Theistic Arguments (A.C. Grayling, 2013)

NOTE: The following article is taken from The God Argument: The Case Against Religion and for Humanism, pp. 65-72


In this and the following chapters I examine ‘arguments for the existence of God’, as the standard phrase has it. I do it in detail, and in the context of other considerations that arise in discussion of these arguments – semantic, psychological and polemical considerations chief among them – something that is too rarely done, although doing it brings a great deal of light to bear on the arguments themselves.

The importance of this contextual approach to these arguments is that apologists for religion wish to sustain the impression that such discussions are rational in the way that discussion of genetics or Roman history is rational, and in the way that discussion of fairy lore, or astrological concepts as employed in assertions such as ‘Mercury is in opposition to Mars in the Third House’, is not. It is rational to discuss the fact that lots of people believe that astrology reveals character and destiny, but it is not rational to discuss whether people born under the sign of Aries are typically more rash than those born under the sign of Gemini. Theological discourse is of this latter type – ‘Is the Son of one substance with the Father?’ and ‘Does the Holy Ghost proceed from the Father and the Son, or from the Father alone?’ are utterances that have caused wars, persecutions, schisms and burnings at the stake; and they are exactly of the same category as ‘Are Aries people more rash than Gemini people?’ – that is, they are part of the internal discourse of the belief system, and they only make sense to anyone who has accepted the premises and parameters of the belief system already.


Let us call discussion about why people believe in astrology an ‘external discussion’ and discussion about the different character traits of Aries and Gemini people an ‘internal discussion’. We then see that ‘arguments for the existence of God’ are intended as external discussions, and therefore an invitation to rational argument about theistic claims. One large problem with this, as we shall repeatedly see, is that to have this external discussion one has to make use of internal discussions in the crucial sense that it is these latter which tell us what the former are about. The question is: what is this thing theists call ‘god’ that these arguments are going to try to prove exists? How can you prove such a thing exists (or: such things exist) unless you know what is under discussion? You can only say what is talked about by reference to the internal discussions; but the internal discussions only make sense if the external discussion has shown that there is something to talk about.

A Teleological Argument Print by Daniel G Walczyk
A Teleological Argument Print by Daniel G Walczyk

This problem has to be borne in mind throughout what follows.

Anyone who studies philosophy or theology immediately meets with a standard set of ‘arguments for the existence of God’. The best known is the ‘argument from design’, called by philosophers the ‘teleological argument’ from the Greek word telos meaning ‘aim, end or purpose’.

The second well-known argument students meet with is the ‘ontological argument’, an argument that claims to establish the existence of a deity by reason alone. It does this by claiming that a deity exists by definition.

A third argument is the ‘cosmological argument’, like the teleological argument an attempt to provide an empirical basis for the existence of a deity, but this time inferring it from the supposed fact that the world cannot be its own cause or ‘ground’.


Other arguments include a moral argument, roughly speaking that there cannot be morality unless there is a deity, and a collection of further considerations intended to establish that theistic belief is rational, desirable, or at very least prudent.

In all these cases an argument is being offered, and arguments invite rational scrutiny. The word ‘argument’ is here being used in its logical sense to mean the derivation of a conclusion from premises. Premises can support a conclusion either demonstratively and conclusively, as in the formal deductive systems of logic and mathematics, or by rendering the conclusion plausible or persuasive to the point of making it irrational not to accept it (and, if relevant, to act upon it). This distinction was explained earlier. Argument of the latter kind is inductive. Arguments endeavouring to establish the ‘existence of God’ can be either deductive or inductive.

I put ‘scare’ quotes around the phrase ‘existence of God’ because, as noted earlier, use of the capitalised word ‘God’ makes it look like a proper name which, in virtue of being one, carries the suggestion that it names something. There is a major debate in philosophy about naming, which we know cannot have straightforward existential implications given that some names, such as ‘Harry Potter’, name fictional entities; and this alerts us to the possibility of question-begging assumptions being involved in arguments that claim to establish the existence of something when talking about that something already seems to presuppose its existence, or at very least the possibility of its existence.

It is better to insist on the formula I began with, namely, ‘arguments for the existence of deity’, although we immediately find ourselves in need of an understanding of what is meant by ‘deity’ or ‘god’ and their cognates. The difficulties relating to this, mentioned earlier, greatly complicate the task of analysing attempted proofs of existence, for how can one prove the existence of something unless one knows what it is supposed to be?

There is a further complication. This is that most religious people do not, of course, subscribe to their religion because of arguments in favour of it, still less arguments establishing the existence of the deity central to it. The arguments examined here are without exception rationalisations of beliefs which are already accepted. In the great majority of cases, people belong to their religion because it is the religion of their parents, learned in childhood and thereafter constantly and in diverse ways reinforced by being present and observed in their communities. When a religion is adopted in later years the impulse for it is almost wholly emotional rather than rational; proselytising of teenagers and adults typically targets loneliness, confusion, failure, grief, anxiety and depression as opportunities for conversion. The psychological support given by the fellowship thus offered is attributed by the convert to his newly formed relationship with that religion’s deity – or so the sceptical observer of this phenomenon would say. It is difficult to find sound empirical data on the length of time that converts remain converted; anecdotal evidence suggests that late conversion, even when it produces the zeal that makes converts more devoted than the (so to speak) natives of the faith, more often than not does not ‘stick’.


Because non-rational motivation plays a much greater role in prompting and supporting religious commitment than ‘arguments for God’s existence’, it merits particular challenge, because having faith – holding beliefs and accepting doctrines either without evidence or even in the face of countervailing evidence, which most religious people actually regard as a virtue – directly controverts canons of intellectual integrity. ‘Faith’ is not a respectable or admirable thing; having been so long paraded as a virtue and worthy of respect, the truth is otherwise: its critics have no compunction in saying that it is irresponsible, lazy and too often dangerous.

Because it would be fruitless to attempt proof of the existence of something which is undefined, ineffable, or too mysterious for finite minds to understand or describe, one has constantly to remind oneself of the difficulty in attaching meaning to the word ‘god’ or ‘deity’ in talking of arguments for the existence of such a thing. As noted in Chapter 2, it is common for religious apologists to reply to critiques of such arguments by claiming that deity is incomprehensible, which puts an end to the debate because, tautologously, there is nothing to be said about what nothing can be said about. Yet as it happens, religious apologists actually find a lot to say about such a thing after all – not just that it exists, but that it has a specifiable nature (it is ‘love’, it is omniscient, omnipotent, morally pure and the like), and they even know with a great deal of precision what it requires of mankind in the way of behaviour and commitments. This does not appear to strike apologists as contradictory, which it is. The ineffability claim is inconsistent with the fact that most religions have scriptures or literatures which say a great deal about the divine being or beings that figure in them. The standard arguments for the existence of deity draw upon this material. For example, the classic version of the ontological argument premises that the deity whose existence it seeks to prove possesses ‘all perfections’, meaning at very least no limits to or defect in any such positive property as wisdom, purity and goodness (the opposites of ignorance, lustfulness and evil). The positive properties are said to include power – indeed, all-power: omnipotence – but power cannot be a positive property unless qualified as such, because presumably evil agencies (the devil and his minions) have some degree of power – they cannot work their evil otherwise. However: the all-good, pure, all-powerful, all-wise being is the traditional conception of deity in the Judaeo-Christian tradition, of which there are offshoots including Islam and Mormonism; and the concept of deity in all these is therefore similar.

Note that the positive characteristics attributed to deity are the ones we humans most approve of and wish we had. Obviously enough we are finite beings, limited both in intelligence and in knowledge, and given to various weaknesses and failings. Our lives are brief and often beset with trouble and illness. Deities tend to have the opposite of these things, and the promise held out to those who will obey their priests is that they will achieve the same at some future time – after death usually, though some sects are more optimistic. Traditional definitions of deity turn on negations of the human lot, and are therefore anthropomorphic in reverse: based on desirable opposites of undesirable human characteristics. Those who say that the nature of deity is ineffable avoid having these considerations offered as a count against theism, but then ineffabilitists are strictly speaking not in a position to offer proofs for the existence of deity, or (more weakly) reasons why anyone might think there is any such thing: if they are serious about the ineffability claim, they have painted themselves into the corner of not being able to say very much, or indeed anything.


This raises an interesting point. Think of trying to get an investor to put money into a scheme which cannot be described or talked about. It is unlikely that anyone was ever persuaded to believe in a we-know-not-what; an indescribable deity has to be arrived at by retreat from a defined deity, a deity of tradition or childhood. The other alternative is that the person persuading someone to believe in a we-know-not-what would have to have great charisma and plausibility; no doubt this sometimes happens. But even here there has to be something to start with, some sense of the word ‘god’, which the persuader can say is not correctly understood (by all the other religions no doubt) or ‘cannot be spoken of or described’.

The standard arguments for the existence of deity are said to belong to what is known as ‘natural theology’ rather than revealed religion, but we see from the foregoing considerations that what natural theology owes revealed religion in the way of a concept of deity is obvious and inescapable. Therefore it is the deity of tradition – eternal, omnipotent, omniscient, morally perfect, in short infinitely possessed of all the positive characteristics that people most admire and wish they had – whose existence the arguments for the existence of deity attempt to prove. This type of deity is to be understood as the subject under discussion throughout the following.

Another point that has to be mentioned in order to be set aside is that the deity of tradition is always referred to as ‘he’ (believers write ‘He’); this is an accident of history, but in personalising the deity it once again forces a circular assumption that such a thing exists even in the process of attempting to prove its existence.


The two most often discussed arguments are the ‘teleological’ and the ‘ontological’. I begin with them, taking each in turn.