Floris asked:

I have a question concerning Kant's epistemology. I have been very confused about the different
pairs of expressions analytic/ synthetic and a priori/ a posteriori.

In short, my teacher has taught me that, according to Kant, an analytic statement is one for which no
empirical experience is required, while a synthetic statement involves empirical knowledge. The
difference between a priori and a posteriori is, so she said, that a priori sentences are sentences in
which the predicate is contained in the concept of the subject, while in a posteriori sentences this is
not the case.

Now I am very confused having read her definition of a posteriori/ a priori being in my view identical to
the definition of synthetic/analytic as described in Thomas Mautner's Dictionary of Philosophy(page
19). Her definition of synthetic/analytic was furthermore identical to Mautner's definition of a priori/a
posteriori (page 33).

As I asked her about this, she said I must have misunderstood what Mautner wanted to say, and, to
my grief, she refused to continue our conversation until I had read Kant's Critique of Pure Reasonor
at least ProlegomenaThen I would see why, she said.

Now I have read the first half of the Prolegomena,but still I can't think but that Mautner's definition is
compatible with Kant's text and hers is not, which, moreover, makes Kant's expressions seem rather
empty (like his questions about the synthetic a priori would be quite useless in my view, if we apply
my teacher's definition to them).

Is the philosophical community not unanimous as to what the exact definition of the two pairs of
expressions (synthetic/ analytic and a priori/a posteriori) is? If not, why is that so?

============

Your teacher is confused. The distinction between analytic and synthetic is a distinction about
sentences (Kant would have said "judgements." It is a logical distinction. The distinction between a
priori and a posteriori is a distinction about how we knowa sentence is true (or false) It is an
epistemological distinction.

Kant said that an analytic sentence or judgement is one whose "predicate is contained in its subject."
For instance, "all dogs are animals." A synthetic sentence is one whose predicate is not "contained in
its subject. For instance,"all dogs are carnivores (eat meat)" A different way of talking about the
analytic-synthetic distinction is to say that analytic sentences are true in virtue of the meanings of
their terms, and synthetic sentences are not true in virtue of the meanings of their terms. On the other
hand, a sentence that is known a priori is notknown empirically or on the basis of sense-evidence. It
is known independentlyof sense-evidence.But a sentence known a posteriori (or empirically) is
known on the basis of (dependently on) sense-evidence.

Kant says that it is clear that all analytic sentences are known to be true a priori, since they are true in
virtue of the meanings of their terms. But,now, what about synthetic sentences? According to Kant,
although some synthetic sentences are know a posteriori or empirically on the basis of sense
evidence,this is not true of all synthetic sentences. It is true, for example, of all dogs are carnivores.
We could not know that unless we observed dogs and what they eat. But what about a sentence like,
Whatever is red is colored? Kant denies that the concept of color is contained in the term red. But he
asserts that although we know this sentence is true, we do notknow it is true on the basis of sense
evidence, a posteriori. Kant insists we know it a priori, that is, independently of sense-evidence.
Another, and more famous example that Kant gives is the sentence "Every event has some cause"
(The law of universal causation). Again, Kant says of this sentence that although it is synthetic, it is
known independently of sense-experience,and so, is known a priori. So, according to Kant, although
all analytic sentences are known a priori, it is false that all a priori sentences are analytic.Some
sentences known a priori are synthetic. After these distinctions are made, and Kant establishes that
there are synthetic sentences known a priori, Kant asks his seminal question: How is synthetic a priori
knowledge possible?It is this question to which the entire Critique of Pure Reasonis devoted to
answering.

Some philosophers, in opposition to Kant, have held that the term "analytic"is co-extensive with the
term, "a priori" and the term, "synthetic" is co-extensive with the term "a posteriori." That is to say,
these philosophers hold that all synthetic sentences are also known a posteriori, and all sentences
known a posteriori are synthetic and that all analytic sentence are known a priori,and all sentences
knows a priori are analytic sentences. Your teacher seems to be one of these. This is a possible and
respectable position, and has engendered much controversy in philosophy since it denies Kant's
claim that some synthetic sentences are know a priori. But where your teacher is confused is in
thinking that the terms "synthetic" and "a posteriori" meanthe same thing (are synonomous); and that
the terms "analytic" and "a priori" meanthe same thing (are synonomous) and that is simply false,
since, as I pointed out when I began, the analytic-synthetic distinction is a logical distinction,whereas
the a priori -a posteriori distinction is an epistemological distinction about how we know.

Besides, to maintain that these pairs of terms are synonomous is simply to rule out Kant's question,
"How is synthetic a priori knowledge possible"by definition, since it would make it a self-contradictory
question. And this would be unfair and somewhat silly.

Ken Stern

I have not read the book by Mautner you mention. It would probably be better for you to forget it, if its
confusing. It seems to me that it is incorrect to define a priori as meaning the same thing as analytic
and a posteriori as synthetic. A priori, literally, means 'without prior experience', a posteriori means
'with experience'.

It is tempting to do what the positivists did and identify tautologies (analytic propositions) with a priori
ones. Surely the only things you know without 'synthetic' experience of the world are trivial logical
facts, such as 'A=A' or 'something can't both be and not be at the same time'. You don't need
experiences to know that these trivial tautologies are analytically true,

Kant's point in theCritique of Pure Reasonis that some non-trivial facts are known without
experience (a priori) but are not simply 'analytic' tautologies. For example, the proposition 'there is
space' is not analytic on your teacher's definition (it isn't a tautology), but according to Kant you
cannot find out the truth of the proposition 'there is space' by looking at your experiences. What he is
saying is that space is pre-supposed (i.e. it is an a priori concept) in order for there to be experiences
at all.

Unfortunately, Kant is one of the worst writers to have ever tried writing philosophy. In particular, his
Transcendental Deduction is amazingly badly written for such a powerful argument. This tends to
disguise the importance and brilliance of what he actually argues. The main thing to grasp with the
introduction to theCritique of Pure Reasonis how you can have 'a priori synthetic' propositions: i.e.
sentences that arenottautologies but whose truth is not derived from synthetic experiences of the
world. Understanding this requires you to ask the Kantian question: whatneeds to be the caseabout
the worldin order for this experience to be thus and so.

You could also read Quine's essay 'Two Dogmas of Empiricism' in which he questions whether these
distinctions are even right (after all, defining an analytic proposition as 'one in which the predicate is
contained within the concept of the subject' isn't amazingly helpful) My experience was that a few
hours worth of serious thought was the best way to grasping Kant on the analytic-synthetic/ a priori-a
posteriori issue.

A. Gatward