A (Not So) Simple Explanation of the Ontological Argument
In recent years, I’ve probably had a dozen or so encounters with the ontological argument. It pops up in Christian circles both casually and as a topic of study. It is, however, notoriously slippery. The argument tends to come and go without any trace. It promises the curious mind a compelling argument but leaves only confusion in its wake. Much of this is due to the history of the argument. It is typically accredited to Anselm of Canterbury, who lived in the 11th century AD. There are similar but distinct arguments made by various other individuals, including some from the Muslim world, which should probably also be considered in a more thorough look at the argument’s history. Suffice to say, the absence of the context in which the argument was penned is enough to cause us modern readers to often miss its force.
There is also a philosophical perspective on the argument; that is, the argument has been given a more formal, exact treatment. The philosophical treatment of the topic only serves to magnify the contextual distance; even determining an “exact” formulation of Anselm’s argument is dubious. Furthermore, many have endeavored to take inspiration from only Anselm’s argument by formulating their own version of the ontological argument. These actually may be more compelling in today’s philosophical or theological context.
While there are logically sound arguments under the “ontological” title, it is critical to unwrap the layers of important but challenging contexts, technicalities, and categories around them in order to truly grasp these arguments in a way that is intellectually satisfying. One of way doing this is, perhaps, simply to start from scratch in order to explain the topic at hand. Nevertheless, I will at least give some indication of Anselm’s argument as I apply my own reasoning.
Anselm’s context was heavily influenced by the notion of transcendentals. These formulations of ideals were rather popular in medieval theology. They conveyed some elements of ancient Greek philosophy as they are, in some ways, a combination of Platonic forms and Aristotelian virtue ethics. Indeed, the transcendentals were the loftiest topics man could consider. It is often said there are three or perhaps four transcendentals: Unity, the Good, the True, and the Beautiful. God may possess these attributes, but more specifically, God has condescended to His creation such that these attributes of Himself are revealed to us. These transcendentals are probably not as distinct as they are channels of revelation of God’s simple, singular Being.
The first step of Anselm’s argument does not leverage transcendentals directly, but rather, asserts something inherent to a transcendent God: He is that than which nothing greater can be conceived. Translating this into modern parlance may require swapping the subject and object: there is nothing conceivable that is greater than God. We will call this premise 1. This statement begins a series of premises which, when taken together as true, is intended to imply a conclusion.
Premise 1 challenges us in various ways. Does this imply that God is conceivable, and if so, is that really the case? If we have a belief about God that is inferior to a god that we might imagine, does that make this premise false for us? A medieval theologian would probably have no trouble answering these questions. Of course God is conceivable insofar as what he has revealed to us, such as the transcendentals, are ideas that the mind can grasp. This requires we accept the category of truth and our ability to grasp it. If our mind believes that God is something other than perfection, other than the source of the true ideals known as the transcendentals, we have been deceived and our judgement is false. Indeed, the whole point of this premise is that if we can conceive of something greater than our idea of God, then our idea of God was probably wrong in the first place, for there is nothing conceivable that is greater than God. The medieval theologian simply took this fact as definitional. So, at least under this brief scrutiny, premise 1 really only requires us to acquiesce to our own ability to “conceive of” God.
For these reasons, premise 1 is probably a bridge too far for postmodern thinkers, and indeed many others. Perhaps helpfully, many modern philosopher-theologians make a similar claim: God possesses all of the great-making properties. I recommend if posing the ontological argument in contexts where premise 1 is a challenge, to make an updated argument starting from the premise that “God possesses all of the great-making properties” instead, but that’s a topic for another time.
Premise 2 is hardly controversial: Some people say that God is not real. Yes, that’s it. The kicker here is that these individuals must have some conception of God, thereby imagining God in some way, even if as sub-God. Quite tricky.
Premise 3 builds on this, claiming that some things exist as ideas only, while other things have “real” existence. We might ask what “real existence” means. The temptation follows to associate reality with physicality—but is God physical? No, that’s not it. Perhaps we might say that real things have the attribute “being”, whereas ideas do not. Then, what is “being”? This is the ontological question, into which we have little capacity to peer. Being “is,” whereas non-being “is not”. It seems as if we cannot conceive of non-being, a perennial topic within philosophy. Nevertheless, the categories under consideration at the present are not being and non-being, but existence as being and existence as an idea.
Premise 4 makes the claim that “real” existence is better than mere existence as an idea. Is this true? It seems intuitive that things that might exist are greater than their hypothetical counterparts. However, from a platonic perspective, are ideas not “higher” than real things? This judgement of “better’ is indeed questionable.
The conclusion, then, is that those who imagine God to exist as only an idea should instead admit to God’s real existence. This is because if they conceive of God, who is definitionally the greatest conceivable thing, and real existence is greater than nonreal existence, then they should logically attribute real existence to God. In this way, the argument was pointed squarely at atheists who believe that God is a mere idea, but who also acknowledge the idea of God to be “that which is the greatest conceivable thing” or similar.
There are two historic counterarguments to this. The first came very quickly from one of Anselm’s peers, Gaunilo of Marmoutiers. The second would come from Kant. Gaunilo objected that perhaps we might make a similar argument about other things, such as islands. Perhaps we might imagine a perfect island. It has all of the islandly perfections: clean sand, tropical fruit, clear water, optimal flora and fauna, warm sunshine, and the incarnation of those strange tiki-dancer bobble-heads only appropriate in hippy vans. Too bad it doesn’t exist; it’s only in your imagination. But, then, it’s not perfect, for existence is also part of islandly perfection! Thus, the island must also exist, for otherwise it would not be perfect, and you couldn’t imagine it. So, it doesn’t exist, but it must exist! A fallacy which must prove the futility of the argument.
Anselm made his reply without much difficulty and said that it should be published as an appendix to his original argument. He explained that his argument works for only necessary beings. Things such as islands are contingent, rather than necessary. While these are philosophical terms that continue to be important and nuanced, Anselm’s use of them was rather straightforward. God’s existence is “necessary,” while islands might not exist. If we take this reply at face value, Anselm is either adding a premise, or noting another aspect of God’s perfection. If it is the former, then the argument falls flat, dramatically, as we then have to believe in the necessity of God as a premise. This would be hardly different than accepting the conclusion prima facia. If it is the latter, it probably doesn’t impact the veracity of the argument, since we already accepted God’s (theoretical) greatness in the first premise. So which is it? My inclination is to give Anselm the benefit of the doubt, as he surely deserves. However, this is less of a question of Anselm’s intention, and more of a question of philosophy: is necessity an attribute (of God, the greatest imaginable thing)? We at least know that the orthodox answer is a resounding “yes”. This is very much so part of what we accept definitionally about God. It is also firmly acknowledged by the philosophical theological tradition.
But there is, nevertheless, something strange about Anselm’s claim that God’s existence is necessary in the context of the ontological argument. As Anselm implies in his response to Gaunilo, the argument does not work without acknowledging God as a necessary being. While we might imagine that this is lumped into premise 1, and thereby does not impact the veracity of the argument, clearly it is not an acceptable premise to the target audience. If the atheists believe that God’s existence is necessary, then it would be illogical for them to believe that God only exists as an idea. In some sense, this was what Anselm was trying to show in the first place, but if so, it comes across as a straw-man of the atheist position. Thus, I find Anselm’s argument to only be acceptable to atheists who admit the necessity of God’s existence — a nonexistent audience. We might try to shift God’s necessity to be an additional conclusion, but this would not hold valid against Gaunilo’s objection.
This is not the end of my concerns with Anselm’s ontological argument. Kant’s objection is also intriguing. Kant claimed that being is not a predicate, for it does not contribute anything to the subject. This is a denial of premises 3 and 4, although not directly. Kant’s objection seems silly to some at first glance, although it has convinced many individuals that Anselm’s argument is invalid. How can we accept that there is nothing predicated about a subject by specifying its existence? Is it the case that saying “God exists” is no different than saying “God”? As Anselm claims, existence takes at least two forms: that of an idea, and that of “being”. Perhaps Kant decided to indirectly address Anselm by making a philosophical assertion on linguistic grounds, rather than trying to counter Anselm on a logical basis. I have heard that the German of Kant’s day accepted sentences having only a subject as valid, just as many languages require us to supply many verbs of being/doing. Indeed, Kant deserves a charitable interpretation at least as much as Anselm does.
What if Kant’s objection is subtly rooted in his views on ontology? I am not sure about this, as I find no obvious Kantian beliefs that help us explain his critique here. However, I can supply some of my own reasoning along similar lines. I believe that Anselm made a category error in claiming that things may exist as ideas. We might understand what it means for things to exist, in that they “have real being”, but for something to exist purely as an idea is enigmatic. Are these ideas thoughts in our heads as Anselm’s argument suggests? Do they have a different sort of existence, as Plato believed? Do they exist in some other way? I find the answer to each of these hypothesis to be “no”. Philosophers disagree on this point, of course. Nevertheless, I find “existence” of ideas to be better described by specifying their “non-existence”, for they have neither place nor being to call their own. Point being, Anselm’s claims that there are two types of existence and that one is greater than the other (premises 3 and 4) are incoherent in my understanding of ideas and being. There are not two different kinds of existence; rather, there is existence and non-existence. Non-existence needs not be mysterious, as it can just be “non-physical,” although this too requires further explanation.
It may seem like only shreds of Anselm’s ontological argument remain. Nevertheless, his work has inspired centuries of thinkers to ponder and marvel at this argument. Many have found it compelling, while many others have found it unconvincing, whether or not they can critique it. The ontological argument lives on in various other forms beyond Anselm’s formulation, and there are even some that are more difficult for me to find flaws in. Ontological arguments are, in fact, not limited to something like Anselm’s formulation, but can be any argument for God’s existence using ontological claims.
When taken as a whole, including his reply to Gaunilo, Anselm’s argument really contributes to the conversation about the necessity of God’s existence more than it works to convince atheists of God’s existence. This conversation has led to “arguments from necessity”, which are often attributed to Thomas Aquinas who lived several generations after Anselm; however, I find this sort of argument to have the longest history of any of the philosophical arguments for God’s existence, going back at least to Socrates himself. Perhaps we can analyze arguments from necessity in a future post.
Interestingly, Thomas Aquinas didn’t like Anselm’s ontological argument, claiming that it made God’s existence self-evident; Thomas believed that God’s existence was not self-evident. He preferred a posteriori arguments (those based on evidence or revelation), as is evident in his arguments for God’s existence called the five ways. Anselm’s a priori reasoning misses the mark for many Thomists (those who ascribe to Thomas’s theology/philosophy) in the same way as it would for modern empiricists and postmodern thinkers alike as it elevates the human rationality to the heavens without giving our epistemic state its proper due. Successful ontological arguments would need to not make the same mistakes.