Quine, Sellars, Empiricism, and the Linguistic Turn

Quine's "Two Dogmas of Empiricism" is thought of by most analytic philosophers as the watershed moment in the destruction of logical positivism. He may have been the first major post-linguistic turn philosopher to identify themself as a pragmatist (and here I'm not counting his Harvard colleague Morton White in the equation). Replaying the death agonies of logical positivism is a good way of seeing where the classical pragmatists tie in with the contemporary, analytic dialectic, and that paper is a must read in that sense.

What Quine wanted to do was purify empiricism, which in its post-linguistic turn phase is what logical positivism was supposed to be a form of. Analytic philosophers, ever since somebody thought it would be brilliant to turn to the study of the way we use words (whether mainline positivists like Rudolph Carnap and A.J. Ayer or Oxford/ordinary-language philosophers like P.F. Strawson and Gilbert Ryle), thought their job was to study "meaning," to find the meanings of words, display how meaning was created, etc. "Conceptual analysis" became what philosophers thought they did, what their special provenance was in relation to the other disciplines. It promoted the thought that, since science was the empirical discipline, science would tell us about the stuff of the world, but philosophy would explain to science what their words meant.

How empirical is that? is roughly the thought that motivated Quine. The very idea of "conceptual analysis" was of basically an a priori discipline, an activity where one didn't have to study anything in the world, but could just sit in your armchair and figure the stuff out--just like Descartes' problems of other minds and the external world. But wasn't the motivation behind logical positivism the destruction and avoidance of pointless metaphysical problems, weird things like whether there was anything going on outside your door ("leave your office" would have been G.E. Moore's reply), which are specifically produced by a kind of armchair speculation that required no input from the world?

Quine's first dogma was the analytic/synthetic distinction, something Kant had made up at the spur of the moment in the First Critique as a way of splitting the difference between the Rationalists and Empiricists (only so named because of Kant's work). Kant's distinction became the basis of the distinction between statements whose meaning and truth was derived only in virtue of their relationship to other statements ("analytic statements" such as "All bachelors are single") and statements whose meaning and truth one not only needed other statements, but also participation of the world to figure out ("rocks fall to the earth," the truth of which can only be determined by a synthesis of the meanings of each individual word and then looking out into the world). Science would do the "looking out into the world" bit, but they would need someone to help them with the other bit, the analytic statements you don't need experiments for. This, for someone like Quine, basically just looks like a Rationalist throw-back, a haven for the Cartesian speculation the sober-minded English Empiricists had wanted to throw cold water over.

By ditching the first dogma, we eliminate the basis for an armchair, a priori discipline like "conceptual analysis," thus putting us back on the path of a thorough-going empiricism. Quine, it is largely thought, didn't quite make it by himself, however. His second dogma, reductionism, was the pernicious reduction of knowledge statements to sensations, or "immediate experience." People impressed by James and Dewey, who still used such locutions, shouldn't be fooled by the term into thinking that Quine's dogma swings at what the classical pragmatists were talking about. James' radical empiricism was supposed to be as much a purification of the terms of empiricism as the post-positivist dialectic was. Quine, though, didn't in the end mean the end of that dogma was much as he should have. He erects in his work the idea of "observation sentences" as opposed other sentences, and the latter end up playing the same role as before. And then there's Quine's over-bearing scientism, his penchant for saying that everything can/should be reduced to physics, the only language that "limns the world."

Philosophers dissatisfied with Quine's tack at reductionism have increasingly turned to Wilfrid Sellars, a by-comparison neglected figure that I believe recently has been receiving a substantial reappreciation (primarily motivated by the rise of Robert Brandom and John McDowell). Sellars' seminal "Empiricism and Philosophy of Mind" was produced at just about the same time as Quine's "Two Dogmas," but received much less attention. Sellars' enemy in that paper is what he calls the Myth of the Given, the idea that there is a bald experience given to our minds that we simply add the hairplugs of language to. This was the attack Quine should have made on reductionism, but didn't.

Sellars, too though, didn't quite make it to a pure empiricism. He, too, liked to talk about science as the end all be all, and his way of making the point was a distinction between our manifest image of the world and the scientific image of the world. The manifest one was the fake one that, while not reducible to the scientific one, needed to be dealt with properly in the face of the true, scientific image--and philosophy could help with its "conceptual analysis."

Rorty liked to say that it was almost as if Sellars and Quine were only able to reject the one dogma of the other, and needed their respective second one to retain their self-image as analytic philosophers. Donald Davidson, Quine's greatest pupil, helped ditch the whole damn thing and finally set us on the path of a fully purified empiricism. He called the third, and hopefully last, dogma of empiricism the scheme/content distinction. This distinction underlaid the others and its motivation was to call into question the idea that language was a kind of scheme we laid on top of experienced content. This, too, was a Kantian relic, that between what he called "concepts" and "intuitions," such that one could say, "thoughts without content are empty, intuitions without concepts are blind."

This is what Kant meant by the mind constituting the world--the world's there, sure enough, but its only presentation to us is by virtue of our mind's constitution of it in our own mind. But this raises the Cartesian spectare again, that we are trapped in our own minds, maybe knowing that there are other ones out there, but not knowing whether we are constituting the world in just the same way--if we are using different conceptual schemes, we might be living in different worlds. Please Obi-philosophy Kenobi--tell us what our conceptual schemes are and how they constitute. Davidson's ditching of the scheme/content distinction again destroys the basis for an armchair, speculative discipline and thrusts us all back into the same world, the one no one has ever really left.

The punchline to the story is that whereas Quine wanted to be both an empiricist and a pragmatist, Davidson isn't sure what's left of either once we pull the underpinnings out. Rorty and Hilary Putnam, the two most prominent post-linguistic turn self-identified pragmatists, themselves have wondered explicitly about what is left of empiricism once one ditches all the dogmas of positivism, though they think the core insights of James and Dewey untouched. (I have a brief talking up of Putnam on this score here.) Even more weird is Brandom calling his Sellarsian inferentialism a kind of rationalism--a much different sort than the 17th century Europeans, but the resurrection of the title bearing out how much damage Brandom thinks Locke's confusion of causation with justification caused philosophy.

The "confusion of causation with justification" is a direct allusion to Rorty, in Philosophy and the Mirror of Nature. In the chapter that opens the second part, the part that deals with representational epistemology, he titles one of the sections "Locke's Confusion of Explanation with Justification," which is roughly between knowledge of where something came from and knowledge that something is so and so. This leads to Davidson's distinction between reasons and causes, such that a reason (part of the web of justification) can be a cause (can change your mind, spur you into action, etc.), but a cause is not in itself a reason. You can be caused to think you see water, but the fact that you see water is not, by itself, a reason to think so--it is only within a network of other beliefs, the whole context of the situation, that a caused-belief can become a reason-to-believe. For instance, the fact that you are in a desert, dying of dehydration, and gored out of your mind on shrooms might give one pause before transferring the sight of water to having a good reason to think there's water over that horizon ahead.

I think it an open question as to whether retro-pragmatists like David Hildebrand are right about there being an important line to be drawn between classical pragmatists like James and Dewey and neopragmatists like Rorty and Putnam, one roughly centering around the "radical empiricism" of the former set, and supposed lack there-of in the latter. I still tend to think that there's simply an unimportant line between the classical tendency to talk about experience and the neo tendency to talk about language, with no further major philosophical implications.

The main thesis for this line of thought is that the role radical empiricism plays for James (with Dewey having an analogous section of his philosophy that could be so-named) is the same role "psychological nominalism" plays for Sellars (with a likewise analogous section in Rorty). We might call the slogan of radical empiricism "everything is experience," and the intuitive appeal of this slogan makes it easy to see why James and Dewey might wield it opposition to those who take, for instance, meanings to be analytic, and therefore non-experiential (for if they were, they'd be synthetic in the requisite sense). Psychological nominalism's slogan, however, is "all awareness is a linguistic affair," which on the surface seems counter-intuitive--am I really aware of that sunset I'm appreciating silently linguistically? I don't think psychological nominalism, however, quite means this kind of thing. It has to be understood in the context of various kinds of atomism, particularly the kinds that surfaced in early analytic philosophy.

My suggestion about the parallel qualities of radical empiricism and psychological nominalism is that both are kinds of holism, and that the only difference between the two is a difference in jargon, in the state of the philosophical dialogue that each arose out of and responded to. We might encapsulate the differences by saying that modern philosophy was birthed out of Greek when talk moved from being about "reality" to being about "experience." Rorty, Whitehead, and Dewey all advanced historical arguments about the lack of an internal "place" called the mind where reality played itself out for us (Whitehead called it the birth of the "subjectivist principle"). This created a divorce between reality and our experience-of-reality such that now we had to deal with problems about just when we were in touch with reality.

The atomistic response are various correspondentisms, philosophical theories about how this bit of experience rubs up against this bit reality, which it does in a one-to-one relation. Dewey and James advanced their ideas about experience in this milieu of experience-talk, and their basic suggestion was to collapse experience back into reality--experience is reality (witness James' "a world of pure experience" and Dewey's appreciation of Aristotle's pre-modern, anti-Platonism). Their's was a kind of holism, for it consisted in the idea that our experiences all relate to each other in a hanging web, and trying to pinpoint connections between this web and something else is pure folly. For the web, our experience of life, works in getting us through life, and it is only a kind of retrograde metaphysical dogmatism that keeps us trying for something more--we need to stop thinking there is something more.

Psychological nominalism, on the other hand, consists in the same move, except now the linguistic turn has made philosophers think that there is a divorce between language and experience. Early atomists like Russell and Carnap spent their time trying to elaborate theories that specified when this linguistic-bit here connected to that experience-chunk over there. Sellars attack on the Myth of the Given was on the idea that our linguistic concepts overlaid bald experiences. "All awareness is a linguistic affair" is simply the analogous collapse of language into experience that James and Dewey did for experience into reality.

What we basically have is a line that looks like this: language-about-experience-of-reality. The Greeks, having just become leisured, reflective individuals, talked about "reality," though they occasionally would stumble down a path where it sounds as if they recognize the modern fact that we each experience reality differently. They did understand this commonsensically, but they didn't charge it with any special philosophical significance. It was only after the march of thousands of more philosophers, trampling down the philosophical terrain, trying to get various theories to work, that Descartes, Locke, and the rest suddenly get the idea that, maybe we should charge the fact that different eyeballs see different ways with philosophically-charged metaphorical significance. And so they began promoting an expansion of what knowing is of from the object "reality" to tacking on "experience-of-reality."

This new philosophical situation took hold in part because of an equivocation in terminology--Descartes and Locke used the word "idea" to denote both (what we might now distinguish as) perceptions and conceptions. As modern philosophy moved passed Kant, who isolated explicitly the two (in intuitions and concepts respectively), and thousands of more philosophers (aided by expansions in population, education, and professionalization) trampled down this new terrain, philosophers began sniffing more and more around the idea that, hey, what really is the difference between a concept and word? Isn't it just a silly Platonism, with a universal Realm of Forms, that would make us think otherwise? (This is what makes sense out of the "nominalism" bit in Sellars' platform name--nominalism was the medieval counter to Plato's Realm of Forms idea.)

The linguistic turn was roughly the realization that when we talk about reality, we are talking about experience or reality, we are using words, language, and that when this or that philosopher, be they Platonist, Thomist, Cartesian, Kantian or Russellian, suggests taking up this or that philosophical position, they aren't changing the world like a bridge-builder does, they are suggesting a change in thought, which is to say, in our conceptions, which is to say, in the words we use to describe the world. And so analytic philosophers began promoting a further expansion of what knowing is of from the object "experience-of-reality" to tacking on "language-about-experience-of-reality."

One way to conceive of what the holists are up to is to see them as trying to get us back to a kind of pre-Platonic common sense position, before the whole obsession over knowing what knowing was even started. The only point in collapsing language-into-experience or experience-into-reality is if you'd also be just as willing to finish the collapse into the remainder, unconsidered term.

The classical pragmatists got the feeling that philosophy was taking us away from life, and they adumbrated radical empiricisms as a way of returning us to the scene of life. Retro-pragmatists see the linguistic turn as another philosophical circumlocution that just gets in the way life. Retro-pragmatists are attracted to the classicals' repeated insistence that their philosophical positions will reestablish life as first in the philosophical order of things.

My thought tends to run, "How could life not come first? Life is everything." As I understand Dewey, his notion of indirect experience was as reflective experience, but direct experience wasn't necessarily non-linguistic (I adumbrate this understanding here). Abstract, reflective thinking pulls us out of "direct experience," but this is just to say that it pulls us out of our typical habits with which we approach the world. The first principle that all three classical pragmatists held in equally tight fashion was Alexander Bain's, that thoughts are habits of action. People who live in their heads live in worlds constructed out of their imagination, an indirect world if you will. But people who write, read, or talk a lot aren't necessarily neglecting direct experience, in Dewey's sense. We can easily achieve equilibrium in our lives, between reading the newspaper, watching sunsets, loving our partners, and writing philosophical papers about the pointlessness of atomism.

I think at the heart of the difference between philosophers attracted to the classical pragmatists but repelled by Rorty is the thought that radical empiricism returns us to the scene of life, a counter to abstract philosophical sterilities. I can empathize with the formulation, to the idea of pragmatism "returning us to the scene of life," a formula I've grown fond of. However, what I think we should rather say in most cases, is that philosophy is abstract by nature--that's what it is--and returning to the scene of life is something that people need to figure out how to do, not necessarily philosophies, or other abstract activities. For instance, why would we necessarily want theoretical physics to do so? Philosophy is Dewey's indirect experience--returning to life is knowing, as Wittgenstein put it, when to put philosophy down.

And yet still, to say as many want to, particularly amateur philosophers, as I am, that we should return philosophy to the scene of life, I have continued sympathy with. But I also have a strong regard for the division of labor, and a suspicion that if we let a thousand flowers bloom, there's a good chance we might find one we like in a garden that we didn't plant. That's what I take amateur philosophy, in particular, to be--one of our major purposes is to steal other people's flowers and bring the ones we like, even if they were grown in a sterile hydroponic farm, back to the grubby garden we got going behind the house. We shouldn't rip on the hydroponic farmers too much--without them, we wouldn't have gotten that flower and, hey, at the end of the day, they go back to their houses, too. I sense this wisdom in Rorty's writings about the relative merits of professionalization in philosophy, and that that is the parallel to James and Dewey's talk about getting back to life. All three talked a lot about setting aside pointless philosophical puzzles. This is just to say that perhaps we've improved this particular flower as much as we can, so perhaps we should move on to some others.

Be that as it may, reading Quine or Sellars or Davidson can help one figure out what kind of empiricist one wants to be by helping see what the destruction of logical positivism pans out to mean. Empiricism began as a reaction to Descartes' attempt to laud science's place in culture. The rationalists thought the New Science was an expansion of Reason's will over us (it was) whereas the empiricists thought the New Science was inauguration of a new experimentalness (it was). They were both wrong in their rightness, and while there may yet be interesting disagreement about the rhetorical upshot of this or that old wine bottle ("Empiricism stands and falls with the idea of unconceptualized content, and to say otherwise is just confused." "Nominalism? Really? Should we really be going back to the Middle Ages?"), what needs to be kept at the forefront is the new wine that everybody is drinking.