A rough-and-ready guide to what the deal is with philosophy, sort of

I agree that a lot of people have misconceptions of what philosophy does, or should do, or what its methodology is. This isn't helped by some philosophers who are arrogant (although this problem can be found all across academia and beyond), or write obtusely (especially in the "continental tradition" and "critical theory"). But a lot of the problem comes from the way in which philosophy is apparently either (1) unreasonably difficult or (2) remote from a potential student's actual life. These latter problems probably arise because philosophy, the real stuff, is difficult, and serious study is required to grasp the subtleties involved---but then electrical engineering is also difficult and requires intense study, and I don't hear anyone complaining about that---and because you have to care about something other than half-baked sophistry when grasping that what matters to your life is more than prescriptions for behavior in everyday situations.

So for instance meta-ethics is important because its subject matter is the foundations of ethics, and we care about ethics, so we should care about meta-ethics (a formalizable ethical theory would have to begin from axioms the existence and discovery of which is part of the meta-ethical subject matter). But not everyone has the time or interest to study meta-ethics; this doesn't make it not worthwhile. Again, this stuff is hard; you don't see people demanding that electrical engineers be required to provide brief and---because of their condensed size---inaccurate tidbits of useful electrical engineering knowledge.

The first point is that there are topics in philosophy that are valuable whether or not it's to your taste to study them.

Humans aren't by nature particularly good at deductive logic; nor are they quick with any sort of complicated statistical reasoning (Bayesian induction with lots of priors made explicit, say). Doing good philosophy requires learning how to understand and make good arguments, and that by itself takes a good deal of time. The fact that good arguments are hard doesn't take away from their effectiveness except in a practical sense. The subtlety and complexity of most good (philosophical) arguments makes them appear to people not equipped to handle them, that is, most people, obtuse, but that does not mean that the arguments aren't cogent. Nor does the difficulty imply that philosophers are doing something wrong; the reason that deep questions don't have easy answers is that the questions are hard and require a lot of work to even understand properly. To think otherwise is simply to demonstrate superficial knowledge of the subject.

The second point is that difficulty in doing philosophy does not in any way make it less than a legitimate research paradigm.

As to what philosophy should do, as a third point, all I'll say is that at the most general level, philosophy attempts to understand and answer questions that can't be answered by the special sciences.

That relationship is interesting. Historically, all branches of inquiry fall under philosophy until an area of investigation is singled out for investigation by empirical means ---philosophy is, broadly, not different from the sciences in that it seeks answers to questions, or truths. (IIRC, Kant was the "Queen of the sciences" guy, and he said it of philosophy, not mathematics.) More recently, probably due in part to positivism, philosophy has come to be seen to be divorced from the physical sciences, but this was not always the case (see for instance that astrophysicist Malcolm Longair is a chair of "Natural Philosophy" at Cambridge). Aristotle did physics and biology as well as ethics and metaphysics; after the Renaissance the "scientific method" was more or less introduced and a priori inquiry was suddenly less relevant to physics, chemistry, and so forth. But note that for example thought experiments are still a methodology philosophy employs in common with these fields. Similarly, psychology began in the 19th century to diverge from philosophy (James might be the last point of genuine convergence). Despite these separations, philosophy is still relevant to understanding the underpinnings of the physical-scientific endeavor; example problems are with respect to physics the problem of induction and the nature of natural laws, and regarding biology the study of natural kind concepts. The latter are interesting because it shows there are topics that can yield truths that the special sciences can't themselves answer. I think these considerations indicate that philosophy in general is not sui generis but is better regarded as one of the sciences, if a peculiar one in some regards.

But because the questions remaining to philosophy are often of an a priori nature, it may seem that the subject matter of philosophy is different from that of the other sciences. I posit that the strongest reading of such a claim is obviously false, since mathematics and logic overlap with philosophy in such a way as to make any distinction among them arbitrary. MPhil has suggested that the subject matter of philosophy is concepts, or perhaps more broadly language and concepts, exclusively. I've suggested to him that he read Timothy Williamson's The Philosophy of Philosophy, and I recommend it to anyone else interested, although you should be aware that it's directed to an audience of philosophers and addresses them as tending to isolate themselves; the introduction and first chapter are a fair taste of what he's up to, if you haven't the time or inclination to slog through the whole of the book. It seems to me fairly straightforward, however, that if philosophy of science is relevant to our understanding of how "the world" is, and if our epistemology is relevant to what we can know about the world, and most especially if our ethical theories are supposed to ultimately issue in prescriptions for action in the world, then the subject matter isn't clearly bounded by what are "the conceptual questions" or to "relations between concepts" or to metaphysical questions, or ones about "analytic truths", just as the special sciences are not clearly bounded but blend into one another, even if there are some (many) questions that clearly belong to one or another of them.

So a fourth point is that it's not clear to me why I should regard a science/philosophy dichotomy as legitimate for any but purposes of crude approximation.

Finally, the purpose of what looks like verbal gymnastics is for the most part either (in good philosophy) required for accuracy, or (in poor philosophy) just as annoying to philosophers because it hides poor arguments. Telling the difference takes work (I think MPhil mentioned the example of P. Churchland's demonstration that the (original) Mary case involves equivocation). "What do you mean by ___ ?" is a perfectly legitimate method for clarifying what's at issue; surely no one advocates beginning a discussion the topic of which one doesn't understand.

[Context: crossposted over here (mostly).]