Previously I have discussed possible reasons as to why philosophy hasn’t made as much progress as science or mathematics. Like science and math, philosophy is supposed to be uncovering certain truths, so why is philosophy taking so long in comparison? I suggested that it was because philosophy lacked a method that allowed disagreements to be resolved (science does this through experiment, and mathematics does this by checking the validity of proofs). Philosophy, being neither math nor science, cannot simply adopt the method of one of these disciplines, but perhaps we can think of something original for it. What follows then are my thoughts on what could possibly be a better way of doing philosophy.
A philosophical position can be divided into three parts, the assumptions, the argument, and the conclusions. Of course it is not always clear what these are in a given philosophy paper, but ideally we should be able to find them. Philosophy, for the most part, is about subjects that cannot be directly observed and so, like in mathematics (since formula cannot be directly observed either), the method can only guarantee the validity of the conclusions by ensuring that the argument and the assumptions are the best they can possibly be. Thus if we wish to improve the philosophical method we must focus our attention on those areas.
The argument is probably the part of the current philosophical method that needs the least amount of improvement. Ideally we would like to completely formalize philosophical arguments in some kind of logic system, and thus be able to check them as rigorously as proofs are in mathematics. And when formal logical was first invented some thought that this might happen. Of course if you follow philosophy you realize that it did not, probably for two reasons. The first is that it is hard to formalize the structure of many philosophical arguments, which are easily expressed in words. Take this very claim for example. I would have to formally state criterions that a formalization must meet in order to be used, and then state formally the properties that the logical system we have possess, including its ability to formalize arguments. And then I would have to show that the criterions for being used aren’t met. Not only is this needlessly complicated, but most of the real complexity becomes loaded into the assumption without justification (for example, the criteria for a formalization to be used). The second reason is that many of the inferences employed by philosophers in getting from their assumptions to their conclusions are not easily captured by the rules of deduction (for example: thought experiments). Thus formalization, while possible ideally, is not actually practical. What is needed then would be some system that captures the valid reasoning strategies that philosophers use “as is” (without reducing them to complex formalizations), which we can then use to examine the validity of the argument in a systematic way. In my mind I envision some kind of flowchart system, but I really have no idea as to the form such a system might take.
What is more difficult, and more important, is to ensure that the assumptions involved are good ones. Much of philosophy can be seen as proposing and defending various assumptions, and not reasoning from those assumptions to conclusions. And this is where philosophy is most likely to get bogged down, because it seems like the definition of a person should be an assumption from which we can derive conclusions, but the definition of personhood would easily the most controversial part of the argument. Natural language philosophy hoped to resolve this issue by simply declaring that the definitions were simply to be what fit the usage best. But unfortunately following such a method means that the conclusions arrived at can only be about language, and not about what is really going on, which I hope most philosophers would agree is unacceptable. I tentatively propose that a better idea is to say that when we define a term, such as “mood”, that the best definition is the one that describes most accurately the kinds of things that motivate us to use the word, or that we are attempting to describe through our usage of the word. This means that sometimes our assumptions may turn out to be incorrect upon further study, a situation analogous to scientific progress. But this proposal is still not fully developed. One possible problem arises when dealing with categories that may include things that we have never encountered before. For example, under this proposal we might be led to define person strictly as “human being”. This might not be a real problem, since we might be able to replace words like “person” in philosophy, words used in a somewhat unorthodox manner, with other constructions, such as “intelligent being” or “possible member of society”. But it is possible that such a solution may not be entirely satisfactory.
If philosophy could somehow be reduced to math or science our task would be easy, but such a reduction is impossible, or at least it should be for genuine philosophy. So I will leave the task unfinished for tonight, but it is something that I will return to.