Coming Soon:

Now Available: Volumes I, II, III, and IV of the Collected Published and Unpublished Papers.



Total Pageviews

Saturday, July 30, 2011


Some More Terminology

Inasmuch as I must clear up some last minute things before flying to San Francisco for a week-long family gathering, and in view of the fact that I have arrived at the most difficult part of the Critique, I thought that instead of continuing today, I would spend a few moments clearing up some more matters of Kantian terminology. So this will be a brief post, explaining Kant's use of three terms, all of which can mislead the unwary: Deduction, Transcendental, and Idea.

Deduction: All of us are familiar with this term, which in philosophical contexts usually means an argument that derives conclusions from premises by the rigorous application of the rules of logical inference. In Kant's day, that would of course mean a syllogistic argument. But there was another use of the term ":deduction," in the law of Kant's day. A deduction in the law was a demonstration of title, an argument showing that someone had a legitimate claim to something. Kant decided to take over this jurisprudential usage, and apply it to a certain kind of argument that was central to his undertaking. A "deduction of a concept," in Kant's sense of that phrase, is a demonstration that the concept has legitimate application to a certain sphere of objects. This is not a demonstration that the concept actually does have instances. Only experience could settle that. But a deduction of the concept. if successful, would show that it is possible for the concept to find correct application in experience.

Now, at the most superficial level, we can provide deductions of concepts like "horse" or "man" or "blue" or "sharp" by exhibiting a man or a horse or a blue thing or a sharp thing. We can also provide deductions of such concepts as "unicorn" by specifying the perceptual evidence that would justify us in applying the concept to an object -- viz. seeing a white horse, with a single horn in the middle of its forehead, that is unusually attracted to virgins. [I say "at the most superficial level" because eventually, Kant will argue that concepts like "horse" and "man." since they involve the concept of substance, require something more elaborate in the way of a deduction. Sorry for all the quotes, by the way. When I was young and impressionable, Willard van Orman Quine beat into me the distinction between use and mention and I never got over it.]

But concepts like "cause" and "effect" and "substance' cannot be given deductions in this same way because, as Hume so powerfully argued, mere examination of our sensory experience is inadequate to demonstrate that we are in the presence of a causal connection or a substance. So a fairly powerful argument is going to be needed to establish that "cause and effect," "substance," and a number of other concepts can find legitimate employment in experience. Kant calls such an argument a "Deduction of the Pure Concepts of Understanding," which is the title of the most important section in the Critique.

Transcendental -- Transcendent: These two terms have completely different meanings in Kant's writings. He draws a very sharp and absolutely clear distinction between them. Unfortunately, he then more or less forgets that, and constantly uses one when he means the other, forcing the reader to keep adding or subtracting an "al" when the term crops up. This is very irritating, no doubt, but Kant never gets confused, he is just uncharacteristically sloppy, so with a little care it is always possible to tell which term he means. "Transcendent" means just what you would imagine it should mean, namely "going beyond the limits of experience." Thus one of the most important conclusions of Kant's argument in the Critique is that we can never have knowledge of transcendent reality. This means that we can never have knowledge of the existence and nature of God, or of things as they are in themselves. As we shall see, one of the consequences of this limitation on the scope of our knowledge is that we can never know that we are free, that we are moral agents. We can believe that we are [he says], we can act as though we know that we are [he says], but we cannot know that we are, because such knowledge would go beyond the limits of experience, it would be knowledge of the transcendent.

"Transcendental" doesn't mean anything like that at all. As Kant uses the term, it means what today we would call "epistemological." That is to say, a transcendental investigation is an investigation into the possibility, nature, grounds, and limits of knowledge. It is a philosophical investigation designed to determine whether we can know anything, under what conditions we can know anything, and what sorts of things we can and cannot know, by virtue of the structure and limits of our cognitive faculties. Since at a superficial level Kant does not think there is any problem about ordinary concepts like "horse" and "man," he does not think we need to give an empirical deduction of them. But as we have seen, he very definitely thinks we need to demonstrate that concepts like "cause and effect" and "substance" can find legitimate application within experience, so we need an epistemological argument to establish that very strong claim, contra Hume. Kant calls that argument a transcendental deduction. Such an argument not only proves that the concepts have legitimate application within experience but also marks out precisely the limits of their legitimate application, so that we are not tempted to try to employ them transcendently [NOT transcendentally], as metaphysicians have been doing since Plato's day.

Once having formulated the notion of a transcendental deduction, Kant was quite taken with it, and it shows up all over the place in his writings. In the Third Critique, for example, he offers a Deduction of Judgments of Taste. The official Deduction of the Pure Concepts of Understanding appears in the section bearing that name, but as we shall see, the real argument is not completed until many pages later, in the section called The Analogies of Experience, and once we understand the argument at its deepest level, we shall see that Kant really has a fully successful argument only for the concept "cause and effect" and maybe for the concept "substance." Still and all, that is not just chopped chicken liver, as they say in the culture I grew up in.

Idea: As I noted in one of the very first parts of this extended series of notes, the philosophers of the seventeenth and eighteenth centuries had a variety of terms they used for the notion of cognitively significant contents of consciousness. Locke uses "idea," Hume uses "perception," and Kant uses "representation." Kant had so much he wanted to say that he felt the need for an elaborate multiplication and differentiation of terms to keep it all straight. First of all, he distinguishes between two mental faculties that, on the face of it, look pretty much the same, namely Understanding and Reason. Both of them introduce unity into a manifold of elements. Understanding introduces unity into a manifold of sense contents, or perceptions, and the "functions of unity," as Kant calls them, by which it does this he calls concepts. Understanding has both a merely logical use and a real use. In its merely logical use, it unifies a diversity of elements or contents of consciousness by formulating judgments. Hence the various logical forms in which it does this are called by him "Functions of Unity in Judgments." [See the first table above.] In Understanding's real use, it introduces unity into a manifold of perceptions and the various forms in which it does this -- we have not yet learned just what these actually are -- he calls Categories.

Reason does the same thing that Understanding does -- that is, it introduces unity into a diversity of elements. But the elements on which it imposes unity are concepts, not perceptions. Reason too has a merely logical employment and a real employment. The merely logical employment of Reason produces chains of syllogistic reasoning -- what Kant calls ratiocinatio polysyllogistica -- one of my all time favorite philosophical terms. Reason, he tells us, has a built-in irresistible thirst for bringing these efforts to completion. It produces out of its own inner resources the notion of the unconditioned, for which it is constantly seeking to find employment. He calls this notion of the unconditioned an Idea to distinguish it from concepts, which are one and all conditioned in their employment by being limited to the sphere of things as they appear to us.

Reason, endlessly busy, elaborates many versions of the Idea of the Unconditioned, such as an Infinite Being [God], a Prime Mover, a First Cause, a First Premise that is not itself the conclusion of any prior syllogism, a Necessary Being, Absolute Unity, Absolute Simplicity, Absolute Totality, the Absolutely Universally Unconditioned Moral Law, and so forth. In fact, Kant concludes, all of Metaphysics if nothing but Reason's ceaseless attempt to apply the various forms of the Idea of the Unconditioned.

Now, this innate tendency of Reason is impossible to root out and eliminate, Kant says, and it leads the mind constantly to overreach, claiming things that it cannot possibly know. Eventually, he will trace all of the Antinomies bedeviling previous philosophy to this fatal flaw. However, waste not, want not. Kant decides that properly understood and contained, Reason's lust for the Unconditioned [my term, not his!] has its uses. It is what drives the forward march of science, for example. When physicists first try to unify all terrestrial motion in Galileo's laws, and all heavenly motion in Kepler's laws, and then seek to derive both the laws of terrestrial motion and the laws of celestial motion from Newton's single set of laws, they are answering Reason's call to seek the Unconditioned. When Einstein spent the latter part of his life unsuccessfully seeking a way to derive the general theory of relativity and the laws of electromagnetism from a single set of equations, he was answering the same call.

So, this is what Kant means by "Ideas of Reason."

And with that, I shall complete my packing and get ready to see my son, daughter-in-law, and grandchildren, and Susie's sons, daughters-in-law, and grandchildren. See you in a week.


J.Vlasits said...

Have a good trip! I did want to say that, although this series of posts may seem an impossible task, I've been enjoying them immensely--as much as I enjoyed your Formal Political Philosophy posts. This will be a great help for me personally because next semester I'll be starting a PhD program in Philosophy at Berkeley and will likely take Dan Warren's Kant class.

Robert Paul Wolff said...

Good luck at Berkeley. If he is still around, say hello to my old friend and fellow grad student, Hubert Dreyfus.