Print

Who made God?

An extract from Professor Edgar Andrews's book

The question ‘Who made God?’ crops up in a probability argument presented by Richard Dawkins in The God Delusion.

And here we must begin with a lengthy note of caution. Probability arguments are often used in debates relating to God, science, creation and evolution, but they are almost always used naively. Probability theory is a branch of mathematics, concerned initially with discrete numbers (like the probability of getting heads or tails when you toss a coin), but extending also to continuous ‘distribution functions’ (like a curve showing people’s average income plotted against their age). Either way, it’s all mathematics. The theory can, of course, be applied to the real world as in the examples given, but there is nothing of the real world built into the theory itself.

For example, its coin-tossing calculations take no account of the possibility that the coin might land on its edge and get stuck in that position — something that could happen quite often if you’re tossing the coin on a soft sandy beach.

Origin of life

A famous example is Sir Fred Hoyle’s calculation of the probability of spontaneously forming a protein molecule from its amino-acid constituents, which he likened to the probability that a whirlwind in a junk yard would assemble a Boeing 747. Good fun, of course, especially for creationists, but it’s just number-juggling and takes no account of physical and chemical realities (such as the presence or absence of water, catalysts and competing chemical reactions). There is no way of knowing whether such abstract calculations have any meaning in a realistic chemical scenario.

Equally, however, the claim that anything (like the origin of life) must happen by chance if you wait long enough is also fallacious. Again, this claim is based on the idea that no mathematically possible event has zero probability. But what is mathematically possible may be physically impossible. Consider the claim that 100 monkeys banging the keys of 100 typewriters (with no lunch breaks) will eventually produce the works of Shakespeare. Not true. The well-established chemical theory of rate processes tells us that all real-world processes are reversible — they can go backwards as well as forwards, and the net result is the balance between forward steps and backward steps. For a process to go forward on average requires a payback of free energy. If there is no such payback, each forward step will sooner or later be cancelled out by a backward step and the system will remain in stasis.

Applying this to the monkeys, we see that progress towards a Shakespeare folio would actually go backwards, not forwards, since the likelihood of the next keystroke being an error is much greater than the likelihood of it being correct (there being 26 letters in the alphabet, one would be correct but 25 would be mistakes). So if you started the monkeys with a typescript lacking just a single sonnet but otherwise complete and correct, it wouldn’t be long before The Merchant of Venice was in total disarray — let alone Much Ado About Nothing.

Instant Shakespeare

The only way the monkeys will ever complete their task successfully is if you artificially introduce a ‘payback’ system in which you remove and store any correct text the monkeys might generate, putting this text into quarantine so that it cannot be corrupted again. Some readers will recognise this as the fundamental principle behind Darwinian natural selection, but the monkeys-to-Shakespeare scenario, being just a numerical argument, admits no such intervention and is completely fallacious.

In order to apply mathematical probability theory to the real world, therefore, requires us to build the real world into the scenario. In physics and chemistry the result is something called ‘statistical mechanics’, which is one approach to the science of thermodynamics. The mathematical theory of probability can only be applied correctly to the real world through the filter of thermodynamics.

The probability of God

In the science of thermodynamics the statistical probability that a given system (or arrangement) will arise spontaneously in nature is related mathematically to the degree of order or complexity of the system. Systems with low complexity (high entropy or randomness) are likely to arise spontaneously, while systems with high order or complexity (low entropy or randomness) are unlikely to do so.

Oh, sorry for the techno-talk; let’s put it this way. If you see a pile of bricks by the roadside, they could have fallen off the back of a lorry (low complexity, high probability of happening spontaneously). But if you see a neat bungalow built of bricks it is most unlikely to have arisen by accident (high complexity, low probability of happening by chance).

Don’t know any thermodynamics? So, let’s learn some. Take a china soup bowl and drop it on a stone floor (preferably without the soup). Very likely it will become less ordered (or less organised) — instead of one piece of china having a nice symmetrical shape, there will be many pieces of different shapes and sizes. Now collect these pieces together and drop them on the same floor. Did they reassemble themselves into the ‘ordered’ form of an unbroken bowl? No? Oh well, that’s thermodynamics for you. Disordered states of matter arise spontaneously; ordered states do not.

Of course, you can sit down with a tube of superglue and painstakingly stick the bits together to reassemble a single bowl. You won’t make a perfect job of it, of course, but you will — after expending much effort and some skill — make the pieces more ‘ordered’ by joining them together in a unique relationship (any other relationship would let the soup leak out). But the increased order of the repaired bowl has only been achieved at the cost of directed energy input and intelligent effort. It could never have happened on its own.

Sticky fingers

Consider another example. If I let a small child get her sticky fingers on my computer keyboard, this page would continue as a random jumble of characters — one meaningless sequence out of an almost infinite range of equally possible sequences. If, on the other hand, I retake control and continue to develop my argument in carefully chosen words, there are only a limited number of symbol sequences that could serve this purpose. The sticky finger jumble is a high probability arrangement because there are so many different ways to achieve it, but it has low complexity because it is random and meaningless. By contrast, my reasoned argument requires a low-probability arrangement of symbols, because there are only a limited number of ways of expressing what I want to say. But because what I say is (hopefully) meaningful, my page is a system of high complexity (high information content) and has a negligible probability of occurring accidentally or spontaneously.

Now we know that the physical world represents a highly improbable arrangement of matter and energy — an extremely improbable arrangement as it turns out. The study of physics has shown that the laws and fundamental constants of nature give every appearance of being fine-tuned to permit the existence of intelligent life on earth (this is known as the anthropic principle).

Dawkins boiled down

The argument for the improbability of God, as advanced by Dawkins and others, seems to boil down to the following reasoning: (1) by common consent, the world is a highly improbable and complex system; (2) if God created the world he must be more complex than the world he created; therefore (3) God is less probable than the world; indeed, he is fantastically improbable; so (4) God probably doesn’t exist. Although produced with a flourish, the argument holds no more water than a sieve.

Firstly, we have to accept the dubious assumption that the science of thermodynamics (or statistical mechanics) — which was developed to describe the behaviour of matter and energy — applies to theology and God. You might just as well apply it to love, music or politics, but I promise you it won’t work.

Sleight of hand

Secondly, there is excessive sleight of hand in the use of the word ‘improbable’. In thermodynamics it refers to the number of different ways a system can be arranged or ordered. The repaired soup bowl is highly improbable thermodynamically speaking because it is uniquely formed — there is only one way to arrange the pieces to rebuild that particular shape. By contrast, the broken bowl can consist of many different arrangements of shards. It could break into just two pieces or ten or 100. An almost infinite variety of shapes and sizes could result from the fracture. There are thus a huge number of possible arrangements for a broken bowl — and that means that the broken bowl is a highly probable entity, thermodynamically speaking. But because the repaired bowl is ‘improbable’ and the broken bowl ‘probable’ it doesn’t mean that all soup bowls are broken. However, let’s skip over these logical death-leaps in Dawkins’s reasoning and suppose that in some way the argument does have relevance to God. What does it actually prove?

Special pleading

Having agreed, presumably, that the world does exist in spite of its extreme complexity and organisation (high improbability), the argument goes on to say that God is unlikely to exist because he is ... well, er, highly improbable. OK, so God is arguably more complex and thus less probable than the physical universe. But by what logic must we accept that one highly improbable entity exists (the universe) while another highly improbable entity (God) does not exist — simply because he is too complex or organised to do so? In my neck of the woods they call that special pleading (when they are feeling polite).

This article is an edited extract from Who Made God? by Edgar Andrews, published by Evangelical Press, and is used with permission.