Recently I have been doing a great deal of metaphilosophy, philosophical reasoning about the foundations and method of philosophy itself. This is simply a manifestation of the intellectual path that all serious philosophers follow at some point. When everyone first comes to philosophy the problems of philosophy seem relatively easy to solve; most intelligent people, having encountered the problems in other contexts (usually scientific), and already have general ideas as to what the solutions to the problems are, ideas that all hang together and justify each other. As such a person answering philosophical questions is easy, because to defend any claim it is simply necessary to appeal to some of your other beliefs, which are often only a step or two away from justifying the claim. But this isn’t really doing philosophy, it is simply explaining what you believe. To do philosophy requires the ability to set aside some of your pre-existing beliefs about the answers to philosophical questions, probably a good many of them, and instead attempt to reach an answer grounded only in an appeal to more generally acceptable (or at least “simpler”) beliefs. To create more robust arguments more beliefs are set aside, even beliefs that may be nearly universally shared, in order to demonstrate that the claim is true, not just implied by the consensus. Obviously this can be extended to more and more beliefs until we are left with what may appear to be no beliefs. In such a position we are like Descartes, trying to construct claims that simply must be true. But there is no reason to stop there. We can throw out even more presuppositions by setting aside our customary methods of investigation, such as intuition and deduction. In such a state we are left with nothing and thus cannot establish any claims.
Most people never take the process so far, but they lack compelling reasons for not doing so. Certainly they can’t know at any point that the principles that remain cannot be wrong or shouldn’t be questioned, because the claim that they don’t need to be set aside must follow from some principles (or be unjustified). And clearly no principles can legitimately justify themselves, otherwise we would have to accept every principle that included a clause claiming itself to be correct, and that is absurd. Stopping for practical reasons is also questionable. It may seem that at some point it no longer serves any purpose to set aside our presuppositions, as they may be so widely accepted that we never encounter problems in convincingly establishing our claims on the basis of them. However, the point of setting aside our beliefs was not to become better arguers (I would hope), it was to eliminate the possibility that we were in error and that our errors were blinding us from seeing that error. And the consensus is no guarantee of correctness. Nor is there a guarantee that someone won’t disagree with you, and may even change their mind on these fundamental (and possibly methodological) issues because they are committed to denying the conclusion, and may argue convincingly enough that what was thought to be a solid foundation is now called into doubt. And so if you wish to universally establish your claims you cannot take anything for granted.
One of the tasks of metaphilosophy is to solve this problem. A successful metaphilosophical theory describes the methods by which a philosophical investigation can proceed, as well as any acceptable principles that can be taken for granted. This stops the infinite and insoluble regress of tossing out all of our beliefs one by one by side stepping it completely; the metaphilosophical theory provides the foundation which every philosopher is supposed to accept, and thus allows us to make arguments for philosophical claims that, if made without error, are the closest we can get to the truth, and must be accepted by every philosopher. Of course the metaphilosophical theory itself cannot be extracted from thin air, and how it can be established might very well be considered the primary problem of metaphilosophy.
But not everyone accepts the validity of metaphilosophy. To me however the value of metaphilosophy seems obvious, even if we are unable to produce a metaphilosophical theory (and even if producing one is impossible in principle). Simply consider how the systematic study of the goals and methods of other disciplines helped them improve themselves, to define themselves more clearly, and to better distinguish good work from bad. Why should philosophy be the sole exception? Obviously there is some difficulty involved in a field studying itself, but that doesn’t make the problem impossible to solve, just trickier. However, the most powerful objections to metaphilosophy are simply philosophical mindsets that avoid the need for any metaphilosophical theory. To conclude I will discuss the three most popular ways of avoiding the problem and why they fail. The first way to avoid the problem is to deny that it is philosophy’s role to produce true philosophical theories or claims. Instead, it is said, we are simply to illuminate the connections of ideas, and show that if certain claims are accepted then other claims follow. This fails in two ways. First it leaves the principles of inference, those by which we determine how one claim is connected to another, without justification. Secondly, it makes philosophy a pointless enterprise. If some of the claims in the network of relationships uncovered can be established by some other discipline then there is no need for philosophy and we should just be engaging in that other discipline. But if the claims cannot be established by some other discipline then there is simply no point in detailing how some of them imply the others. Another way to avoid the need for a metaphilosophical theory is to produce intentionally obscure philosophy, philosophy that can be interpreted in a large number of ways (instead of just one). This avoids part of the problem, by producing claims that are acceptable to everyone simply because they are free to interpret them in ways that agree with their existing beliefs. But I cannot stomach this solution because I have some degree of intellectual honesty, and this solution amounts to pretending to produce philosophical claims while really doing nothing of the sort. And finally we come to the third solution, which is simply to deny that there are right and wrong answers in philosophy, but that both sides are right and that the “Truth” (with a capital T) results from a synthesis of both sides. Again, this approach seem unpalatable to me because it involves a level of intellectual dishonesty, on two levels. One level of intellectual dishonesty springs from the fact that while many might espouse this theory few actually live by it. Synthesis is fine and good until we come to the positions that those who endorse it care about, at which point suddenly one position on the issue is correct and the other wrong. The other, more serious, kind of intellectual dishonesty comes from the fact that the theory is not applied to itself. If we were being honest we would synthesize the theory itself with its denial, and that new theory with its denial, and so on. Whatever that supertask yields would be the appropriate philosophical paradigm to operate under. Since they can’t adopt that position proponents of synthesis should accept that other approaches, those that deny synthesis, have equal validity, at which point any outside observer will conclude that by endorsing its opposite synthesis has denied itself.