Is there a convincing form of Is there a convincing form of natural theology?

Natural theology is that branch of theology that is concerned with the proofs for the existence of a deity, in western tradition, generally the Christian style God. It takes three principle forms. Firstly, there is the Ontological Argument, which asserts that it an inherent and necessary property of God that he(?) exists. Secondly, there is the Cosmological Argument, which asserts that there necessarily exists a self-causing entity, and that anything which can cause itself fulfils our criterion as God. Thirdly, there is the Teleological Argument, which asserts that the universe can only have turned out as ordered as it is if it were created by an intelligent being, who would be equivalent to God. Once I have examined all these arguments, I must then decide whether or not they are in fact convincing, or even coherent. They do all seem to be mostly coherent, but I am unsure that they could be termed convincing. They only serve as supporting arguments for people who are already sure of their outcome. They seem to all hinge upon points where subjective views make all the difference.

To begin with I shall examine the Ontological argument. This can be succinctly summarised in four steps:

  1. God is a perfect being
  2. God has every perfection
  3. Existence is a perfection
  4. God has existence

As with every logical argument, there are two ways of attacking this, either by disputing the truth of the premises, or by disputing the validity of the structure. The structure for this formulation is valid assuming the premises (points 1 and 3), so assuming acceptance of them, the argument is coherent.

I shall begin by testing the assertion of point 1, that God is a perfect being. Our evidence for this is twofold, firstly that of the bible, which cannot really be admitted as evidence, since it is itself in question. Our second source of this conception that God is a perfect being is that the very concept of God requires perfection, so one cannot talk of God as not necessarily being perfect. If someone was to attempt to talk of a God that was not perfect, they would be misunderstanding the concept. However, there is a problem with this point. It seems that point 1 may in fact be too strong an assertion. All we can really assert is that our idea of God is as a perfect being. The rest of the argument can still hold from this premise, but not if we consider it further. If we consider the ambiguity of such a sentence, we discover that it can in fact mean two different things. It could mean 'There is a perfect being God of whom we have an idea', or it could mean 'We have an idea of a perfect being God'. Whilst the first possibility maintains the coherence of the argument, it also presupposes the conclusion, i.e. that God has existence. The second possibility leads us to a weaker conclusion about the properties of our theoretical God, but cannot make any claims about whether there is in fact a referent for the idea.

Point 2 is in many ways implicit in point 1, but still merits explicit stating. In his third Meditation, Descartes considers the possibility that God has only some perfections. However, he makes the point that completeness is in fact a perfection, and therefore it is necessary for a truly perfect being to possess every perfection.

Point 3 is where the real contention takes place. The first assertion that existence is a perfection came with Anselm when he first proposed the ontological argument. His argument is based not necessarily around God, but rather around ä being greater than which nothing can be thought". He points out that if that is our criteria for God, then we must necessarily accept existence as one of the properties, as a being as great as we can imagine that doesn't exist is clearly less perfect than the same being that does exist. In fact, it is contradictory to think of it as not existing, we can only think of it as existing. One could also argue that existence is more perfect than non-existence as a non-existent object could only theoretically fulfil its role, whereas an existent object could also fulfil its actual role. One could argue that there are some things that when they exist are not more perfect, pain for example, but the response to that is simply that they are a more perfect as an instantiation of pain. Another counter to the idea that existence is a perfection is that we are simply making a mistake when we talk of existence in those terms. Kant tells us that 'exists' functions grammatically as a predicate, but not logically. It is not the same sort thing to say that a thing exists as it is to say that a thing is yellow. Frege and Russell are able to express this more specifically by demonstrating how 'exists' functions as a second order predicate, it only serves to ascribe properties to other properties. Thus one can say 'x is yellow' and alter your conception of x, whereas to say 'x exists' does not add anything to our conception of x. Therefore, you cannot talk about existence as ascribing a property to an object, and thus to say ëxistence is a perfection" is an incorrect usage. A being is no more or less perfect for existing under this system of logic.

From this examination, it seems as though we can only say for sure that the argument can stand up if we start out by assuming the conclusion, making the argument invalid. However, there does seem to be a weaker form of the argument that cannot be wholly refuted, but it only runs to saying 'If there were a God he would exist'.

Next I shall consider the Cosmological argument, as first proposed by Plato. He began by noting that every event has a cause, and then goes on to assert that there must be a first cause, and that that cause must have some special status, since it does not itself require a cause. The full version of the argument can be considered as having three possible outcomes, assuming its premises. Either, one can say that the first appearance of physical substance was self-causing, or that the first appearance was caused by an finite chain of non-physical events, or that the first appearance was caused by an infinite chain of non-physical events. Alternatively, one can claim that there was no first appearance, and that physical substance had no beginning.

I shall begin by examining the idea that there was a finite chain of non-physical causes that led to the creation of physical substance. This solves the problem of what caused physical substance, but leaves us with a question hanging over what causes non-physical events. There are two possible answers to this. Firstly, we can assert that non-physical events do not follow the same laws as physical events, a reasonable assertion, and can then postulate that non-physical events do not necessarily require causes. However, if we are going to sweep away all knowledge of how these non-physical events behave, then we cannot surely meaningfully talk about them. As an example, what evidence do we have for the idea that non-physical events do not require causes? It would seem we have none, as no-one has an example of a non-physical event that they can demonstrate. Some would suggest that the mind is non-physical, but it is not demonstrable in an appropriate manner. This line of reasoning seems then to be meaningless speculation. An alternative hypothesis is that the first non-physical event was self-causing. This is the line of reasoning that leads us towards suggesting that a deity exists, as something that can cause itself must be virtually, if not actually, omnipotent. If we have this self-causing non-physical event, then the rest of the chain seems to fall into place. However, if we can bend the rules for a non-physical event and allow it to be self-causing, why can we not do so for the first physical event. We can postulate that the first appearance of physical substance was the cause of itself with the same validity as we can postulate that the first non-physical event was the cause of itself. This could be said to lead to a Spinoza type God as Nature position, but that is only if we require that causation requires intelligence, or some form of mental activity. This holds for self-causing non-physical events also, as there is no reason why a cascade of non-physical events cannot have been begun by a self-causing non-intelligent 'thing.' The argument for an infinite chain of non-physical events is also open to this same style of comparative attack, as one can say why if we can have infinite non-physical events can we not have infinite physical events. These do not destroy the argument, they simply make it inconclusive, and therefore useless.

The conclusions of the argument are not the only weak points within it. The premise of causality is also vulnerable to attack. Our belief in causality is an example of our use of inductive reasoning. The only reason for saying that x causes y is that when we perceive event y, event x immediately precedes it. This is no reason to assume that y only happened because of x. Hume points out that it could simply be a case of coincidence that y happens after x. Even if we assert that every time y happens, x precedes it, and every time x happens y follows it, there is still nothing inherent in that that means there is a causal relationship, we merely assume one in order to aid our interpretation of the world.

This argument is perhaps one of the weakest for any sort of recognisable deity, as all it can possibly prove is that there is a powerful entity that set off a chain of events that led to our current state. The nature of that entity is almost entirely undetermined by this argument. It does not even require that creation was a deliberate act, nor that this entity is even aware of physical substance, as it would, by necessity be purely non-physical.

I shall now move on to an examination of the Teleological argument. This argument rests on the idea that order can only be the result of intelligence, and that as the universe is ordered, it must be the result of intelligence. Possibly the best known formulation of this argument is in Hume's Dialogue's Concerning Natural Religion, where Cleanthes postulates that "the curious adapting of means to ends, throughout all nature, resembles exactly... the productions of human contrivance." Since the universe can be seen as having order in it, in the form of recognisable laws of physics that allow for continuity, there must be intelligence behind it. To me this is a very weak argument, as we can see examples of order that are not the direct result of intelligence, such as plants, the cause of which are other plants, which do not appear to have intelligence. The only way to say that they are the creation of intelligence is to go all the way back to our deity, and say that since he instigated order, plants became possible. However, we then have no reason to say that order is the result of intelligence, as we can have no comparison. We cannot say that since intelligence x made ordered thing y, we can see that order comes from intelligence, as the only possible example of this is the first one, the creation of order itself.

From the examination of these three principle arguments I draw the conclusion that none of them provide a definitive proof for the existence of God, certainly not the Christian style God. The Ontological argument can only give information about the potential God, the Cosmological argument is susceptible to Schopenhaur's objection that people tend to stop saying event shave to have causes once we reach non-physical events, and the Teleological argument seems to rely far too heavily upon our own subjective interpretation of order in the universe. Simply because we see it as ordered doesn't mean it is, it is quite likely to be merely our own imposition upon it. For me, none of the arguments are convincing, as the best they can muster is to be inconclusive, that is to show that we cannot outright deny the existence of God.


J. Cottingham, Descartes: Selected Philosophical Writings (New York: CUP, 1999) J. Hick, The Existence of God (New York: Macmillan, 1976) D. Hume, Dialogues Concerning Natural Religion (London: Penguin Books, 1990)

File translated from TEX by TTH, version 2.92.
On 24 Mar 2002, 23:50.