One question that frequently confronts the New Atheists (especially those with a science background) is whether a religion and science are incompatible. The stock answer is that many religious leaders accept science as a good way to understand the natural world and conversely, many scientists have a religious faith (Ken Miller and Francis Collins come to mind). In a previous blog post I talked about how sociological research had revealed that about half of American scientists are able to both perform cutting-edge science and maintain a religious identity. An even larger proportion is still interested in matters of spirituality despite daily engaging in rational, empirical inquiry.

These facts show there is, at least, a kind of ‘brute compatibility’ between science and religion; a single person can hold both ideas simultaneously. However, the obvious counter to ‘brute compatibility’ is to point out that in certain cases the findings of science conflict with specific religious claims about the nature of the world. For example, if you claim that the world is 6,000 years old, science says you are wrong. According to empirical data, the world is more like 4.5 billion years old and anyone who says the scientific evidence shows otherwise is simply mistaken. Because science can only conflict with specifically defined religious claims, I call this ‘specific incompatibility’. Although this type of incompatibility is important, and probably accounts for a large proportion of science’s moderating impact on religion, it does not completely contradict all types of religious claims. Again, this answer is too superficial; the original question is asking something more fundamental – are religion and science incompatible at the deeper, philosophical level?

Here we must start with a rigorous understanding of the exact nature of science, its epistemological limits, and the assumptions it makes. First science assumes that the universe exists and is, broadly speaking, observer independent. This assumption avoids the problem of solipsism, where I could construct an argument based on the idea that the universe is simply a figment of my (hyperactive) imagination. The second assumption brings in causality, scientists must assume that causes and effects are empirical (observable and measurable) and, more importantly, natural. That these causes must be empirical is fairly self-explanatory. If we cannot observe and measure we cannot draw any inferences, offer any explanations, or say anything at all about them! Basically, we would not be able to do science. Less clear might be why science can only approach natural causes rather than supernatural ones.

Let’s use an example to help highlight the problem of supernatural causes. A nice set of evidence for evolution is that of the fossil record. Without going into too much detail, the arrangement and transitions of different fossil types is empirical evidence for evolution occurring in the real world (the universe out there). However, along comes a supernaturalist, and he to us “yes, I agree with your observations and measurements of the fossil evidence, however God (or other supernatural cause) made it so the fossil record appeared to support evolution but in reality the theory is wrong.’ This is a serious and unsolvable problem for science and it can be applied to any other explanation or conclusion derived from methods based on the two assumptions above. If supernatural causes occur, the best of verifiable, empirical science will frequently give the wrong answer. If God (et al) always makes the observed and measured evidence look like it is supporting the wrong conclusion no appeal to empiricism can save a scientific theory – by definition. This is why science must reject supernatural causation and become methodologically naturalist at the outset. Without this assumption we would not be able to do science.

Methodological naturalism, therefore, means that science cannot ever make a judgment on supernatural claims. Science assumes that supernatural causes don’t exist and gets on with its job of figuring out the observable universe. As religion, for the large part, is based on such supernatural claims (God caused the universe, Karma causes ill luck, Boobs cause earthquakes) science has very little to do with the majority of religious assertions. I call this ‘philosophical compatibility’, as an understanding of the philosophy of science shows that science and religion are not in conflict. I accept that in specific cases religious claims may contradict with the findings of science and in those cases religion is wrong, but we can always take the step back to the philosophical level and show that such ‘contradictions’ may not be problematic if we allow for supernatural causes.

There’s one last level I want to discuss and that is ‘metaphysical incompatibility’. Working from science and its assumption of methodological naturalism one might take the eminently reasonable position of philosophical naturalism – supernatural forces positively do not exist in the real world. Note that this is not a scientific position but a metaphysical one, albeit one informed by scientific understanding. Science is insufficient to get us to philosophical naturalism, the move also requires reason and logical arguments (examples would include the argument from suffering of which I am fond). Philosophical naturalism is in clear contradiction with a metaphysic infused with religious supernaturalism. There is also a secondary incompatibility at this level but Feynman said it best so I’m going to turn the last word over to him:

“As a matter of fact, the conflict is doubly difficult in this metaphysical region. Firstly, the facts may be in conflict, but even if the facts were not in conflict, the attitude is different. The spirit of uncertainty in science is an attitude toward the metaphysical questions that is quite different from the certainty and faith that is demanded in religion. There is definitely a conflict, I believe – both in fact and in spirit – over the metaphysical aspects of religion.’

And that is the heart of the science/religion incompatibility in a nutshell.