I’m constantly told that Science requires a foundation of Methodological Naturalism to function and that you cannot do science without it, while being told that this doesn’t really mean Philosophical Naturalism and so on. However I don’t think the proponents of Methodological Naturalism have every really considered what it means for science as a discipline if we take this position seriously.
One of the basic problems any scientist will face is that theory is always undetermined by data. This is more generally a problem but it certainly applied to the sciences. At heart the problem is that for any given set of data there is a very large and possibly infinite set of theories that will explain this particular set of data. So how is a scientist to whittle down the theories to find a workable set to choose from? They can use various heuristics to whittle the field down, things like Ockham’s Razor or the idea that the right theory will also be beautiful and so on.
Methodological Naturalism as an assumption can be used in this fashion as a heuristic for trimming the field somewhat. The problem with such heuristics is that they do not always work, you may exclude the right answer because it fails to match up to the rough filter you have set up. This isn’t necessarily a problem though. Given under determination you need some way to thin the field a bit and find a starting place to work from. If Methodological Naturalism is used in this fashion then it presents no problem.
However it seems some people think that Methodological Naturalism is a hard rule of science that any theory must fit within or it no longer counts as science. But this is where you will run into a problem. If it is an inviolable rule then you have artificially restricted the field of possible options. Without some definitive proof of the closed causal nature of the universe there is no way this artificial restriction is legitimate if science if still to be about the way the universe works.
There is a pair of terms, Realism and Anti-Realism that describe two basic approaches to a discipline like Science. The Realist says that science gives us genuine information about the universe, that when we hypothesis an electron there is such a thing as an electron whizzing around atoms and so on, these entities are real. This isn’t to say that science as a discipline cannot make mistakes and it is likely that many of the things we think may exist that we have discovered through scientific inquiry may turn out to be wrong or incomplete, but the task of the scientist is to discover real truths about the universe. On the other hand you have the anti-realist who says that this quest for truth is misguided and that when scientists hypothesize an electron it doesn’t really matter if an electron exists or not, it is a useful idea that makes for practical application. On Anti-realism science essentially becomes a game that seeks to explain what it can in ways that are useful, but the truth of the theories is immaterial, only their practical use counts. There might really be tiny dancing hamsters but if electrons work then they will do.
Douglas Adams illustrated this idea in a talk he gave with a concept from Feng Shui where the ideal arrangement of furniture in a room is determined by laying the room out in such a way that a Chinese dragon will have the ability to most easily move around the room. It doesn’t really matter if Chinese dragons are real, they are a useful device for laying a room out in a way that humans find appealing and comfortable. The Anti-Realist is on board with this sort of an approach to science.
Now do you see why Methodological Naturalism as a hard and fast rule entails an Anti-Realist approach to science?
If not, then let me put it together. If Methodological Naturalism is an inviolable rule then automatically any explanation that strays outside of that principle cannot be considered “scientific” and must be excluded from the sphere of scientific explanation. But what if it is true? You haven’t checked or tested it you have simply excluded it by fiat. Whatever “scientific” explanation you offer in its place will be wrong because there is a right explanation but it violates the rule you setup as a hard and fast filter on explanations. Ultimately this is the approach of the Anti-Realist not the approach of the Realist in science.
Now this isn’t strictly true, you could say science is limited as a field to only things that can in principle have explanations compatible with Methodological Naturalism, but this will make the scope of scientific investigation tiny, possibly non-existent. You can do that if you like, but don’t tell me how science can unravel all the mysteries of the universe in the next breath.
So which is it? Is science an anti-realist enterprise or is methodological naturalism just a rough heuristic?