Wednesday, January 10, 2007

Nietzsche on Kant

He was proud of having discovered a new faculty in man -- the faculty of making synthetic a priori judgments ... It is the ambition and rivalry of all the younger philosophers to discover "new faculties"! ... How are synthetic a priori judgements possible, Kant asked himself. And what was actually his answer? By virtue of virtue -- but unfortunately not in five words but so complicatedly, respectably, with such a show of German profundity and sinuosity, that one failed to hear the funny German simple-mindedness inherent in such an answer.


Beyond Good and Evil, trans. by Marianne Cowan, section 11. (I have no idea how to cite this.)

Unfortunately I haven't read enough Kant (I've only read the Grounding for the Metaphysics of Morals), but Nietzsche's criticism does strike me as true.

3 comments:

ADHR said...

Actually, it's not only not true, it verges on incoherence. For Kant, the fact of synthetic a priori judgements follows from a couple of claims. First, that there are judgements that are a priori -- judgements about the truth of mathematical statements, for example. Second, that there are judgements that are synthetic -- judgements about the truth of mathematical statments, for example.

Okay, facetious example, but the point is that we have clear cases of judgements that fall on either side of the a priori / a posteriori distinction and judgements that fall on either side of the analytic/synthetic distinction (Quine notwithstanding). In the former case, we can think about judgements about mathematical statements vs. judgements about counting objects in a room. I can know that 1+1=2 without comparing it to the experienced world, but I can't know how many things are in the room without counting them. In the latter case, we can think about judgements about the meaning of tautologous sentences ("a bachelor is an unmarried man") vs. judgements about the meaning of other sentences ("Rawlings is not a bachelor"). I can know that it is true that "a bachelor is an unmarried man" just by knowing what the terms mean. I have to actually find the person referred to by "Rawlings" and determine if he is a "bachelor" in order to tell if "Rawlings is not a bachelor" is true.

Kant's question, then, is just a question about whether these distinctions go together (any analytic judgement is a priori, any synthetic judgement is a posteriori, and vice versa) or whether they cross boundaries. He thinks that mathematical judgements are a paradigm case of synthetic a priori judgements: you don't have to go out into the experienced world to know whether they're true, but you can't figure out what they mean just by inspecting the terms involved.

That, at least, is my stab at the gist of what Kant was getting at. It's in the opening bit of the First Critique.

What Nietzsche is trying to say, I have no clue. By virtue of virtue? Zuh?

undergroundman said...

It's the same argument you pulled on me. Why is our conscience right? Because it 'seems right.' It's an "immediate" judgment which tells us what's right and wrong. Why is the categorical imperative right? Because it feels right.

In a little philosophy handbook I have, some examples of synthetic a priori judgments are "every event has a cause", "moral responsibility presupposes freedom of action", and "colors have dimension". Each one I would take issue with.

Kant argues that the categorical imperative is an the true morality - why? Because his conscience tells him that is the case.

I think that's what Nietzsche is getting at.

And, yeah, you're examples are a bit strange. :p Mathematical judgments are both? What are some things that are one but not the other?

I can know that 1+1=2 without comparing it to the experienced world

Arguably you wouldn't understand what that meant without the experienced world - the difference between a rationalist and empiricist? (I expect you'll brush that off as well.) Is it relevant to bring up the fact that our a priori judgements are necessarily connected to the existing world?

He thinks that mathematical judgements are a paradigm case of synthetic a priori judgements: you don't have to go out into the experienced world to know whether they're true, but you can't figure out what they mean just by inspecting the terms involved.

He's not talking about mathematics at all, as far as I can tell. Kant's synethic a priori logic was directed at other things as well (principally, even - or am I wrong?).

Incidentally, have you ever read a book called Logic by Immanuel Kant? I've got it sitting by me and I'm thinking about tackling it.

ADHR said...

You can take issue with particular examples of synthetic a priori judgements, but that doesn't refute the category unless the category is held to be incoherent. This is something like what Quine was trying to do in "Two Dogmas", in arguing against the analytic/synthetic distinction. If there's no such distinction, then there's no synthetic judgements whatsoever -- just a priori and a posteriori ones.

Perhaps the basic appeal, at the end, is to intuition (in the philosopher's sense). But I'm not sure why that would be a mark against a system. Every system, sooner or later, comes down to "it seems right". It's only if you think that more is, in principle, achievable that this starts to look inadequate.

FWIW, Kant rejected the rationalist/empiricist distinction on the grounds that, while nothing could be known unless it could be experienced, experience itself was structured by the mind. So, while you had to experience something in order to know what 1+1=2 means, determining whether or not it's true can be done a priori.

So, are a priori judgements "connected to the world"? Sort of, but here we get into Kant's trancendentalism (and here I start to have serious doubts about the adequacy of his argumentation). The idea, I think, is that we can judge the transcendental conditions of knowledge (and other things) a priori, but everything else must be judge a posteriori. That is, by inspecting our experiences, we can infer back to what conditions must have been in place for those experiences to occur in the way they did. This is an a priori matter, for Kant, and results in what he calls "transcendental" conditions. Within these conditions, though, we make a posteriori judgements. So, if you see or hear something, your judgements about it will be a posteriori; but, how you can see or hear anything at all (not just talking about details of the visual system here, but about how a sight or sound can be something you experience) is a matter of a priori judgement.

So, it's not just about mathematics, although that is one of his stock examples to try to persuade us to take synthetic a priori judgements seriously.

I haven't read the Logic. Unless you're looking to increase your historical knowledge, or you're a logician, I don't tend to think there's a lot of value in reading old logic books.