Faith and Evidence
Dawkins' emphasis on evidence-based reasoning leads him to adopt a strongly critical attitude towards any beliefs that are inadequately grounded in the observable. `As a lover of truth, I am suspicious of strongly held beliefs that are unsupported by evidence.' One of his core beliefs, repeated endlessly in his writings, is that religious faith is `blind trust, in the absence of evidence, even in the teeth of evidence.' Faith, Dawkins argues, is `a kind of mental illness,' one of the `world's great evils, comparable to the smallpox virus but harder to eradicate.' This is to be contrasted with the natural sciences, which offer an evidence-based approach to the world. And quite rightly so. But I wonder if his own strongly held atheist views are quite as supported by the evidence as he seems to think?
Dawkins here opens up the whole question of the place of proof, evidence, and faith in both science and religion. It is a fascinating topic. But is it really quite as simple as Dawkins suggests? I certainly thought so during my atheist phase, which ended towards the end of 1971, and would then have regarded Dawkins' arguments as decisive. But not now.
Let's begin by looking at that definition of faith, and ask where it comes from. Faith `means blind trust, in the absence of evidence, even in the teeth of evidence.' But why should anyone accept this ludicrous definition? What is the evidence that this is how religious people define faith? Dawkins is coy at this point, and adduces no religious writer to substantiate this highly implausible definition, which appears to have been conceived with the deliberate intention of making religious faith seem a piece of intellectual buffoonery. I don't accept this idea of faith, and I have yet to meet a theologian who takes it seriously. It cannot be defended from any official declaration of faith from any Christian denomination. It is Dawkins' own definition, constructed with his own agenda in mind, being represented as if it were characteristic of those he wishes to criticise.
What is really worrying is that Dawkins genuinely seems to believe that faith actually is `blind trust', despite the fact that no major Christian writer adopts such a definition. This is a core belief for Dawkins, which determines more or less every aspect of his attitude to religion and religious people. Yet core beliefs often need to be challenged. For, as Dawkins once remarked of Paley's ideas on design, this belief is `gloriously and utterly wrong'.
Faith, Dawkins tells us, `means blind trust, in the absence of evidence, even in the teeth of evidence.' This may be what Dawkins thinks; it is not what Christians think. Let me provide a definition of faith offered by W. H. Griffith-Thomas (1861-1924), a noted Anglican theologian who was one of my predecessors as Principal of Wycliffe Hall, Oxford. The definition of faith that he offers is typical of any Christian writer.
[Faith] affects the whole of man's nature. It commences with the conviction of the mind based on adequate evidence; it continues in the confidence of the heart or emotions based on conviction, and it is crowned in the consent of the will, by means of which the conviction and confidence are expressed in conduct.
It's a good and reliable definition, synthesising the core elements of the characteristic Christian understanding of faith. And this faith `commences with the conviction of the mind based on adequate evidence.' I see no point in wearying readers with other quotations from Christian writers down the ages in support of this point. In any case, it is Dawkins' responsibility to demonstrate that his skewed and nonsensical definition of `faith' is characteristic of Christianity through evidence-based argument.
Having set up his straw man, Dawkins knocks it down. It is not an unduly difficult or demanding intellectual feat. Faith is infantile, we are told - just fine for cramming into the minds of impressionable young children, but outrageously immoral and intellectually risible in the case of adults. We've grown up now, and need to move on. Why should we believe things that can't be scientifically proved? Faith in God, Dawkins argues, is just like believing in Santa Claus and the Tooth Fairy. When you grow up, you grow out of it.
This is a schoolboy argument that has accidentally found its way into a grown-up discussion. It is as amateurish as it is unconvincing. There is no serious empirical evidence that people regard God, Santa Clause and the Tooth Fairy as being in the same category. I stopped believing in Santa Claus and the Tooth Fairy when I was about six years old. After being an atheist for some years, I discovered God when I was eighteen, and have never regarded this as some kind of infantile regression. As I noticed while researching The Twilight of Atheism, a large number of people come to believe in God in later life - when they are `grown up'. I have yet to meet anyone who came to believe in Santa Claus or the Tooth Fairy late in life.
If Dawkins' rather simplistic argument has any plausibility, it requires a real analogy between God and Santa Claus to exist - which it clearly does not. Everyone knows that people do not regard belief in God as belonging to the same category as these childish beliefs. Dawkins, of course, argues that they both represent belief in non-existent entities. But this represents a very elementary confusion over which is the conclusion and which the presupposition of an argument.
The highly simplistic model proposed by Dawkins seems to recognize only two options: 0% probability (blind faith) and 100% probability (belief caused by overwhelming evidence). Yet the vast majority of scientific information needs to be discussed in terms of the probability of conclusions reached on the basis of the available evidence. Some have argued for assessing the reliability of probability of a hypothesis on the basis of Bayes' theorem. Such approaches are widely used in evolutionary biology. For example, Elliott Sober proposed the notion of `modus Darwin' for arguing for common Darwinian ancestry on the basis of present similarities between species. The approach can only work on the basis of probability, leading to probabilistic judgements. But there's no problem here. It's an attempt to quantify the reliability of inferences.
One of the most striking things about Dawkins' atheism is the confidence with which he asserts its inevitability. It is a curious confidence, which seems curiously out of place - perhaps even out of order - to those familiar with the philosophy of science. As Richard Feynman (1918-88), who won the Nobel Prize for physics in 1965 for his work of quantum electrodynamics, often pointed out, scientific knowledge is a body of statements of varying degree of certainty - some most unsure, some nearly sure, but none absolutely certain. Yet Dawkins seems to deduce atheism from the `book of nature' as if it were a pure matter of logic. Atheism is asserted as if it was the only conclusion possible from a series of axioms. Yet given that the natural sciences proceed by inference from observational data, how can Dawkins be so sure about atheism? At times, he speaks with the conviction of a believer about the certainties of a godless world. It is as if atheism was the secure and inevitable result of a seamless logical argument. But how can he achieve such certainty, when the natural sciences are not deductive in their methods? Others have examined the same evidence, and come to quite different conclusions. As will be clear from what has been said thus far, Dawkin's insistence that atheism is the only legitimate worldview for a natural scientist is an unsafe and unreliable judgement.
Yet my anxiety is not limited to the flawed intellectual case that Dawkins' makes for his convictions; I am troubled by the ferocity with which he asserts his atheism. One obvious potential answer is that the grounds of Dawkins' atheism lie elsewhere than his science, so that there is perhaps a strongly emotive aspect to his beliefs at this point. Yet I have not come across anything that forces me to this conclusion. The answer has to lie elsewhere.
I began to find an answer to my question while reading a careful analysis of the distinctive style of reasoning that we find in Dawkins' writings. In an important comparative study, Timothy Shanahan pointed out that Stephen Jay Gould's approach to the question of evolutionary progress was determined by an inductivist approach, based primarily on empirical data. Dawkins, he noted, `proceeded by elaborating the logic of Ôadaptationist philosophy' for Darwinian reasoning.' This being the case, Dawkins' conclusions are determined by a set of logical premises, which are ultimately - yet indirectly - grounded in the empirical data. `The very nature of a valid deductive argument is such that, given certain premises, a given conclusion follows of logical necessity quite irrespective of whether the premises used are true.' In effect, Dawkins uses an essentially inductive approach to defend a Darwinian worldview - yet then extracts from this worldview a set of premises from which secure conclusions may be deduced.
Although Shanahan limits his analysis to exploring how Gould and Dawkins arrive at such antithetically opposed conclusions on the issue of evolutionary progress, his analysis is clearly capable of extension to his religious views. Having inferred that Darwinism is the best explanation of observation, Dawkins proceeds to transmute a provisional theory into a certain worldview. Atheism is thus presented as the logical conclusion of a series of axiomatic premises, having the certainty of a deduced belief, even though its ultimate basis is actually inferential.