Language and Mind

When I was in graduate school in the early 1990’s, we were told that philosophy of language was over, and philosophy of mind was the future of philosophical research. Here was the standard story. In the dark days of Twentieth Century philosophy, language had a privileged place; some philosophers even held that all philosophical problems were fundamentally problems about language. People soon realized that this was incorrect. Even after we were perfectly clear about the grammar and meaning of the philosophically controversial claims, many philosophical problems remained (such as whether the perfectly clear controversial claim was true). But the excessive focus on language remained. In investigating the philosophical problem of intentionality, the problem of how representational elements come to be about the things in the world they represent, philosophers focused in the first instance on linguistic representations. But this was a mistake.

     Paul Grice came on the scene, and showed how to reduce linguistic meaning to lots of facts about intentions. This led philosophers to recognize that questions about the intentionality of language – and indeed philosophical questions about language generally – were really misplaced questions that genuinely should be about mental states. Looking at properties of language was looking at the wrong place. As my thesis advisor put the point to me once, looking at language for elucidation about intentionality involved making the same mistake as the man who looks under a streetlight for the keys he dropped, on the grounds that the street where he dropped the keys had no streetlight. One of the central morals of Twentieth Century philosophy, according to this story, is that there really was no philosophical interest in studying language. Now that this confusion was cleared up, philosophical progress could really be made, as everyone turned their attention to studying the mind (if you want to get the flavor of those times, look at Tyler Burge’s contribution “Philosophy of Language and Mind: 1950-1990”, in the centenary issue of The Philosophical Review).

     As far as I can tell, the future that was supposed to happen never occurred. Though the philosophy of mind is a healthy sub-field of philosophy, no great revolution occurred there in the 1990’s, and the philosophical study of language (and its allied field, philosophical logic) continue to provide a rich and interesting source of problems and insights. Furthermore, considerations from philosophy of language continue to be deeply relevant in a wide variety of disciplines (e.g. epistemology, metaphysics, and metaethics). This is not to deny that interesting things have occurred in the philosophy of mind; the study of consciousness has flourished, stoked in large part by the innovative work of David Chalmers and others. But much of the debate in this area too involves reconfigured debates from the philosophy of language. 

     So what happened? Why is philosophy of language still a major source of philosophical insight and activity, and philosophy of mind (at least of the relatively a prioristic variety) perhaps somewhat less active than it has been in the past? I think there are several reasons. Looking back on the last thirty-five years, perhaps the most interesting and philosophically fruitful inquiry outside of political philosophy and ethics was the intensive study of conditional and counterfactual sentences. Real progress was made in understanding these ubiquitous linguistic constructions, and genuine puzzles emerged (such as David Lewis’s triviality results), with which we are still grappling. Some people took these puzzles (rightly or wrongly) to motivate radical new theories of content (expressivism). In general, intensive focus by the community on conditionals and counterfactual constructions clearly led to philosophical progress, both in understanding these constructions, and in recognizing new possibilities. It is this kind of work that inspired philosophers to look for other linguistic constructions, the study of which could be equally fruitful. In contrast, it’s not clear how to use the insight that the intentionality of mind is prior to the intentionality of language to break new ground. The intentionality of the mental is prior to the intentionality of language: now what? 

     Secondly, philosophy of language and philosophical logic have clear methodologies. In producing a systematic account of the meaning of a certain kind of linguistic construction, there are wrong answers (lots of them). If your theory of X in metaphysics entails certain facts about the meanings of some sentences, those are consequences that we know how to evaluate. In contrast, the methodology of (relatively a prioristic) philosophy of mind was the thought experiment, and the thought experiment proved to be a difficult methodology to control. Indeed, much of philosophy of mind became the study of this methodology, and why it is so difficult to control (i.e. questions about the relation between conceivability and possibility).

     I’m not sure what moral to draw from this tiny slice of the history of philosophy I’ve lived through. Perhaps it is that having a reliable methodology is the best path to breaking new ground, and if you don’t, what you’re going to be spending your time doing is wondering about the methodology. 

     UPDATE: I've received too many interesting e-mails about this, so now I'm opening up comments...

-Jason Stanley

Comments

[djc]

This doesn't seem quite right to me. The philosophy of mind is probably more active than it's ever been. Apart from consciousness, there's been a huge renaissance in the philosophy of perception, the metaphysics of mind is as active as ever, and there are all sorts of newly thriving subareas such as emotion, self-knowledge, and so on (and of course major growth in the philosophy of cognitive science). The only area that might have slowed down a bit is the area that neighbours the philosophy of language, i.e. the study of intentionality and mental content. This is still very active, but maybe not quite on the same scale as in the 1980s (due in part to the dead-ending of research programs on naturalistic theories of mental content). I guess that's the bit that seems most salient from a philosopher of language's perspective, as it was the part that was allegedly allegedly going to put the philosophy of language out of business, but that was always an over-the-top idea.

As for methodology, I think that what's really going on is that philosophy of mind is a much bigger field with many different methodologies in place in different subareas and different communities. That's a good thing in some respects, but it does make for some fragmentation, compared to the relative unity and uniformity of approach in the much smaller community of philosophers of language.

In the spirit of Jason's trash-talking above, one could also argue that recent developments in the philosophy of language, while really interesting and often methodologically sophisticated, have been more narrowly focused than the developments that put philosophy of language at the centre of philosophy earlier in the 20th century. While there are still interesting connections to the rest of philosophy, one couldn't really say that philosophy of language is serving as "first philosophy" in the way it did for the philosophical community circa Russell, Quine, Kripke, and Davidson. But likewise, philosophy of mind hasn't taken over as "first philosophy" either. Probably that's a good thing, since philosophy is all the richer without a "first philosophy".

[Josh Dever]

Well, perhaps not surprisingly, as a philosopher of language I rather liked the trash-talking. A blog box doesn't seem like the place for subtlety in viewpoint, so I'll make the very broad claim that it looks to me like the conceivability and knowledge arguments in the philosophy of mind eventually work their way down to being disputes in philosophy of language, and that it's unclear to me that the contemporary discussion in mind has moved much beyond the point to which Kamp, Segerberg, Kripke, Kaplan, and Stalnaker brought us in the 1970's. As I see it, the central question (in this dispute) right now is whether a workable two-dimensional semantics for a natural language can be given (which is in turn the question of whether natural language contains Kaplanian monsters). The answering of that question seems to be occurring largely in philosophy of language (e.g., Philippe Schlenker's recent piece on monsters, Barbara Partee's examples of monstrous constructions, and Scott Soames' forthcoming book raising compositionality concerns about two-dimensionalism).

[Alva]

Dear Jason,

Thanks for this interesting and somewhat surprising post on Language and Mind. There are no first philosophies. It was a mistake to think it was epistemoloogy back in Descartes' day, and it was a mistake, back in the sixties, seventies, and eighties, to think it was language and logic. Philosophy is more integrated than that, more unified in its concerns. Themes linguistic are as central now in philosophy as they ever have been, to be sure. But your remarks about the philosophy of mind seem parochial (at least from my point of view). We are, I think, in a golden age of philosophy of mind, and what drives it is not disappointment with philosophy of language. What drives it is the development of an empirical cognitive science that rightly sees philosophy as an integral component. Philosophical and empirical problems about the mind are interpenetrating. The birth of cognitive neuroscience, and developments in cognitive psychology, robotics and philosophy have opened up a field of enormous potential and excitement. One of the features of the new philosophy of mind is its break with unsatisfactory pictures of the relation between philosophy and science, on the one hand, and also its movement beyond the confines of analytic philosophy on the other. (This is partly due to the fact that German and French scientists are playing a role in the cognitive neuroscience revolution, and they all grew up reading Merleau-Ponty.)

Here's a canned history that is perhaps closer to the truth. The new philosophy of mind is rising out of the ashes of an earlier cognitive science that was too a-biological, a cognitive science whose roots were logico-linguistic. It isn't that functionalism is dead (or that it should be) -- not at all -- but it is the case the more we learn about the brain/body/environment complex, the harder it is to take seriously philosophers' blithe talk of multiple realizability and the like. To the extent that today's best philosophical work on the mind tends to revolt against a Chomskyan or Fodorian picture of cognitive science, to that extent it's natural to see the mind-orientation as counter to the language orientation. But that's a superficial opposition, I would say.

Personally, I think it is outrageous that there isn't more integration of the mind and lanuage worlds. I go to all the mind/consciousness meetings, and there's almost no language presence. Why is this? I'm not sure.

--Alva

[Ralph]

Unfortunately for me, given that I am primarily a metaethicist and epistemologist rather than a philosopher of language, I think that there is a good sense in which philosophy of language can claim to be the closest thing that there is to "first philosophy". At least I can't see any dialectically effective starting point for metaethics and epistemology other than an investigation of our ethical and epistemic discourse.

In my view, philosophy of language is not first in order of explanation - far from it - but at least in metaethics and epistemology, it is methodologically first (it has the "first word", as Austen said, even if it certainly doesn't have the "last word"). This is because very often, the least controversial data that a philosopher can appeal to is that he or she is strongly inclined to accept certain utterances in certain circumstances.

This is the basis on which one then might try to answer other questions. E.g. on this basis one might try to determine: whether some of the utterances in question express beliefs, or non-cognitive states of some other kind; what exactly is the nature of the mental states that those utterances express; whether the fundamental theory of the logical behaviour of those utterances must appeal to the idea that those utterances can be true and false, or whether some other sort of theory might be preferable; whether there is anything that makes those utterances true, when they are true; what the nature is of whatever it is that makes those utterances true when they are true; and so on.

It would clearly be hopeless to start with ethical thinking rather than ethical discourse, since it's controversial whether ethical thinking consists of genuine beliefs with propositional contents or whether it consists of something else altogether. It would be equally hopeless to start by asking the metaphysical questions about what if anything makes ethical statements true, when they are true -- since it is controversial whether these statements can ever be true at all. Moreover, the importance in epistemology of the essentially linguistic questions about whether or not contextualism is true points to a similar lesson in epistemology, I believe. In short, the linguistic data seems methodologically the most secure starting point that the metaethicist or epistemologist is likely to find.

[Alva]

Ralph,

I take your point. There is a way of framing problems in metaethics and epistemology that takes problems about language as prior. I'm prepared to admit that this is so, and that the metaethics, etc, that arises in this way is worthwhile. But surely that's not the only way to proceed. I would have thought that there is a way of proceeding in this area that pays no attention to utterances, or linguistic intuitions at all. For example, one might hold that the least controversial data that a philosopher can appeal to is that he or she is strongly inclined to judge certain thoughts to be true (or to make certain judgments or to believe certain things about the right and the wrong) in certain circumstances.

Granted, if you assume that the subject matter of metaethics concerns the status of utterances -- e.g. whether they express beliefs or non-cognitive states, etc -- then it would seem that metaethics comes second to the study of language; indeed, it would then seem that the subject of metaethics is nothing more than the phenomena exhibited in a certain region of linguistic discourse.

But now it does begin to sound as if we're back in Oxford tutorials learning about the new analytic first philosophy of language...

Whether philosophy of language can play this role depends, I assume, on substantive questions in the philosophy of language as to the relation, say, between language and thought. I think those questions remain undecided. And those questions, I hasten to add, might just as well be taken as questions in the philosophy of mind.

--Alva

[Jason Stanley]

Alva,
One point I think remains correct from the Oxford tutorial years (at least when Dummett was teaching them) is that it is difficult to investigate the structure of thought independently of the structure of language. It seems that when we try to tell each other what the thoughts are that we're supposed to have intuitions about, we tend to use language. Of course, this doesn't mean that language has any kind of metaphysical priority over thought (this I take it was one of the mistakes of 1970s Oxford). Rather, language is the tool we use for expressing our thoughts, and it doesn't seem we have another vehicle for expressing our thoughts. As a result, investigation into the content of linguistic utterances has provided us with the models we have for debates about the contents of thoughts.
So, the point is not that (as Dummett perhaps concluded from the above epistemic points) that language is somehow more 'fundamental' than thoughts. It's that we developed very specific models of content by looking at a wide range of linguistic constructions, and we couldn't have perhaps developed these models by just looking at thoughts. These models can then be used to give an account of the content of our mental states, not just our linguistic utterances. I take it one of Ralph's points is that when we are adjudicating between these models of content, we tend to appeal again to how we frame the disputed class of thoughts in language.