You’ve read one of your four complimentary articles for this month.
You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please SUBSCRIBE!
Science, Ockham’s Razor & God
David Glass and Mark McCartney say Ockham’s razor doesn’t cut it with God.
The idea that science has explained God away is very popular. The suggestion is that as science explains more and more about the world there is less and less need for God. Sometimes this is expressed in terms of Ockham’s razor. William of Ockham was a medieval philosopher and theologian, and his famous ‘razor’ is the idea that “It is futile to do with more things that which can be done with fewer.” Applied to science and God, the implication seems to be that if science can explain the world around us on its own, there is no need for science and God. There is no need for two explanations when one will do.
This kind of reasoning is central to the New Atheism. The late Christopher Hitchens appealed explicitly to Ockham’s razor as part of his case against God and the same idea is found in Richard Dawkins’ The God Delusion (2006) where he claims that “Historically, religion aspired to explain our own existence and the nature of the universe in which we find ourselves. In this role it is now completely superseded by science.” In claiming that science explains God away, Dawkins focuses in particular on “Darwin’s destruction of the argument from design.” Outside the New Atheism, cosmologist Sean Carroll sums up the view clearly, “Over the past five hundred years, the progress of science has worked to strip away God’s roles in the world … Two thousand years ago, it was perfectly reasonable to invoke God as an explanation for natural phenomena; now, we can do much better.” (The Blackwell Companion to Science and Christianity, eds J.B. Stump and A.G. Padgett, p.196, 2012).
But some caution is needed. If Ockham’s razor is used properly it can be a very helpful tool, but if used incorrectly it can become a dangerous instrument. First, consider a scenario where Ockham’s razor works. Suppose your car won’t start. Two possible explanations are a) that the battery is faulty and b) that the starter motor is faulty. You call a mechanic who quickly determines that there is indeed a problem with the battery. Since this would explain why the car didn’t start, there is no longer any reason to think that there is a problem with the starter motor. Ockham’s razor removes the need for the further explanation. Or to put it another way, the truth of the faulty battery hypothesis has explained away the faulty starter motor hypothesis. Of course, it is possible that there is a problem with both the battery and the starter motor. That cannot be ruled out until the car starts, but it would be really unlucky; based on the evidence so far and the application of Ockham’s razor, one explanation is sufficient.
Now consider another scenario. A road traffic accident has taken place near a set of traffic lights. Based on preliminary evidence relating to the scene and the condition of one of the drivers, the police propose two possible explanations: a) that the driver of one of the cars went through a red light and b) that this same driver had been drinking. Sure enough, further evidence confirms that the driver had indeed been drinking. Could we now apply Ockham’s razor as before and conclude that there is no longer any reason to think the car went through a red light? No, in this case it is quite plausible to think that the two explanations go together. The two hypotheses are needed to account for all the evidence and alcohol consumption could explain why the driver went through a red light. It would be inappropriate to apply Ockham’s razor in this case since doing so could very easily lead us astray.
So how do you decide in a particular case whether Ockham’s razor can be applied legitimately to use one hypothesis, A, to dismiss another hypothesis, B? In some recent work (‘Can evidence for design be explained away?’, in Probability in the Philosophy of Religion, eds J. Chandler and V. Harrison, 2012; and ‘Explaining and explaining away in science and religion’, Theology and Science, 12(4), 2014), we’ve proposed both a formal account based on probability theory and an informal account based on answering a number of questions including the following:
Question 1: Are the two hypotheses A and B mutually exclusive? If so, then accepting A requires rejecting B and indeed vice versa. Clearly, this can give rise to a trivial version of Ockham’s razor, but it is much more interesting to consider cases where A and B are not mutually exclusive.
Question 2: Does one of the hypotheses depend on the other? If so, that reduces the potential for applying Ockham’s razor.
Question 3: How likely is it that the available evidence would have been produced by each hypothesis on its own? If the answer is ‘very likely’, that increases the potential for Ockham’s razor.
Question 4: Is there good reason to accept hypothesis A? If not, this reduces the impact of Ockham’s razor in cases where it could occur.
Let’s apply these questions to our two scenarios. In the case of the car not starting, while the two explanations were not mutually exclusive (Question 1), the answers to Questions 2-4 help us see why Ockham’s razor is appropriately applied. It seems that neither hypothesis depends strongly on the other (Question 2), either explanation could account for the evidence on its own (Question 3), and the mechanic has found out that one of them is indeed true (Question 4). In the case of the road traffic accident, the two explanations are not mutually exclusive (Question 1), there is a dependence between them since one could explain the other (Question 2), and both are needed to account for all the evidence (Question 3). In this case, confirmation that the drink driving hypothesis is true (Question 4) does nothing to undermine the other explanation.
William of Ockham’s actual razor, recently discovered in a drawer in a Belgian monastery
How does all of this relate to science and God? How can we use it to evaluate the claim that via an application of Ockham’s razor the progress of science explains God away? Let’s consider each of our questions in turn.
In response to Question 1, it is clear that science and theism are not mutually exclusive. Sometimes atheist popularisers might claim that they are, but there’s no good reason to think that science and theism are logically incompatible. If one is to argue that science removes the need for God, it will have to be an argument – like that of the New Atheists – that involves a non-trivial application of Ockham’s razor. The point would be, not that science logically refutes God’s existence, but that it makes God redundant.
Question 2 asks whether there is a dependence between science and theism. Many theists have argued that there is indeed. Some versions of the cosmological argument claim that the very existence of the universe, and hence science, depend on God. Similarly, an important version of the design argument defended by Richard Swinburne in The Existence of God (2004) is that theism provides the best explanation for the order in the universe expressed in the laws of science. The point of that argument is not to undermine science, but just to say that it cannot do the impossible since whatever the most fundamental scientific laws (or law perhaps) turn out to be, they cannot themselves be given a scientific explanation. Yet, maintains Swinburne, it still makes sense to search for an explanation for the fact that the universe operates according to such laws and the best explanation is provided by theism. Here, the idea is that theism provides a different kind of explanation; a personal, not a scientific, explanation. Design arguments based on the fine-tuning of physical parameters would provide a further reason for thinking that science depends on theism, as would theistic arguments based on the comprehensibility and mathematical nature of the universe. Others have argued for a historical dependence between science and theism on the grounds that modern science developed in a theistic context and that it was because of theistic beliefs, such as belief in God as a rational creator and God’s freedom in creation, that science took off.
If there is in fact any kind of dependence of science on theism and hence a positive answer to Question 2, the case for applying Ockham’s razor to explain God away through science is very weak indeed. Proponents of applying Ockham’s razor in this way will be quick to object to the theistic arguments in the last paragraph; our purpose here is not to defend them but to draw attention to the fact that a number of them, as well as objections to them, are relevant to our discussion and that serious philosophical points are put forward on both sides. The point is that proponents of applying Ockham’s razor against God can’t simply assume that these theistic arguments fail if their use of Ockham’s razor is to have any force. Maybe there are serious flaws in these theistic arguments, maybe not, but to avoid begging the question in their application of Ockham’s razor, its proponents need to engage in detailed philosophical work to show that these theistic arguments fail, or else risk a positive answer to Question 2.
This weakens the appeal of the strategy. It might have been thought that a combination of science and Ockham’s razor would somehow shortcut this process by avoiding the need to refute detailed philosophical arguments for God’s existence, but that is not the case. It’s worth comparing this argument against God with the problem of evil (the claim that the existence of suffering in the world makes it unlikely that there is a God who is simultaneously all-good, all-powerful and all-knowing). Whatever the merits of the problem of evil as an argument against God, it can be considered to some extent in isolation from other theistic arguments. One could argue that the problem of evil has some force against the existence of God while setting aside the question of whether other arguments might count in favour of God’s existence. Given the importance of Question 2 in determining whether Ockham’s razor can be applied, it seems that this isolation from other theistic arguments isn’t available in the current context. It’s interesting to note that while using science and Ockham’s razor to explain God away is very popular in the New Atheism and in Internet discussion, such an argument is rarely to be found in the philosophical literature. Atheist philosophers tend to focus on more standard objections to belief in God such as the problem of evil and are perhaps aware of just how difficult it is to make a convincing argument based on science and an application of Ockham’s razor.
Question 3 is more difficult to answer in an abstract sense, but becomes more relevant in particular cases where science might be thought to explain God away. We will consider it presently in a specific discussion of evolution. It is important to note that theists often think about theistic explanations of features of the world as working not in isolation from science, but via scientific laws. Natural phenomena are to be explained directly by science and only indirectly by God. One way of putting this is to say that God acts through causes which can be described scientifically. Think of this is in terms of a causal chain where A causes B and B in turn causes C; A’s influence on C comes via B. An interesting result of our formal account of the applicability of Ockham’s razor was to show that it does not apply to causal chains. Here’s a simple example. Suppose your friend has arranged to get the train and meet you at the railway station after work, but he isn’t there at the specified time. One explanation might be that he missed the train while another might be that he was delayed at work. The evidence shows he missed the train, but even so, Ockham’s razor clearly wouldn’t dispose of the hypothesis that he was delayed at work. Why? Because his being delayed at work could have caused him to miss the train. Similarly, if the above account of God acting through natural causes is correct, then Ockham’s razor cannot be applied against God even if natural causes appeared to explain all of the physical evidence available to us.
What about Question 4? We surely do have good reasons to accept what science tells us about the world, while also accepting that science is fallible. In some cases, the truth of a hypothesis could make the application of Ockham’s razor much more plausible, as it did in the earlier example of the car not starting. In other cases, like the road traffic example, Ockham’s razor does not apply and so Question 4 is not relevant. Learning the truth of one hypothesis in this case does not undermine, and may well enhance, the other. Even in cases where Ockham’s razor might be applicable, the mere possibility of one hypothesis being true is not sufficient for its application. For example, in the case of the car not starting, Ockham’s razor can be applied once we have good reason to believe that there is a problem with the battery, but the mere possibility of there being such a problem would not merit application of the razor. This is relevant here because sometimes ideas that are not scientifically well established are appealed to by atheists in just this way. For example, in response to the apparent fine-tuning of natural laws as evidence for God, it is sometimes claimed that the possibility of a multiverse removes the need for God. Whatever the merits of fine-tuning arguments, it should be clear that pointing to the mere possibility of a multiverse is inadequate as a response.
The most famous specific topic in science which might be thought to explain God away is, of course, evolution and design. Given the success of Darwin’s theory of evolution by natural selection (and so a positive answer to Question 4), does it remove the need for God as an explanation of intelligent life? In response to Question 1, evolution and God are not mutually exclusive. Some people might claim that evolution requires atheism, but this far outstrips any claim that scientific evidence can establish. Besides, those like Dawkins, who use evolution to argue against God do so not on the basis of their logical incompatibility, but in terms of Ockham’s razor since evolution is claimed to remove, not disprove, God.
With regards to Question 2, there are multiple reasons that could be given by theists for thinking that evolution depends on God. Clearly, evolution requires an orderly, physical universe with appropriate scientific laws in place as well as the very fine-tuning of its physical parameters. All of these features of the universe form the basis of some of the main theistic arguments found in the contemporary literature. As with our earlier discussion of Question 2, proponents of an evolutionary case against God will reject these arguments, but given their relevance they cannot be set aside without begging the question. In other words, showing that evolution explains God away will require detailed arguments to show that the features of the world that are necessary for evolution do not also support a belief in God.
With regards to Question 3, the likelihood of intelligent life having arisen solely through unguided evolutionary processes is far from clear. This is not a criticism of the theory of natural selection, but just recognises that the evolution of intelligent life might well be dependent on some highly fortuitous events having occurred in the history of life on Earth. Richard Dawkins, for example, says of the origin of life as well as other evolutionary transitions that they might have needed to be ‘bridged by sheer luck’ (The God Delusion, p.140). Irrespective of whether, and if so how, God might have guided natural processes to bring about intelligent life, this suggests a negative answer to Question 3 and so raises a serious problem for attempts to apply Ockham’s razor in this context. This brief discussion of evolution suggests that it is very doubtful that Ockham’s razor can be applied here. And if it can’t be applied here, it seems unlikely it can be applied successfully anywhere in an attempt to explain God away.
In conclusion, we have attempted to show why using science and Ockham’s razor to explain God away is very unlikely to be successful. No doubt it is a tempting strategy for some atheists. Its appeal both to science and to a legitimate tool of scientific reasoning – the razor itself – seems to offer the possibility of a powerful and straightforward argument against theism that circumvents the need to get involved in the details of standard philosophical arguments for the existence of God. However, our analysis of the valid uses of Ockham’s razor suggests that nothing could be further from the truth.
© David H. Glass and Mark McCartney 2016
David Glass and Mark McCartney are Senior Lecturers in the School of Computing and Mathematics at Ulster University. This article is based on work that was carried out as part of a project funded by the John Templeton Foundation.