Kant's Criticism of Reason and the "Reductio ad Absurdum&quot

[Acount Overdrawn]

Kant's Criticism of Reason and the "reductio ad absurdum"

by Roderick Fitts

I'm not currently studying Immanuel Kant--I'm reading Nietzsche and I just started with Aristotle's "Categories." For instance, I commented on Nietzsche's view of concepts in my Facebook group "Objectivist Epistemology," contrasting it with the Objectivist theory of "measurement-omission."

However, as I mentioned in one of my summaries on Objectivism, I've read some of the beginning parts of Kant's Critique of Pure Reason, and I think I've read enough to state Kant's metaphysics in terms of essentials.

Kant posits that there are two realities: the noumenal and the phenomenal--or: "things in themselves" and "things in relation to subject" (a being with a rational faculty). One could not use any type of rational process in order to know the noumenal world exists, and it cannot be perceived.

Kant never really provides evidence for his claim that such a (noumenal) world exists, he simply defines reason, knowledge, understanding, concepts, etc. in such a way that they are restricted to the phenomenal world, and then claims that the noumenal world, because of the identities of those mind-related terms (the words I just listed), cannot know a noumenal world.

In other words, Kant's method of making a distinction between the noumenal and the phenomenal is a case study in rationalism: it proceeds deductively from limited premises (the definitions he gives), and it is presumed to include all the relevant considerations, and yet it ignores where and how it received the inductions necessary to deduce anything. [1] Though I'd like to follow up on this, I simply have not read enough yet to argue much further on this point: this point really isn't what this note is about anyway, just a lead into what could become a more fleshed and stronger argument. Back to the purpose of this note:

The question of the moment then is: how does Kant know that there is a noumenal world?

For him to posit a noumenal world (let's say it's "N") that cannot be inferred or shown to exist by reason, I infer that he must be using a non-rational means of knowledge, let's call it "K." [2] Now, if K exists, then it must possess identity, and because it is a means of knowledge for a subject, it can only know N in relation to K, not "N in itself."[3] (This is an application* of Kant's criticism of reason to his own system, from what I understand of it.)

And with a small edit, I'll finish my point:

One could go further and claim that the only way to know that K is not a means to know N in itself is to have a non-rational, non-K means of knowledge, let's say F--and so on. It would never cease.

In the end, you get Kant's implicit statement: "I know the noumenal world (N) exists in itself by K means." and "I cannot know N in itself by K means." Which is a contradiction.

The reductio ad absurdum fallacy involves taking a premise (or set of premises), deriving a contradiction from the premise, which establishes it as false. Since I used multiple premises, a contradiction would only show that at least one of the premises is wrong. Here are the premises I think I used:

My attempt at constructing the argument as a reductio ad absurdum

Premise 1 (P1): "Kant (or anyone) can know N-in-itself through non-rational means K" (This is the premise I'm trying to show is false)

Premise 2 (P2): K, as a means of knowledge, has a specific identity. (If it did not possess identity, it would not exist.)

co-premise to 2 (Cp): K has a specific human identity [4]

Premise 3 (P3): If P1, P2, and the Cp, then we cannot know N-in-itself, only N in relation to K (as a human process).

Conclusion: P1 and P3 lead to a contraction--therefore P1 is false.

Anyone taking a logic course that wants to help me out here is more than welcome.

References and Notes:

[1]: Like Kant's insistence on "a priori judgments"--which is taken to mean "independent of experience."

[2]: Kant rejects this notion through his dismissal of positive noumenon, however he still does not offer a means of knowing about the existence of the noumenal world. Rather than end the issue here, I'd like to draw out an absurd conclusion from his earlier premises.

[3]: Hattip to Harry Binswanger for pointing this part out regarding Kant. It's really an example of Ayn Rand's point that Kant's system is "a negation, not only of man's consciousness, but of any consciousness, of consciousness as such..." (Introduction to Objectivist Epistemology, p. 80)

[4]: Since the "reductio ad absurdum" is in formal logic, I'm not sure if I can use informal logic (co-premises), but here's my go at it.

*: Edited by Acount Overdrawn (I know it's spelled wrong; I think I ran out of room while typing it.) The unedited version read: "abbreviation," but I meant to type "application."

Edited December 12, 2007 by Acount Overdrawn

[cilphex]

Thanks for posting this. I've just started a course called "Critical Thinking" with the theme of Skepticism, which is being geared largely as an introduction to philosophy class. Reviewing some history, we've started with the sophists and are progressing onward through Socrates, etc. I suspect we'll eventually reach Kant. We were introduced to reductio ad absurdum today, but apparently the professor didn't explain it very well, only describing it as "making the claim you are trying to deconstruct seem absurd" instead of "following the claim logically to the contradictions on which it is based."

For him to posit a noumenal world (let's say it's "N") that cannot be inferred or shown to exist by reason, I infer that he must be using a non-rational means of knowledge, let's call it "K." [2] Now, if K exists, then it must possess identity, and because it is a means of knowledge for a subject, it can only know N in relation to K, not "N in itself."[3] (This is an application* of Kant's criticism of reason to his own system, from what I understand of it.)

What do you mean by "...and because it is a means of knowledge for a subject, it can only know N in relation to K"? How does K being a "means of knowledge for a subject" necessitate that K is a rational means of knowledge? This must be what you're claiming for that sentence to be true, no? Because if it isn't what you're claiming, then K remains, as it started, an irrational form of knowledge, which would mean that it could know "N in itself", right?

Is "subject" a synonym for "human," and because K is a human faculty, it must be rational? If so, was this according to Kant, or for another reason? (Or, should I have inferred the answer to this already somehow ?)

These are just some preliminary questions... I haven't studied logic and I don't have a great background on Kant, so I'm trying to get a better understanding.

Edited December 12, 2007 by cilphex

[Robert J. Kolker]

Did I read you right? You say P1 and P2 imply P3. P3 is false therefore P1 is false. There is the error. In fact you yourself said it. If a conjunction of premises lead to a false conclusion one of the conjoined premises must be false. But which one? In the specific instance all you can say is either P1 or P2 is false.

To quote a certain Russian born novelist --- Check your premises.

Bob Kolker

[Acount Overdrawn]

What do you mean by "...and because it is a means of knowledge for a subject, it can only know N in relation to K"? How does K being a "means of knowledge for a subject" necessitate that K is a rational means of knowledge?

I did not say that K is a rational means of knowledge. I said it was explicitly a non-rational means of knowledge, since "rational knowledge" in Kant's system could not grasp the noumenal world. In his writings, Kant implies that he knows the existence of this noumenal world, despite himself possessing "reason" (which can only know the phenomenal world), so I inferred that he must be using some other means of knowledge, a "non-rational" means, and I called it "K."

Because if it isn't what you're claiming, then K remains, as it started, an irrational form of knowledge, which would mean that it could know "N in itself", right?

By Kant's logic: no. "N in itself" is just that: N in itself, out of relation to any consciousness, human or otherwise. If K is what I said it is, a non-rational means for humans to know the existence of the noumenal world ("N"), then it is "N" in relation to a human consciousness with a "K" means of knowledge, which is not synonymous to "N in itself."

Is "subject" a synonym for "human," and because K is a human faculty, it must be rational?

I think "subject" in Kant's system can apply to any being with a consciousness, but in this case it is only referring to humans. No, as I said, in Kant's system a "rational" means of knowledge cannot know "N," so in order for "K" to be a means of knowing "N," it must be non-rational.

Thanks for the questions.

[Acount Overdrawn]

In the specific instance all you can say is either P1 or P2 is false.

Hmm, I was responding to your post on how P1 is false, when a thought occurred: Maybe P2 is false. Kant's point on consciousness and the noumenal world is that the former can never know or understand the latter. In whatever way Kant knows the existence of the noumenal world, it has no identity: if it possessed identity, then it would have a "human" identity, and P3 of my argument would follow. Kant implies (or maybe says outright) that we cannot shed our human identity to know the noumenal world, and yet, at the same time, he claims to know of the noumenal world's existence. Aargh!

So what is your opinion on this? Kant must then be saying that he has a way of knowing the noumenal world which has no identity, some type of mystical construct that we mere "non-Immanuel Kants" do not possess, and can never hope to understand. Anyone is free to respond to this, btw.

[cilphex]

Hmm, I was responding to your post on how P1 is false, when a thought occurred: Maybe P2 is false. Kant's point on consciousness and the noumenal world is that the former can never know or understand the latter. In whatever way Kant knows the existence of the noumenal world, it has no identity: if it possessed identity, then it would have a "human" identity, and P3 of my argument would follow. Kant implies (or maybe says outright) that we cannot shed our human identity to know the noumenal world, and yet, at the same time, he claims to know of the noumenal world's existence. Aargh! banghead.gif

Why do you even need to prove anything wrong? Isn't this just one big contradiction?

- Kant claims to know the noumenal world exists.

- According to Kant, the means of knowledge through which he can know the noumenal world exists must have no identity.

- Something without identity cannot exist, so Kant cannot know the noumenal world exists.

Am I oversimplifying?

[Benpercent]

Why do you even need to prove anything wrong? Isn't this just one big contradiction?

- Kant claims to know the noumenal world exists.

- According to Kant, the means of knowledge through which he can know the noumenal world exists must have no identity.

- Something without identity cannot exist, so Kant cannot know the noumenal world exists.

Am I oversimplifying?

Nothing was lost in translation, so I see no reason to say it's oversimplifying . If you simplify to the point where specifics are lost, then youl should worry.