What is philosophy, again?

by Dave Maier

There has been some interesting recent discussion, both here and elsewhere, about what philosophy is and should be. Here are two shiny pennies from my own purse.

LeiterIn the introduction to his recent anthology The Future for Philosophy, Brian Leiter laments that “[p]hilosophy, perhaps more than any other discipline, has been plagued by debates about what the discipline is or ought to be.” This is strong language. To call such debate a “plague” is not simply to regret its occurrence as a necessary evil, but to see it as an alien force, infecting the host body from outside. On this picture, to debate the nature and ends of philosophy is akin to putting the car up on the rack; when this is happening, no progress is being made. If we have to spend all our time figuring out what philosophy is, we'll never get around to actually doing any of it.

Maybe we should simply shrug the question off. After all, no matter what you said philosophy was, one could always respond “okay, so what these other people do isn't 'philosophy' by your definition; but it's still worthwhile – maybe even more so than what you do under that name.” Coming up with a new name for what we have been calling “philosophy” seems even less pressing. Who cares what something is called, when what is important is whether and how to do it?

However (you knew this was coming), I think these debates can be quite enlightening – if you know what to look for. In any case that is our subject today.

1: Ontic science vs. the linguistic turn

So what is philosophy then? One common answer, usually just assumed but occasionally spelled out, is that philosophers try to discover the basic features of reality, just like science does. However, while physical science determines the nature of observable objects and processes, and thus contingent matters of fact, philosophy concerns itself instead with matters of metaphysical necessity, inquiring into the ultimate entities and structures underlying the world as we encounter it. This conception of philosophy is what Colin McGinn endorses in renaming it “ontic science”: non-empirical inquiry into the real.

On this view, the end result of our inquiry – as it must be if it is to be inquiry at all – is truth (specifically, true doctrines or theories); and the proper method in reaching it is precise, rigorous argument from universally accepted premises to an unambiguous, substantive conclusion. This is naturally easier said than done; and it is no secret that the list of universally accepted philosophical doctrines is – well, let's just say it's not long enough really to count as a “list”. However, I don't think that the lack of universally accepted doctrine shows all by itself that “ontic science”, which we might also call metaphilosophical “dogmatism”, is all wrong; a perfectly good response by my lights would be “well, after all, we Westerners have only been at this for 2500 years – what did you expect?”

If we can't agree, though, then maybe we're doing it wrong; and a natural place to look for the problem is in our instruments: our minds, and in particular, our language. Might these things be systematically distorting our view of reality? This question, along with important formal developments in logic and semantics (I condense and oversimplify here), led at last, in the mid-twentieth century, to what has become known as the “linguistic turn” in analytic philosophy (continental philosophy having broken off some time before).

Rorty linguistic turnFollowing a line of thought which may or may not be found in Wittgenstein, philosophers like Gilbert Ryle and P. F. Strawson, and more recently Peter Hacker, have reconstrued the task of philosophy as one of examining not the world directly – a fantasy by their lights – but instead the conceptual scheme we use to talk about it. Importantly, we do this not to improve our concepts, so that we may more accurately portray the objective world, but instead to understand them, and more to the point, to understand what it makes sense to say, and what it does not make sense to say (or seem to say). If we are asking questions which, given the actual meaning of the concepts we use to formulate them, turn out to be of the same form as “when are the round squares?” or even “is the number of stars prime?”, then rather than continuing to try to answer them – and worrying about the (natural!) lack of consensus – we should give them up as meaningless.

So far so good, surely; no one wants to bang their head against the wall trying to square the circle. Indeed this looks like the opening moves in a familiar attack on metaphysical “inquiry” as fatally non-empirical and thus (ultimately) meaningless; but the upshot is importantly different. For “Hackensteinians,” the importance of Wittgenstein's Philosophical Investigations was to show that philosophical questions are questions of meaning as opposed to matters of fact. It's easy to conflate this view with its logical-positivist cousin, but Hacker's view is more subtle. It's true that traditional metaphysics is “meaningless” when construed as a demand for philosophical doctrine on the model of science (“ontic science”), and in this sense our conceptual clarifications may indeed result in giving up this demand. However, the task is not simply negative; for what we will find, on Hacker's view, is that in gaining “a clear view of the use of our words” (PI §122), and seeing our dogmatic task as misguided, we gain the very understanding we had confusedly thought we needed a doctrine to achieve. That's why “linguistic philosophy” isn't just one big argument for the claim that “metaphysics is meaningless so give it up.” It's an endless series of careful case studies – Wittgenstein's image is that of “therapy,” which has made it seem like he thinks metaphysicians are just nutcases – aimed at unpacking the complexities of our conceptual scheme in just the right way, so as to soothe the platonistic metaphysical urge by providing a satisfying yet at the same time non-illusory substitute.

Or that's the idea anyway. Linguistic philosophy was very big for a while, in Britain at least, but in this country the post-positivist resistance to metaphysics has been more naturalistic (e.g. Quinean) than anything else; and there has been a (perhaps inevitable) renaissance in analytic metaphysics too, in the work e.g. of Saul Kripke and David Lewis. So we shouldn't be surprised when the “linguistic turn” is itself subjected to critical attack by its main targets: ontic scientists.

WilliamsonTimothy Williamson is a big name in contemporary, well, ontic science; and in his 2008 book The Philosophy of Philosophy he has some strong words for fans of the linguistic turn. As you can imagine, Williamson’s main gripe is that instead of talking about the world, linguistic turners talk about language – that is, about how we talk about the world – leaving the world itself behind. But it’s the world that we’re really interested in. That’s why we were talking about language in the first place: because we thought that we might have been taking features of language (our own contributions to thought, i.e., its subjective part) for features of the world (the world’s contribution to thought, i.e. its objective part). The idea was to fix things up by speaking more clearly, so that we wouldn’t do this – that when we talked about the world we would use concepts that picked out only objective phenomena, so that when we used them in inquiry, we would (when inquiry was successful) state only objective truths (that is, truths). But we seem to have forgotten that the point was to get back to doing inquiry into the world. Instead, we’ve gotten stuck talking about language.

This is a common song nowadays, sung with varying degrees of appropriateness about hermeneutics, deconstruction, and every flavor of literary theory under the sun, and its appeal is undeniable if you have been waiting to hear someone come out and sing it with gusto. Williamson does not disappoint his fans. He takes full rhetorical advantage of the apparent middle path between the empiricist rejection of “armchair philosophy” on the one hand, and on the other, the single-minded focus on linguistic rules characteristic of the linguistic turners. The details of his argument are beyond our scope here, as this is not a review of his book; but its rough outlines should be clear enough. In an interview he stakes out the traditional “commonsense” understanding of metaphysics, which one may interpret, depending on one's commitments, as either reiterating what we should have been saying all along, or as simply doubling down: “For example, philosophers of time are interested in the underlying nature of time, not just the word ‘time’ or our concept of time. […] As for the principles that we implicitly accept simply in understanding words or grasping concepts, I argue that there aren’t any.” Well, okay then.

My point in mentioning Williamson in particular – he is of course not the only philosopher to defend the traditional picture against these foes – is because his defense of philosophy takes the particular form it does: a self-consciously philosophical one.

2: The philosophy of philosophy

It should be pretty obvious that the question of the nature of philosophy is a philosophical question. Who else is going to be able to answer it if not philosophers themselves? This very fact, however, is more significant than people think.

Consider the phrase “philosophy of [X]”, as in philosophy of science, or philosophy of mind or language. Surely if this denotes anything, it's a type of inquiry (a philosophical one) into a particular phenomenon or aspect of reality. But types of inquiry, as philosophy is, are themselves phenomena or aspects of reality; so naturally we can put “philosophy” in for X and get “the philosophy of philosophy.” But here's an interesting difference: that last phrase isn't simply the “philosophy of [X]” schema with “philosophy” in for X; it's an “[X] of [X]” schema with “philosophy” put into both spaces.

This gives it a rather different form, and puts some unexpected constraints on what we could possibly be doing when we do this. In particular, it requires us to recognize the type of inquiry as on a par with – not on the opposite side of a methodological gulf from – its object. At a minimum, it requires us to be consistent: we cannot appeal to different principles or doctrines in determining the ultimate nature of our object as we do in describing the object we are investigating. Yet we cannot lose sight of the fact that our [X] is playing an importantly different role in the two spaces of the schema. We are looking at the same thing in relation to itself, and thus in two different aspects simultaneously. Ideally, our inquiry should reflect this.

How does this work for the philosophers we have been discussing? Not surprisingly, but a bit disappointingly now that it comes up, it turns out that the nature of philosophy, for both Hacker and Williamson, is pretty much straightforwardly determined by that philosopher's own previous substantive commitments: as if simply to describe the object under the magnifying glass were ipso facto to determine the proper nature of the inquiry into that object itself, and thus into everything else under that same glass. For example, Hacker's ground-level semantic theory of linguistic rules turns out to be pretty much indistinguishable from his conclusions about the nature of philosophy as inquiry (or, as he would rather say, as conceptual clarification). But when he tells us that, what is he doing? What is the status of his own semantic theory? Is it really just a Wittgensteinian “reminder” of how we use our words? If so, why are Hackensteinians the only people on the planet who talk this way? Naturally consistency is a virtue, all things being equal; but I was really hoping for something more enlightening from an inquiry of this provocatively self-referential form.

Similarly, Williamson's conception of philosophical inquiry is scarcely distinct from his “classical realist” metaphysics: “the world is largely independent of us”, he claims, and in describing it philosophically, we are simply doing what we always do in inquiry: saying how it really is by determining an ideally accurate linguistic representation, one from which all subjective contamination has been carefully expunged (what I like to call the “bulletin-board” conception of objectivity: here it is, take it or leave it, no matter who you are or why you care). This, again, is a substantive (and controversial!) philosophical commitment, and it seems that the nature of philosophy as a whole is not particularly helpfully identified with what seems to be a mere application to itself of the settled doctrines of one of its subfields. At the very least, we might expect the two to be held far apart enough, long enough, for a new conception of each relatum to, as they say, settle into place.

Still, this is not a refutation of either philosopher. Maybe the answer to the question of philosophy's nature really is that boring, and one or the other philosopher has things (whether that be the world, for the one, or the actual rules of our language, for the other) exactly right. I'm running out of space here, but let me just sketch the barest outline of an alternative.

Naturally I have my own philosophical commitments which I bring to the table, and my point is not that we should give them up when we consider what philosophy is. Instead, we should use this inquiry to align them so that they reinforce and complement each other in the most mutually beneficial way. For example, I think the metaphysical/conceptual dualism of subject and object, as manifested both in Williamson's realism and Hacker's linguistic therapy, has outlived its usefulness. This if anything is a metaphilosophical claim (as its pragmatic form suggests); but it is manifested, in the lower-level philosophical commitments it governs, in ways which, I hope, help make sense of that pragmatic attitude itself. For instance, it tells us not only to avoid dualistic doctrines (of which Cartesian mind-body substance dualism is only the most easily dismissed), but also to carry out that disposal in a characteristically non-dualistic way. Again (and I must apologize for this all-too-brief treatment; I hope we can revisit the topic some other time), the idea is not to allow the one to dictate the other, but instead to align them so that they may helpfully interact.

One last quick example: Davidson has shown us how to align the concepts of belief and meaning so that we can see semantic interpretation and inquiry into the world as two sides of the same coin. That fulfills the content of our metaphilosophical desideratum that subject/object (here, belief/meaning) dualisms be overcome; but it also refers back in its form to that principle itself: it is by looking to overcome dualisms in this way that we see the philosophy/metaphilosophy dualism as another instance of that same thing. By reconstruing the two as not identical, but not different either, we see how the lower-level process can be generalized and made to apply to itself as well. The view is then just as self-consistent as either Hacker's or Williamson's, while actually allowing the two spaces in our schema to influence each other dynamically instead of simply mirroring them faithfully at another level. It is the dualism of level itself that the “philosophy of philosophy” allows us to overcome.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone