While doing metaphilosophy, one can come across some rather hostile attitudes. This is particularly true if one dares to consider such questions as how is, can, or even should philosophy be practiced. Some say that such methodological considerations are a non-starter since philosophy, by its very nature, should not have any constraints. Philosophy is about bringing to light unarticulated presuppositions and then transcending them. It is the ultimate “meta” discipline, and attempts to curtail it will just end up severing it.
Such attitudes are understandable, yet they miss something important. It is true that taking any particular methodological restriction as given will curtail philosophy, but having some methodology at all does not. This is because you cannot properly do philosophy without doing it somehow. Philosophising is an activity after all.
To be precise, you can philosophise without a methodology, but it will quickly turn into something trivial. If there are no methodological restrictions, then all ways of justifying any claims are allowed. This would mean that we could prove any claim in any way we desire. Thus, a single utterance of, say, “plii”, is enough to prove any claim. But that same “plii” is also enough to disprove the same claim, which means that without methodological restrictions, you cannot have any practical difference between true and false claims. Truth becomes arbitrary and, hence, also the idea of a proof itself. This way we end up with everything and nothing at the same time. This is why having no methodological restrictions at all is to walk in a barren wasteland of arbitrariness.
There is no escape from having some methodology, but this does not mean that philosophy should adopt, say, the methods commonly used in chemistry or ornithology. Avoiding arbitrariness leaves a lot of methodological room and it is the job of metaphilosophy or metamethdology to work out exactly how much and at what price.
Just to give a short example, consider the problem of heaps showcased by Ilkka in a previous blog post. The predicate “heap” seemed to be vague and this leads to potentially metaphysical problems. From a methodological point of view, however, we can note that the initial problem comes with a strong presupposition. It takes a notion that is used for large quantities of objects (“heaps”) and then analyses it on a level of individual objects (grains). It takes two notions from two different levels of description and then combines them into a single context. The result is that the notion of “heap” becomes vague.
From a practical point of view, this is totally expected and unproblematic. In a context of having heaps of things, the notion of a heap is not vague. The distinction between different heaps or some other objects of similar size is often not arbitrary. For the notion of a “heap” to be usable for those contexts, it does not need to be well defined for all possible contexts. Similarly, the notion of a “grain” does not need to be well-defined at the level of description of “heaps”. Indeed, when we are dealing with heaps, the number of grains they have, is not something we can often evaluate, and is thus vague.
From the perspective of methodology, then, the problem of heaps comes from mixing two different levels of description. Noting this allows us to recognise something non-trivial. By accepting a mixing of contexts of the aforementioned kind, one is effectively assuming that concepts are well-defined in all contexts and thus that there is or should be some sort of a universal context. Thus, if anything, the argument from vagueness is an argument against certain varieties of metaphysical realism and not a problem otherwise. So, Ilkka is right in claiming that we do not need metaphysics to deal with it, even though it can show us something about certain metaphysical views.
By paying attention to how different ways of analysis can lead to problems, what sort of restrictions they come with, and how analysis is always tied to some practical activity of analysing, we can arrive at similar results and deepen our understanding of philosophy as a whole.
Methodological scientism is an example of such an activity. It tries to enhance philosophical inquiry that has some epistemic goals. What methods work for achieving these goals? Why? Why not? Methodological scientism draws from known cases of epistemic success, namely science, while also taking into consideration the domain specific nature of methods. This is why the focus is on general methodological principles such as evaluability or definability. Methodological scientism has the advantage of being able to draw from the success of science and from the rich methodological expertise scientists have. It is a good place to start one’s metaphilosophical journey.
While any specific form of methodological scientism can be opposed, rejecting the approach altogether is just to embrace wilful ignorance. It is to assume that nothing methodologically useful can be learned from the success of science before even making the attempt.
Studying the limits of analysis is a fruitful field of philosophical research and should not be shunned merely because it can potentially be truly fruitful.