There is much debate over the nature of philosophical theories. Are they like scientific theories? Is so, to what extent? Wittgenstein thought that no realm of phenomena is the special business of the philosopher. Paul Horwich in his article "Was Wittgenstein Right?" published in The Stone Reader, says that for Wittgenstein there are no phenomena about which a philosopher "should devise profound a priori theories and sophisticated supporting arguments." Further "There are no startling discoveries to be made of facts, not open to the methods of science, yet accessible 'from the armchair' through some blend of intuition, pure reason and conceptual analysis."
So this is what I think. Philosophical theories and arguments are stories which are more or less illuminating depending on the context. They are a priori only in the loose sense that they are not tethered close to sense data or scientific experiment. But they are based on experience, and there is nothing to keep philosophers from developing their theories in a way that is at least consistent with contemporary science. Philosophical theories are based on philosophical dialogue (whether implicit or explicit.) Sure, philosophy tends to be done in armchairs, although that's true for a lot of science too. And philosophers are not horribly limited by that since their armchairs are generally in front of computers connected to the web. And usually those armchairs are in well-stocked libraries.
So the question remains whether there are startling discoveries that philosophers can make. Probably not, or not often. However there are moments of inspiration. There are discoveries. I like to think of these as sudden revelations of essences, although, in an unorthodox way, since I see essences as evolving through history. It is more like "I now see how to say something interesting about art that is relevant to our own times" rather than "I now see art's eternal unchanging essence."
Horwich describes Wittgenstein's attitude as "in stark opposition to the traditional view, which continues to prevail. Philosophy is respected...for its promise to provide fundamental insights into the human condition and the ultimate character of the universe, leading to vital conclusions about how we are to arrange our lives." I partly agree with the traditional view and partly do not. Yes, Philosophy should be respected for its promise to provide fundamental insights. However I doubt there is just one human condition, or that there is just one ultimate character of the universe. Philosophy can just reveal essential things about the world as we experience it, and that world changes. Doing this, revealing essences, however, is enough to lead to vital conclusions about how we should arrange our lives.
Wittgenstein thinks we are bound to be disappointed if we take the traditional view, and I agree. But Horwich goes on to say that the perennial controversy of philosophy is an embarrassing failure. I would say, taking more of a Hegelian line, that it is not embarrassing at all...something more like a wondrous success story. Each successful philosophical project, i.e. the ones that ultimately make it into the history books, is both insight and invention.... a further development in the ongoing evolution of essences.
Horwich/Wittgenstein (for now I treat them as the same) believe that traditional theorizing must be replaced by "painstaking identification of its tempting bu misguided presuppositions and an understanding of how we ever came to regard them as legitimate." This is a fine thing to do. But this is part of the dialectic, and there are other parts to the dialectic. It is the process of destroying the thesis. But dialectical process, and hence Philosophy, is not complete until it provides an antithesis and a synthesis.
So Wittgenstein thinks that philosophical investigation should only destroy everything interesting. I think it does this, but only to make new interesting things. He thinks the point of Philosophy is "clearing up the ground of language on which they [the so-called interesting things] stand." I think that this is just one sub-project in philosophy.
So, unlike Horwich and Wittgenstein, who are extreme pessimists about Philosophy, I am an optimist. (Although I must say that I am a pessimist about a lot of other things.) The difference between me and Wittgenstein here is that he is a disappointed absolutist. I never was an absolutist and so, hopefully, avoid the illusions of being a disappointed absolutist.
So I have no problem with the primary goals of traditional philosophy, which Horwich describes as "to arrive at simple, general principles, to uncover profound explanations, and to correct naive opinions..." I just disagree with anyone who wants to see these things as much like what scientists do under the same names.
The place where I most agree with Horwich is when he says: "our concepts exhibit a highly theory-resistant complexity and variability. They evolved, not for the sake of science and its objectives, but rather in order to cater to the interacting contingencies of our nature, our culture, our environment, our communicative needs and our other purposes." Hear hear! Except that I disagree with "theory-resistant." This evolution is theory-dependent, or better, it just is the development of theory. I also agree that "the commitments defining individual concepts are rarely simple or determinate and differ dramatically from one concept to another." The "general principles" we look for here are very unlike the ones we look for in science: they are best seen as philosophical definitions offered as part of the ongoing dialectic.
But it is fundamental to Wittgenstein to avoid theory-construction and to be merely therapeutic, and this will be an exposing of irrational assumptions on which theory is based. I just do not see why this is an either/or situation. Theory-construction IS story-telling that is therapeutic. To expose some assumption of an opposing theory is just what one does in dialectic. Be prepared for that to happen to you too, later down the line.
Horwich then goes on to apply this thinking to the evolving idea of "truth." His point, as always with the Wittgensteinians, is that the entire history of the concept is a history of failures. But no theory ever really fails, it just recedes, or is sublated. It goes underground, for awhile. It can be reinterpreted and enter the fray again. True, "truth" is not like an empirical concept such as "red" or "magnetic." It is, to follow Gallie's terminology, an essentially contested concept. So it doesn't quite stand for a property: it stands for an object of philosophical contest, an essence. But this means that Horwich is profoundly wrong when he concludes that "Truth emerges as exceptionally unprofound and exceptionally unmysterious" because he accepts the disquotational theory of truth, i.e. ""It's true that E = mc2" is equivalent to E=mc2."" This is just ignoring the problem of truth. It is a deliberate avoiding of Philosophy, and I think there is something inauthentic going on here.
No, all of the essentially contested concepts are profound, all are mysterious. Diving deep into the nature of anything is diving deep into a nested collection of essences and thus, for example, into an entire culture. What he calls a "wild goose chase" is really the point...except it is never just one goose.
Of course belief in the disquotational theory of truth is belief that there is one final definition of truth, one that, it turns out, is marvelously unmarvelous. This is the path of a disappointed absolutist who is still, really, an absolutist at heart.