![[Home]](http://www.bayleshanks.com/cartoonbayle.png)
Copyright 2007 Bayle Shanks released under the Creative Commons Sharealike NC 2.0 license.
This should go without saying when i post notes on a book, particularly on something that i understand as little as philosophy, but: i'm probably misinterpreting some things and so some of this is probably wrong.
Note that most statements found on this page are my interpretation of what Kant says, not statements about what I personally believe. Indeed, when I say "I" below, I am usually speaking as Kant, except in my aside, and in my footnotes.
Some of my interpretations are close to Kant's phrasings. Not quoting something does not mean that I claim credit for coming up with a totally different, better phrasing, it means only that I do not assert that it is exactly what Kant (as translated) said.
Even when I quote Kant (double quotes), I make very liberal use of ellipsis (...). Sometimes I even skip over things that slightly change the meaning, such as qualifications that I consider unnecessary.
I want to note somewhere that a friend has suggested that Kant provides a blueprint for the mind of a rational being (my words, not his), and, as a student of cognitive studies, this is my primary interest in reading Kant.
The main question of this book is: is metaphysics possible?
I (Kant) got to thinking about this stuff after reading David Hume. Hume studied the the connection of cause and effect.
Defn "cause": If A causes B, then if A is posited, B necessarily must also be posited.
Hume asked if the connection of cause and effect between two events can be deduced a priori using reason. His answer was no. Everyone misunderstood Hume. People thought he was arguing that we should give up thinking in terms of cause and effect, which he was not. Hume agreed that, when thinking about the physical world, cause and effect is indispensible. Hume's question was just, is the connection between cause and effect an a priori concept borne of pure reason?
Hume's argument was that the connection between cause and effect contains necessity. If you can conceive of two different things A and B, then it doesn't make sense to say a priori that just because A is, B must necessarily also be.
Hume then concluded that the concept of the connection of cause and effect has nothing to do with a priori pure reason, and is really just an a posteriori observation. I say, rather, that the concept of cause, considered in general, is in fact a priori.
In terms of method, I don't agree with philosophers who want to appeal to "ordinary common sense" to decide what is correct philosophy. Common sense is great for everyday life, but when you are debating people with different points of view, you need to use rigor. Hume's opponents said that cause and effect are real just because common sense says it is so. By contrast, I aim to give a rigorous ground for my views.
I have come to the conclusion that there are also other concepts which, similar to the concept of cause and effect, which are a priori. In the Critique of Pure Reason, I prove a formal deduction of these concepts. I also rigorously determine the boundaries of correct a priori reasoning.
I have written this Prolegomena as an prologue to the Critique of Pure Reason. The Critique is difficult to read and so I thought it would be helpful to have an overview of it. The Prolegonmena gives the plan for the work that is actually carried out in the Critique. However, please look at this work as only preparatory exercises for reading the Critique.
In the Critique, I have attempted to completely describe the boundaries of the domain of pure reason. Because pure reason is such an isolated domain, there is nothing outside of it that could correct our judgement within it. Therefore, the validity and use of each part depends on the relation in which it stands to the others within reason itself. And so I think a Critique of Pure Reason must be entirely complete to be convincing, and that in the domain of this faculty one must determine and settle either all or nothing.
Even if you don't agree with my solution to the problem of how metaphysics can exist, this Prolegomena will still be useful because it certainly shows that the question of whether and how metaphysics can exist must be answered.
"reason": a faculty
"the understanding": a faculty
"a posteriori": reasoning that make use of things which are given in experience
"a priori": reasoning that is not a posteriori; "has an inner truth independent of all experience". Often Kant puts "a priori" after a noun phrase where I would have put it before (as a modifier of the noun phrase); for example, at once place later on he says that such-and-such can consist only of "judgements a priori", whereas I think the normal usage nowdays is to say "a priori judgements".
"pure": synonym for a priori
"pure reason": the faculty of a priori reasoning
"apodictic": an apodictic statement is one that is irresistibly convincing (i.e. any reasonable person will be convinced of it)
"science": an area of study which is irresistably convincing
"natural science": an area of study including both what we today would call "science", i.e. the deduction of natural laws from a posteriori experimentation, and also what we today would call part of philosophy, i.e. a priori reasoning about concepts used for the interpretation of the world. This is a little confusing because Kant says that there are parts of natural science which are necessarily true and which can be taken as a priori. He doesn't mean physical laws which we deduce from experimentation, though, he means various deductions that we would today call "philosophy" rather than "science".
By definition, the source of metaphysical cognition cannot be empirical, and must therefore be a priori.
What's the difference between metaphysics and mathematics? The reader is referred to pp. 712 f. of the Critique of Pure Reason.
Propositions are usually either analytic, or synthetic, or explicative.
defn Explicative propositions: propositions that merely re-explain without adding any new content.
defn Analytic judgements: "say nothing in the predicate except what was thought already in the concept of the subject, though not so clearly nor with the same consciousness".
The common principal of all analytic judgements is the principal of contradiction. For example, the definition of "body" is "something with extension", and from this you can derive "every body is extended" and also "no body is unextended". Another example: the definition of gold (actually, he says "the concept of gold") includes that gold is a body, is yellow, and is a metal. So you can say "gold is a yellow metal" and this is an analytic judgement. The second example shows that analytic judgements may concern themselves with empirical concepts.
defn Synthetic judgements: those which require a principal other than the principal of contradiction.
So, if you make a (content-ful) judgement that isn't analytic, then that judgement is synthetic. This isn't to say that synthetic judgements can contradict analytic ones; they have to be consistent with analytic judgements, but they go further.
Note that synthetic propositions might still be necessarily true.
I know of at least two ways to interpret Kant's analytic/synthetic distinction. Part of this discussion will follow Wikipedia (http://en.wikipedia.org/wiki/Analytic-synthetic_distinction; http://en.wikipedia.org/w/index.php?title=Analytic-synthetic_distinction&oldid=150868325; 13 August 2007) Before giving the alternate interpretations, let's restate Kant's definition of the terms:
analytic proposition (according to Kant acc. to wikipedia): a proposition whose predicate concept is contained in its subject concept
synthetic proposition (acc. to Kant acc. to wikipedia): a proposition whose predicate concept is not contained in its subject concept
The first way, which I tend to agree with, is that what Kant meant was this (ignoring explicative propositions):
analytic proposition: a proposition whose truth depends solely on the meaning of its terms
synthetic proposition: a proposition which is not an analytic proposition
A very similar definition would be:
analytic proposition: a proposition that is true by definition
synthetic proposition: a proposition which is not an analytic proposition
Note that under this interpretation, a mathematical proof (a deduction using accepted formal rules of logic, once the axioms are taken as given) is analytic. Such a deduction is necessarily true just because of the formal definitions of things, given the axioms.
But Kant says that mathematics is not analytic1. How can that be? Under this interpretation, we read that as meaning that the choice of axioms, and in that, relation of mathematical concepts to the real world, goes beyond analytic. For instance, the decision that space is 3 dimensional (which is involved in choosing axioms for the mathematics of space) is a synthetic one. We could do formal mathematics with different axioms, but they wouldn't describe actual space.
A friend of mine has a different interpretation. In this interpretation, synthetic is when the representation (a representation is here thought of as a data structure) of the conclusion of the theorem is different from the representation of the premises. Therefore it's not the case that the conclusion of the theorem says nothing "except what was thought already in the concept of the [premises]". So, in hir interpretation, a formal proof need not be analytic. An example of something analytic is when you start with one representation, and then present just a restricted part of the same representation; for instance, if gold is defined as "yellow metal", then you can say, "gold is yellow".
Under this interpretation, a formal deduction might be considered synthetic even if the axioms are taken as given, if the conclusion of the proof is put into a different representation than the premises.
The Wikipedia page is helpful here.
Wikipedia says that "One common criticism is that Kant's notion of 'conceptual containment' is highly metaphorical, and thus unclear." Wikipedia goes on to say that, "the logical positivists drew a new distinction, and, inheriting the terms from Kant, christened it the "analytic/synthetic distinction."
However, my friend presents an interpretation of the terms that is quite different from the logical positivists', yet still consistent with Kant's definition.
Therefore I conclude that Kant's definition is indeed underspecified, due, as Wikipedia suggests, to its metaphorical nature (specifically, the notion of "conceptual containment"; Kant's definition uses the notion of "what was thought already in the concept of the subject", but since he hasn't defined (at least not prior to this definition in the text of the prolegomena) the terms "thought", "concept", or what it means for a thought to be "in" a concept). This metaphorical nature permits multiple very different interpretations to be consistent with Kant's words.
Luckily, although Kant states that the distinction between analytic and synthetic is essential for metaphysics, in my opinion very few of the arguments in the Prolegomena require the reader to make use of this concept. So the reader may continue onward without pausing to take firm stance on this issue.
I note that as a matter of hermeneutics, my friend proposes that the way such interpretation questions should be resolved is to choose whichever interpretations allows Kant not to contradict himself or make any errors.
Not to bias the reader's choice of interpretation of Kant, but merely to inform the reader of how the term is commonly used today, the current usage of the term is given by Wikipedia as "Analytic propositions are those which are true simply in virtue of their meaning while synthetic propositions are not." (i.e. the first of the definitions that Wikipedia says were put forward by the logical positivists).
There are three types of synthetic judgements:
Properly metaphysical judgements are simply defined as those metaphysical judgements which are not analytic.
Another example of a judgement that is synthetic (although Kant doesn't say whether or not he agrees with it) is the principal of sufficient reason.
judgements: i dunno, but at the beginning of "On the distinction between synthetic and analytic judgements in general", he says that "metaphysical cognition must contain nothing but judgements a priori", so I guess a body of knowledge is made of judgements.
"principal of sufficient reason": "The principal of sufficient reason proposes that some sort of explanation must exist for everything" (http://www.crossroads-cc.org/app/w_page.php?type=section&id=19).
If there were a book or body of knowledge on the topic of metaphysics that argued from pure reason and was irresistibly convincing, then that would serve as an existence proof that metaphysics is possible. But since we don't have such a book, we'll have to consider the question by other methods.
The highest aim of metaphysics is "knowledge of a supreme being and a future life, proven from principals of pure reason".
What has been done so far? Many analytic metaphysical propositions have been derived. But the synthetic propositions which have been presented (for instance, the principal of sufficient reason) have not been "proved a priori" 2
Although we must not assume that metaphysics is possible, we do know that at least some a priori synthetic judgements exist. There are a priori synthetic judgments in mathematics (see above) and also in the philosophy of nature.
problematic: i think that by this he means that some concept has the status that we don't even know if it's possible or not. We can't just use the phrase "a possible concept" because maybe later we'll prove that it's impossible. So instead of calling it "a possible concept", he calls it a "problematic" concept. Not sure about this though.
OK, actually, everyone can see how a priori analytic judgements 3 are possible 4