct of this way of thinking is that God is not the only target of this linguistic approach to metaphysics (which can be understood as a very constrained form of nominalism), which leads to Maz’s first critique:

The most significant issue is Robertson’s misunderstanding of the divine simplicity, among other aspects, of the God of classical theism.

Now in my original paper, I had said:

He is described as not one good thing among many, but goodness itself. Thomas Aquinas said God is ipsum esse subsistens — the persistence of being itself. And perhaps most importantly, he is one.

Before alleging that I have misunderstood Divine Simplicity, one should ask: what is the difference between “simplicity” and “being one”?

Classical divine simplicity argues that God is free of any kind of composition, that his attributes are indistinct and singular in kind and essence. My argument that he is “one” was not a numerical claim, contra polytheism (although that is also there), but I specifically clarified that:

The unity of God in his nature demonstrates a unity of God’s ontological essence 

In other words, I was speaking of the oneness of his being and nature — the exact, classical meaning of divine simplicity, if I understand the concept correctly.

When Maz says that:

Robertson treats “goodness,” “knowledge,” “being,” “truth,” etc. as though they were independently constituted properties or categories and asserts an entity called God can stand in relation to all of them,

he is both missing the inclusion of divine simplicity itself over a slight difference in language (“one” instead of “simplicity”), and misunderstanding general the direction of the argument. I am not, for example, trying to describe motion and stillness as properties God holds, thereby revealing a contradiction in God’s nature, but am offering motion and stillness as examples of linguistic descriptions of the sort which might lead to transcendental arguments demanding a grounding for such descriptions, and where such descriptions must inevitably lead. The hypothetical nature of this explanation was intended with “Suppose we were to make our own version of Zeno’s paradox…”; “such an entity we would call God” is the final conclusion of our hypothetical syllogism, but I am not (knowingly) quoting any actual Christian theologians, beyond a cheeky invocation of Aristotle/Aquinas’ “first mover”. I am not treating God as an individual who possesses properties, but as a claimed foundation for these properties, and as an idea arrived at by way of ideas. It is not the idea of God himself, but the ideas by which we arrive at God that I am first concerned with. 

For example, let us look at this concept of divine simplicity.

The idea of simplicity predates Christianity, and goes all the way back to Plato, who says:

Do you suppose the god is a wizard, able treacherously to reveal himself at different times in different ideas, at one time actually himself changing and passing from his own form into many shapes, at another time deceiving us and making us think such things about him? Or is he simple and does he least of all things depart from his own idea? (Republic, 380d)

But here, simplicity is not something observed, but something put forward conceptually, as an answer to a hypothetical question about what God must be in order to be really perfect (which is to say, really God). Absolute simplicity itself — as an answer to a question concerning hypothetical perfection in a hypothetical God — is itself hypothetical. We see simple things, as opposed to complex things, but the absolute oneness of essence is not something we see in any real thing in the world. This issue is further compounded by the idea of the trinity, where a supposed singularity of essence despite separation of personhood is thought to be a profound answer, when it is really just a word invented to create a distinction when there is none: to be separate people is to have separate essences, to some degree. To be three is to be a composite.

Here, simply accepting certain ideas as mysteries beyond human understanding is a perfectly valid route, but such an acceptance would destroy any right of the Christian to challenge the foundations of other people’s worldviews or beliefs, with charges of “self-defeating” etc.

In either case, my position remains that both the idea of God and the idea of divine simplicity which supposedly describes him find their roots in language, and that this language describes things which do not and cannot exist, and that there is no other conceptualization of God, and finally, from this, we can know both the essential nature of God as originating in language (John 1:1), and from this nature that God, as a pure idea, like absolute simplicity, does not exist, based on the nature of the language by which we came to conceptualize him.

Grounding-Exemplification Error

As with the argument regarding divine simplicity, Maz is slightly misunderstanding the argument for how we arrived at an idea of God with a positive, metaphysical claim about God’s nature in relation to these conceptual “attributes.” It’s not that “God exists, and he has goodness, truth, justice, etc,” but rather, “we think we see goodness, truth, justice, etc, therefore God,” and I am challenging the move from an experienced instance of a quality to the generalization of that quality into a kind of entity. I am not saying “they say God is the ground for goodness, therefore God must be good,” but observing that theologians and philosophers have attempted to derive God’s nature from what he grounds, deducing the very concept of God from the supposed singularity of goodness. God’s exemplification of the attributes which believers describe him by is not my claim, but the very mechanism by which a single, ultimate, good God is conceivable.

This was what I was alluding to when I first accounted for and attempted to address this concern in the original piece:

And if this foundational deity is not like the things for which he is a foundation, then we ourselves would have no ground for claiming any knowledge of his nature whatsoever.

If Maz is correct and classical theism rejects this (which is not my impression — if Aquinas’ ipsum esse is permissible as evidence), then it must assert some other explanation for how we came to believe in one God. If this alternative explanation is revelation, then we must consider the epistemology and psychology of revelation, and compare the revelations of Christianity to those of the variety of other religions around the world.

I know of no explanation for the origins of the idea of God which is not either based in reason or in revelation.

3. Epistemic Leap

I believe Maz’s misunderstanding of the argument may be at the root of his claim of an unjustified epistemic leap. If he assumes my argument is true, then it does imply that God cannot exist — if the very idea of God arises from misapplications of language, such that it becomes clear to us that his very nature is a misapplication of language, then we can say not just probabilistically, but with certainty, that God does not exist, just as we can say that a “round square” does not exist with certainty once we are enlightened to the experiential meaning of “round” and “square”. Seeing the origins of the language from experience reveals the impossibility of their combination, and simultaneously shows us the origins of the phrase “round square” in language.

The law that Maz is invoking — roughly, “a failure to prove P doesn’t mean not-P” — is a perfectly valid law, but there are other kinds of arguments against P which do not purport to disprove an argument for P, but to disprove P itself. My argument is one such argument.

I am not entirely sure what “predicated upon human epistemic standing” means, but if it means that my argument hinges upon humans being able to know things, infer things, and make arguments, then I would grant this as trivially true. But the ontological argument against God is not making an argument against certain evidentiary claims for God; it is beginning with the foundations for the idea of God, and pointing out that the very idea does not make sense. Whether other great powers, creators, or simulation coders exist is unknown, but the God described by monotheism does not begin with idle speculation regarding the origins of things, but begins through logical deduction. It is in this original logic, and not my observation about the language used within this logic, that an epistemic leap is to be found.

4. Phenomenological-Real Leap

Maz understands my argument to be mistaking the phenomenological constructs by which we know things for the things themselves, as if seeing the process of experiencing and expressing the external world disproved the external world, simply because the map was not the terrain.

My argument is not — and nothing like —  “experiences can’t be real if they’re in your head.” I am not saying that the concept of God is phenomenologically constructed, but linguistically constructed, which is an immensely important difference, and the confusion conveys a fundamental misunderstanding of the nature of my argument.

My claim follows a genealogical order which goes as follows:

1. We have a phenomenal experience (a red apple is delicious)
2. The phenomenal experience is expressed (“that apple was good”)
3. The expression is isolated and solidified into a standalone concept (“What is ‘good’ anyways?” — this is sometimes called “reification”)
4. The concept, being treated as a thing, is now thought to require an external foundation
5. This foundation is asserted to be God

Both the steps from 2 to 3 and from 3 to 4 are experiences we can recognize firsthand, with 4 to 5 simply being a continuation of the same kind of linguistic error, compounded. It is a process of transforming internal, subjective experiences into external conceptualizations by language.

Now Maz invokes the examples of Mt. Fuji and electrons as external things which we refer to, but can only know by internal phenomenal experience (i.e., ‘our experience of Mt. Fuji’), and compares this to our relationship with God. But this conflates the conceptual essence of a thing with our knowledge of it. Our phenomenal experience of Mt. Fuji is not the result of a word, but from visual data. To compare Mt. Fuji with God would be like beginning with descriptions of Mt. Fuji — “large,” “mountainous,” “snowy,” etc — and seeking Mt Fuji from these attributes, and then finding every instance of that descriptor as a sign of Mt. Fuji (making the Himalaya Mountains more like Mt. Fuji than Mt. Fuji). Though we still know external objects by phenomenal, internal experiences, we nevertheless find our foundation for our belief in them in pre-linguistic experience. God is not like this.

We can, in theory, go on ad infinitum in parsing which parts of knowledge refer to outside phenomena, and which to internal, subjective interpretation; the two not only interact, but intermingle. But at a very basic level, the distinction between an experience seeking description and a description seeking a distillation transcends the interweaving of external stimulus and internal interpretation. The idea of Mt. Fuji is the former; the idea of God is the latter (we can discuss whether electrons belong somewhere in the middle, if the conversation meanders in that direction).

I am not saying that the order of knowing must be like the order of being, as I acknowledged in my original examples from the periodic table (see gallium and scandium, Sec 1, par 2). Rather, I am saying that the things theists claim to know about God are incompatible with external existence. The order of knowing is important for isolating these essential natures and understanding what it is that we are talking about, and not confusing ourselves, but the argument concerns God’s nature, and the nature of this nature. It does not directly concern the order of knowledge, nor does it treat God as a phenomenal entity, but rather begins with the phenomenology of language.

5. Unsupported internal-external taxonomy

Maz argues that I make an unsupported distinction between internal and external groundings of objects.

Merely dividing concepts into two classes does not demonstrate that the division tracks a metaphysically significant distinction. Not only is the justification unstated, but there appears to be ambiguity between the two categories. 

To this, my own understanding of language forces me to concede… at least where we are speaking technically. This might mirror Kant’s concession when he ceded the distinction between phenomena and numena to Hume, but whereas Kant was seeking to salvage the possibility of knowledge in the absolute, here I am simply acquiescing to the impossibility of knowledge of the absolute. At the same time, I would question the similarity between that sort of “absolute” knowledge, and the kind of knowledge by which we navigate every single other activity in human life, beyond the mere word “knowledge”.

I had intended to invoke a distinction between external and internal groundings as a way of illustrating the difference between, on the one hand, objects which we perceive and then attempt to describe, and on the other, objects whose existence we infer based entirely on the description of other things. The former I referred to as “externally grounded,” the latter “internally grounded.”  Maz has, I think, correctly pointed out an inadequacy of my language in properly creating a metaphysical distinction between kinds, since at some level, all knowledge is phenomenal (as his examples of Mt. Fuji and the electron also illustrate). But this is a failure in language more broadly, which illustrates my larger argument about the difficulty of doing metaphysics with language. The “independent establishment” Maz demands cannot be provided by any argument, both because the referents we refer to are inescapably subjective in knowledge, and the medium by which we attempt to describe them — or make sense of them, even to ourselves — functions entirely by metaphor and association and is subjective in mechanism. Maz’s point about an abstract understanding of someone else’s pain, thus blurring the line between the “internal” and “external” of pain, illustrates this mercurial and changing nature of language, since someone else’s pain, in concept, is essentially something different then our own pain, just as a dull ache can be understood to be different in kind from a sharp sting, yet we refer to it all — in idea and in experience — by the singular word “pain,” and somehow we still manage to understand each other (sometimes).

Perhaps my argument fails to establish a hard, metaphysically solid taxonomical distinction. But if so, then this is a symptom of failures in all metaphysical arguments to deduce grand and absolute truths about the universe from language alone. Such a failure should indicate the difficulties in drawing grand metaphysical conclusions from abstract, linguistic categories in general, and God is, incidentally, the greatest of these grand metaphysical conclusions. I hope even in failure, the distinction between externally-grounded and internally-grounded ideas — at least as a directive gesture — may still help us understand a meaningful difference in being between things which we experience and then attempt to describe, and things which we only come to think of because of descriptions, which was the intention of the taxonomy to begin with. Illustrating the limitations of language and of the analytical mode of philosophy was not the intention, but an apparent and useful byproduct.

6. False Dilemma, analogical predication 

Maz argues that I present a false-dilemma, in that God either resembles the things he grounds or he does not, but argues that classical theism instead says that these resemblances are analogical, without necessarily having to precisely resemble that for which he is contingent, nor being completely unknowable in lieu of precise resemblance.

But if our descriptions of God are analogies — that “it is like he was goodness itself…” — then we are still left with a problem in explaining how we came to know of this being in the first place, since analogies speak of object A in terms of object B. Moreover, concepts of extremity like “absoluteness” and “perfection” are, linguistically, different in kind than what can be conveyed by analogy, since analogies indicate directionally, while the characteristics of God by which he is described are absolute and deductive.

My argument presumes that the theist sees our perception of good, order, and design in the world as a kind of direct window into the effects of God’s nature, and we can know him — indirectly — by these effects. This is not an argument for precise resemblance, but of some degree of knowability by effect, and broad (perhaps analogous) resemblance in nature to that which he creates. But when these effects begin with broad descriptors of internal states — like beauty, perfection, goodness, etc — then we are outside the realm of pure analogy, and into something closer to psychological projection. And if analogies — which are here advanced as the totality of our description of God’s nature — are allowed in articulation of the metaphysics of the universe, then assertions of God are brought down to the level of an idle dreamer’s speculation, no different then those who say that maybe the world is like a great wheel (after the invention of the wheel), or like a book (after the invention of books), or that perhaps it’s like a simulation (after spending too much time on a video game). These too are analogical explanations, with all the weight that an analogy has to offer. Analogies can illustrate and help people conceptualize, but they do not argue or demonstrate. If God’s nature is only gestured to analogically, then — again — we must admit that we don’t know God’s nature, or his existence (and the God which countless generations have spoken of by his nature remains the target of my inquiry). 

7. Self-undermining argument

Maz argues that my argument is self-undermining because it falls into its own ungrounded taxonomy and must either be objective independent of human experience (how do we reach it?) or internally grounded (and therefore with no external referent, or at the very least, it has no standing to make objective claims).

I readily and happily concede that some truth (in fact, all truth) is not internally-grounded, insofar as we are talking about “truth” as “that which is.” But knowledge is a different matter than truth; knowledge is never the same as truth itself, any more than the map can “be” the terrain, but is useful and good as knowledge insofar as it corresponds with what is true, and this we gauge by its predictive value. That knowledge is always interpretive (and therefore, somewhat internal) does not mean that the truth about which it speaks is internal, as was discussed before. Nor should we be confused by the different senses of “objective” which might be invoked, one referring to things which are, so to speak, ‘absolute,’ and the other sense referring to things which are merely outside of ourselves.

My claim that God cannot exist is externally grounded in the second sense, in that it refers (in the negative) to a concept stated to be outside of ourselves. But at the same time, I observe that those who claim to articulate God’s nature (and to infer his external existence from his nature) invariably describe a nature that can only be derived from a certain expansion of language, which itself finds its origins inside human subjective experience (perfection, justice, goodness, consciousness, etc). This confusion of internal and external grounding, pertaining to my argument itself, is resolved when we agree with Aquinas that we can know negatives with greater certainty than positives, and that we can even know God’s nature via negativa.

But when we understand that God’s nature is not merely grasped, but arrived at initially through internally grounded experiences, we should begin to see. These internally-grounded experiences become conceptualizations when they are articulated, and those articulations lead to distilled abstractions and ultimately, into identities of their own — not a fast bird, but speed itself; not a beautiful painting, but beauty itself. Whereas other entities are inevitably known by way of internal experience, God is asserted from the language of internal experience itself, and from this language he is asserted as something that is not just external, but the very grounds for externality itself. In this linguistic genealogy of the idea of God, we see an error that gives us as much ground for certainty in the negative as subjective knowledge permits, and this perception is not similar in kind to the positive claim of a single, unifying, transcendental God. When we remember this distinction, and when we remember to distinguish truth from knowledge, as it pertains to objectivity, the supposed self-refutation evaporates.

Conclusion

I want to thank Maz again for his thoughtful response, and his helpfulness in clarifying these issues. A secondary unfortunate byproduct of the nature of language is that there is no clear and perfect way to convey a single message to everybody, for all time, so much clarification is always needed in conveying the intended meaning of an idea. But I hope the responses to his major criticisms has provided at least some of that clarification.

His minor rebuttals, being minor, I will leave unaddressed for now, though we may get to those later if they are invoked in larger ways. Suffice to say, all of them are addressable — the first of these (an unsupported assumption of non-theism) being addressable by simply asserting that I am not an atheist, and do not believe that there are no Gods, but that in fact there are many. It is only one god — or rather, one particular kind of asserted God — which is at issue. There is room for many more, of many other types, beyond just this somewhat single-minded view of what a “God” can be.

Overall, the point, once understood, I think remains intact. The argument that by his nature, we can know God does not exist really begins with the nature of language and the relationship between this nature of language and the nature of God. Knowing that language comes from us, and that the ideas by which we conceive God are linguistic in origin (“justice” as ascribed to God is not like the instances that we call “just” in our experience), we can see the emergence of God as a third-order expression of reification, from descriptive naming, to abstraction, to ascription. These compounding corruptions of language do not disprove some presupposed outside entity, but conclusively demonstrate the linguistic origins of the very idea of a singular, total, absolute God, and in showing this, show his internality to our own consciousness — in other words, his negation.