At a workshop on Wittgenstein’s methodology which I was at recently, Marie McGinn made a point of underlining the ethical stakes of much philosophical work undertaken in a Wittgensteinian spirit. I won’t try and rehearse exactly what she said here, and will instead examine the topic of naturalism which she raised in this context, but some of what I will say, ultimately, I take to be deeply sympathetic to her view (but be that upon my own head and not hers).
The question of ethics came up here in light of Wittgenstein’s remarks about philosophical problems arising when ‘language goes on holiday’ or is like ‘an engine idling’ rather than doing real work. For if we take philosophical problems to share this form — however diffuse their manifestation and origins — then it seems we are led to a conception of philosophy as a therapeutic set of practices which simply “bring words back from their metaphysical to their everyday use.” (PI 116)
This makes philosophy look like a purely negative activity, and there are certainly places where Wittgenstein appears to embrace this idea. Take PI 118-9, for example:
Where does our investigation get its importance from, since it seems only to destroy everything interesting, that is, all that is great and important? (As it were all the buildings, leaving only bits of stone and rubble.) What we are destroying is nothing but houses of cards and we are clearing up the ground of language on which they stand.
The results of philosophy are the uncovering of one or other piece of plain nonsense and of other bumps that the understanding has got by running its head up against the limits of language. These bumps make us see the value of inquiry.
So conceived, philosophy returns us to the everyday. We get back to baseline. This is not something just to be sniffed at since the temptations to fall into philosophical error are deep (and such inclinations are not products of stupidity), and the baseline we get back to is not forever fixed but shifts as our linguistic usage does, requiring some acuity in grammatical investigations to recognise. Nevertheless, something important can seem to go missing here.
Such a philosophy can appear very conservative, lacking the sort of critical function which has animated much of the best philosophical work. For those of us with an affinity to Wittgenstein as well as the post-Kantian tradition, through Hegel and Marx to the Frankfurt school, this presents something of a problem. How do we ensure that our philosophising respects the insights of a therapeutic approach and yet remains able to interrogate our everyday assumptions?
One attractive answer to that question would be to deepen our understanding of the sources of philosophical problems, not resting content with a linguistic turn alone. For language is, of course, a practice — one that takes place in a wider social world. Following this line, there may be room for a marriage of critical theory and therapeutic philosophy. For example, such an approach might try and trace a connection between the perennial temptation to forms of Cartesianism and the alienation engendered by the conditions of life in modernity. The upshot of such an approach would likely not be a philosophical therapy that tried to return the wayward philosopher to ordinary linguistic usage, but rather identified what social conditions would need to be changed in order to stem intuitive but misleading forms of thought. I don’t know much Adorno, but my suggestion here I think might end up sharing some aspects of his approach.
Without going this far though, there are still important tasks that Wittgensteinian methodology can be put too. Here, perhaps the most important is holding the line against virulent forms of reductive naturalism. Recourse to grammatical investigation can be a tool in defending a kind of philosophical humanism: a position which takes human life to be just as substantial and respectable as the domain studied by the natural sciences. Our ordinary activities are shot through with appeals to values, to our dispositions, to the contingencies of our history (a history which no less unfolds in nature than that of supernovae or trilobites). Anatomising these sort of appeals in the manner of a grammatical investigator can help us understand the place of humanity in a natural world, and can be drawn on in resisting the rabid reductive naturalist who wants to evacuate meaning in favour of mechanism. Bare appeals to the phenomenology of human experience are cheap, but grammatical investigation in a Wittgensteinian vein can help draw out the underlying patterns of human activity in a more substantial way. This sort of rich understanding of the role of our human qualities as something which are (and should continue to be) drawn upon without embarrassment in our explanatory endeavours can be employed to stave off the sort of naturalist for whom all this is merely folk psychological self-delusion. It is, of course, not enough to say that ‘this is just what we do’ and expect the reductive naturalist to be satisfied, but this can be an important first step in resisting the breezy dismissal of human attitudes as no more than mere projections onto an indifferent world which a properly scientific cast of mind can see through.
Grundlegung 
That kind of thing is what I think I find most worthwhile in Wittgenstein’s actual writing, as opposed to things like the reconstruction/invention rule-following argument by Kripke.
I do have worries about negativity though. It would be very easy to simply be in the business of attacking some views in cognitive science, say, and at the end to have nothing to replace it with. Maybe that would show that there is just nothing to be said about the relevant issues, but to me that looks like a pretty bleak place to end up in.
I don’t think the conclusion should be that there is just nothing to be said about certain issues, say, in the philosophy of mind or language; it isn’t a ‘whereof we cannot speak, we must remain silent’ situation. What I take Wittgenstein to be doing in many cases is questioning why we think we ought to give a theoretical account of some phenomena. So, I don’t think anyone is saying that we ought not to do cognitive science, or that the results of cognitive science can’t have a bearing on philosophical issues. Rather, the thought being challenged is this: that unless we have an account that systematically explains the causal mechanisms or relations of supervenience that are in play with respect to mental phenomena, say, then our explicitly philosophical understanding of the situation must be incomplete.
Take Brandom as an example (who despite himself is rather un-Wittgensteinian). The way he approaches topics like meaning or normativity is in the spirit of a mystery that needs to be explained. Initially, they are treated as phenomena which appear to jar with a naturalistic picture of the universe, but then Brandom comes to the rescue by showing us how, after all, we can understand the complex behaviour of animals to constitue meaningful utterances or correct performances. So, for example, he gives us a theory of semantics which is reductionist, excluding all appeal to prior semantic concepts in its explanations. But I think this is already too late in the day: the very idea that talk of meaning is not on a firm foundation until such an account has been given is a highly corrosive thought. Of course, being able to give a succesful account of this form would be neat — there would be real explanatory value to it (though I am suspicious of such an account’s chances of success). However, the idea that such an account has to be out there, and that we are in trouble if we don’t find it — that important philosophical questions go unanswered until it is found — is undermotivated.
I don’t think there is a knock-down argument for why such inquiries can’t work (though I think there can be plenty of reasons to oppose specific projects, like semantic reductionism), and I think Wittgenstein somewhat unsatifyingly urges us to look at the details of language carefully enough and we will realise that a theoretical account is superfluous as against a set of reminders about how we use words. However, as Wittgenstein repeatedly remarks, it is an almost impossible task to get the ordinary philosopher to even start considering the sort of ‘grammatical’ details that Wittgenstein thinks are sufficient. According to him, the urge is to think that such details simply couldn’t satify us, so there is no point looking at them — that only a conceptual argument or causal mechanism can do the trick. (This lack of a general methodological argument against traditional philosophy is part of why Wittgensteinians can seem so cultish. They simply adopt an alternative perspective, and find that the world makes more sense that way: the proof is in the pudding; and there is little in the way of general deductive arguments forcing such a shift, only a torrent of micro-benefits which appear retroactively.)
I suppose this passage sums up much of the above:
I agree that we don’t want to think of philosophy as a “purely negative activity,” and that Wittgenstein sometimes says things that make it seem like that’s what he was after. But I don’t think that can be right in general as an interpretation of Wittgenstein. Same for “the linguistic turn” (think Davidson rather than Strawson).
So I don’t think we’d be looking in the right place if we turned, with critical theorists, to look for some particular social conditions of modernity which are responsible for our philosophical problems. Not that there couldn’t be interesting things to say about that; but you’d still have to go back and do the philosophical work, not only for its own sake, but also to show that your sociological analysis was really relevant to it. That would, I think, bring us right back to where we left off. (On the other hand that might help – even if it wouldn’t be strictly Wittgensteinian any more.)
For myself, I’d rather use Wittgenstein against such nihilistic (skeptical) senses of “therapy” than attributing them to him and trying to supplement him with other things (at least so directly). This is tricky though, no doubt about it.
I like what you say about Brandom though (in the comment). I’ve never understood his appeals to Wittgenstein. He seems the most unWittgensteinian of the whole bunch of post-analytics.