Why should we care about Kant?

by Dave Maier

A reader writes in to ask (“naïvely,” as he thinks):

In Kant's transcendental idealism, what is at stake? And, if we take it that it is a form of anti-realism and a source of later anti-realisms in philosophy, what is at stake in anti-realism? […] I see that one answer is that anti-realism lends itself readily to relativism, and if we want the comfort of believing that what we hold to be true or valuable is true or valuable “in fact”, then realism seems attractive and anti-realism problematic, whereas on the other hand we might think that we empowering our mental faculties in some important way if we adopt an idealist or anti-realist conclusion. [Also,] if truth has some kind of epistemic criteria and does not reduce to correspondence to a reality, but these criteria are objective in the sense that they enable us to establish that a claim is indeed true or correct, this might be thought to facilitate the meaningfulness of our moral and aesthetic claims against the possibilities of emotivism and expressivism. […] Maybe this move owes a debt to Kant, because Kant's idealism gives epistemic criteria for the truth even of empirical propositions while nonetheless maintaining an “empirical realism” according to which there are objective truths.

Dear reader:

That is one heck of a question, and not at all naïve. It would be a better and much different world if everyone were clear on realism, anti-realism, their relation to each other and to the other views which try to get beyond them, not to mention Kant. I am not a Kant scholar, but as a pragmatist I do have some things to say about realism et al, so let me give your question a necessarily compressed as well as highly contentious go.

Immanuel_kant As I see it, the best way to approach Kant is to see him as trying to get past a traditional impasse (not of course always seen as such by the tradition itself): that our intellect is such as constantly to pose questions about its relation to the world that it cannot answer. It's not that these questions are simply too hard – there's no philosophical problem with that, as we run into such problems all the time – but that they are incoherent. That is, in reflecting on itself as operating in the way that it in fact does, our intellect is inevitably drawn to certain conceptual train wrecks like a moth to a, um, train wreck. In Kant's own striking image, we're like the dove, who, tired of dealing with wind resistance, yearns for the ideal perfection of a vacuum – which would of course make flight impossible. (For more on the dove image, see here – a much better post than this one, if I do say so myself.) According to Kant and those of us following him this far, the trick is to get off the train at the right time – not following the moth into disaster, but also resisting nihilistic calls to abandon train travel entirely. As you can imagine, this can be a tricky business indeed.

As modern philosophers tell it, this platonic urge toward transcendent metaphysics was supposed to have been overcome by the naturalistic turn to empirical science; but as Hume showed, this move simply makes the same problem take a different shape. I think of this shape as basically Cartesian. That is, contemporary platonism takes a Cartesian form: it gets rid of the transcendental world of forms in favor of the natural world, but then in the new picture, the latter takes on many of the former's attributes. (Indeed, in some ways it's even worse: at least the forms aren't “independent of our minds” [i.e. qua logos] in the bizarrely incoherent way the Cartesian “real world” is supposed to be, which contains us qua object but not qua subject, whatever that means.)

On the other hand, at least the modern picture allows us to make better sense than the medievals did of empirical science as gaining us knowledge of a reality not of our making, which is nothing to sneeze at. However, it's not 1650 any more, and the very idea of modern empirical science and its characteristic virtues no longer needs the sort of philosophical promotion it relied on in pushing against the medieval/Aristotelian paradigm, and which is now causing trouble, sort of like a booster rocket that won't detach.

Kant sees this, and he wants to turn a crisis into an opportunity. The crisis was the Humean alarm bell which famously awoke Kant from his “dogmatic [ = Cartesian] slumber.” As you might imagine, opinions differ about the opportunity Kant saw. Like any such project, that of moving beyond Cartesianism has a negative part and a positive part: you show why and how the old view is wrong, and you substitute a better one. But by now especially, there's nothing specifically Kantian in the very idea of rejecting the Cartesian picture. In fact some ways of doing so paint Kant as just as much a villain as Descartes.

MW cover The most common way to do this is to see Kant as primarily responding to the skeptical challenge (which of course he is in some way). On this view, Kant brings empirical reality within the reach of our knowledge, but only at the cost of pushing real (“noumenal”) reality farther away: not only can't we know such a reality, we can't even conceive it, as it escapes our cognitive faculty entirely. Naturally this is no good. Barry Stroud, for example, himself completely entranced by the Cartesian skeptical dialectic, can neither let Kant subjectivize so unacceptably the reality we know, nor render reality as it really is so definitively inaccessible. Even philosophers sympathetic to Kant's basic strategy can read Kant as making this mistake. John McDowell, in Mind and World, accuses Kant of spoiling his perfectly good point about the conceptual nature of empirical reality by unnecessarily enclosing it in a “transcendental framework” (see Lecture II, section 9 (pp. 40-44).

McDowell's accusation provoked a heated response from some of Kant's contemporary defenders, including Graham Bird and Henry Allison, who read Kant as rejecting the Cartesian picture in a very straightforward way, unfortunate obscurities arising only when Kant tries to put his own constructive spin on the results. On this basic view (I am here running together a number of people, none of whom should be held responsible for the hash I am about to sling), there is no such thing as “reality-in-itself” in the relevant sense, which is what Kant has been trying to tell us all along, all that talk about the unknowable noumenon notwithstanding.

Remember the original formulation: that our intellect, simply in seeing itself going about its daily business, is fatally tempted by the Cartesian siren song when it tries to reflect on its activity. In one sense, it's perfectly natural to see ourselves as in search of an “objective” view of reality: that's one which is not distorted by our necessarily subjective perspectives on it. Seeing this danger, we use various techniques, including scientific experiment and mathematical rigor, to reduce this distortion as much as possible. The ideal of absolute objectivity gets closer and closer, until it seems almost within our grasp. If we could just stay perfectly still and hold our breaths … ! Once we have made the fatal identification of really true knowledge with knowledge of a world-in-itself, then we can find it impossible to settle for anything less – until, facing the inevitable, we make a virtue of necessity and turn away from “metaphysics” to the modest, revisable claims of empirical science.

But this is either a temporary respite – indeed, today we find the most robust realists within the halls of science, not without – or a sort of philosophical Stockholm syndrome, as when we make fallibilist empiricism as consistent as it can be made, it turns out to be skepticism once again, presented as virtue rather than tragedy. If everything we “know” is carefully regarded as possibly false, then the skeptic might as well be right: we can regard nothing – nothing not guaranteed by formal abstraction – as known without doubt. As a rule of thumb for investigators, empiricist skepticism (regarded, again, as a virtue, as in “scientific skepticism” of pseudoscientific balderdash) is perfectly appropriate; but in our context it's a philosophical dead end.

Summarizing madly here, we can think of this pro-Kantian view as claiming that Kant's “idealist” talk is exhausted by his rejection of the Cartesian conception of a potentially knowable (and thus disturbingly unknown) world-in-itself as an incoherent conflation of two distinct ideas: the objective world which is the “object” of our knowledge, and the “objective” world as a world from which the subjectivity has been abstracted away. It is that conflation which Kant sees as the sad legacy of the Cartesian/Faustian bargain of modernity. We will emerge at last into the light, he thinks, when we embrace this “Copernican Revolution” and stop trying to eliminate our own contribution, as subjects, to our knowledge of objective reality.

That story is one of triumph, made unfortunately obscure by Kant's opaque writing style. And it's right as far as it goes. For a while I was reluctant to go any farther, as Bird et al are certainly right that most readers of Kant don't get even this far with him, and that this is an entirely salutary lesson that we all must learn before we can dig ourselves out of the Cartesian hole. However, I have reluctantly come to believe that we can't really leave it at that.

Bird's sort of Kantian responds to the typical accusation that Kant is an “idealist” (indeed, as in the words of one writer, “a confused Berkeley”) with a forceful denial. He rejects “realism,” all right, but that sort of “realism” is a transcendental fantasy better done without, a dove's naive yearning for the vacuum. He uses the term, yes, but no other “idealism” is intended. I wanted to believe this, even as I saw it as an ineffective criticism of McDowell, who is more or less in agreement on the key points, but has other fish to fry as well. But I still keep coming back to the idea of the “Copernican Revolution.” Kant really seems to be making much more of it – the idea that objects conform to the mind rather than the other way around – than simply diagnosing the Cartesian confusion, as if he were a 20th-century linguistic philosopher like Austin. And that Transcendental Deduction! If Kant's point is so deflationary, so … Rortyan, then why go to such impenetrable lengths? We can't just say, oh well, that's the sort of floundering around (and flirtation with “idealism”) you had to do before we saw the linguistic roots of these puzzles … which we do now thanks in part to Kant himself, etc.

Pippin No, I think we need to bite the relevant bullet and allow that Kant is an “idealist” after all, and to see his positive project as not simply an overcomplicated way of defending the negative (anti-Cartesian) one, virtuous as that latter is, but an attempt at the very sort of metaphysics to which, as he says elsewhere, the rest of his argument is simply a “prolegomenon.” If we limit our purview to Kant's own views, however, it is very difficult to get a satisfactory perspective from so close to. This explains the importance of Robert Pippin's groundbreaking book Hegel's Idealism. Since it came out in 1989, scholars have been trying to realign the “idealisms” of Kant and Hegel rather than seeing Hegel as primarily a critic. Pippin's book is about Hegel, of course, and naturally enough the Hegel community has plenty to say about this non-standard and seemingly excessively Kantian Hegel. I am incompetent to judge that debate. But the flip side of seeing Hegel as a post-Kantian is seeing Kant as a proto-Hegelian, which might give us the focus we need. This is what McDowell, Pippin, Robert Brandom, and a few others seem to be up to.

Unfortunately I cannot competently judge or even explain this either. What I can do is draw a moral or two and steer the conversation back in more congenial directions. The first moral is that anti-Cartesians need not be so skittish about “idealism” (or “metaphysics” either, for that matter). Those are indeed terms in which contemporary Cartesians (malgré leur as they may be) will attack us, in defense of realism and empiricism. And, like “relativism”, they can indeed refer to philosophical errors. But – as Kant saw, even if we don't follow him all the way to … wherever that is exactly – just as we must board the train as “realists” of some kind, we can't effectively disembark without seeing ourselves just as much as “idealists.” My own background in analytic philosophy makes me deeply uncomfortable at this prospect, so I prefer (barring intelligible results from the neo-Hegelian camp) to think of this as rather the necessity to take up an idealist perspective for a particular purpose, rather than to propound a stable “idealist” doctrine. These may turn out the same in the end (or at least come within shouting distance, which may after all be preferable). Even here, though, I have simply exchanged one form of realist abuse for another, for my own “perspectivism” can seem just as relativistic as “transcendental idealism” (or Hegelian “objective idealism”) is, well, idealistic.

However, I do think – and I better save this for another time – that so far at least, a post-Wittgensteinian version of the Kantian lesson makes it easier for us to keep our eye on the rapidly moving ball. 20th-century linguistic philosophy may not be necessary for the success of an “idealist” project, whether Kantian, Hegelian, or whatever. However, in the light of this close examination of language's actual workings, “perspectivism” can be seen as not at all the same as relativism. To take up a “perspective” on reality is, well, to take up a perspective on reality, and not some relativistic substitute. It is the unfortunate necessity to fight this sort of fight against realist abuse that makes it difficult to embrace anything called “idealism,” and I imagine something similar is true for “idealists” re: perspectivism (or pragmatism). But this view of language use itself allows us to put these differences into the proper relation. Or at least I hope it will, when the time comes – especially since McDowell makes so many of his other salutary points in Davidsonian and/or Wittgensteinian terms. However, I do think it will be a while before we can really hook up properly. Stay tuned!

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