Unconditioned, So If We Thought of Our A Priori Knowledge As Concerning Things in Themselves, We

Summary of the Critique of Pure Reason 1
Summary of the Critique of Pure Reason
John R. Searle
The Prefaces1
The basic idea of the prefaces is to announce the Copernican Revolution. Just as
Copernicus shifted the point of view from the earth as center to the sun as center, so Kant wants to
shift the point of view from objects as the source of a synthetic a priori knowledge to the human
mind as the source of such knowledge. Instead of thinking of the mind as having to be responsible
to objects, we should think of objects as having to be responsible to the a priori conditions set by
the mind.
In order, however, to carry out this project, we need to make a distinction between
appearances and things in themselves. The notion of things in themselves is the notion of the
unconditioned, so if we thought of our a priori knowledge as concerning things in themselves, we
would have a contradiction, because the notion of a priori knowledge would set a condition on the
unconditioned. But as Kant says, the contradiction vanishes the moment we make appearances the
subject matter of our knowledge because the a priori conditions are conditions on appearances. We
can have no knowledge of things in themselves, but can at least think of things in themselves.
Kant shows how the contradiction between free will and determinism is removed by making
the distinction between the will as a thing in itself and the will as appearance. Causation is a
condition on the will as appearance, but this does not prevent the will, as a thing in itself, from
being free.
There are two distinctions in this Preface that are fundamental for everything that follows.
One is the distinction between things in themselves and appearances and the second is the
distinction between those parts of the contents of the mind that he calls "sensible intuitions" by
which he means perception, and those parts which he calls "concepts" by which he means the
apparatus necessary for thinking. The distinction in short, is between perceptual intuition and
1 Note to the reader. I have tried to separate my critical comments from my summary of Kant's
theory. [My criticisms are in square brackets].
understanding. There is also a passage which is ominous and decisive. It is on page 22 and there he
refers to "objects, or what is the same thing,... the experience in which alone, as given objects, they
can be known". That is to say, that like Hume, he equates objects and experience. [That is the
fundamental disaster from which everything follows.]
Though all knowledge begins with experience, it doesn't derive from or follow out of
experience. Empirical knowledge may be made up of two components, that which we get through
the senses, and that which we contribute on our own, through our faculty of knowledge. Do we
have knowledge independent of experience? Such knowledge would be a priori, as opposed to
empirical, and what is the same thing, a posteriori. We need a criterion for distinguishing between
pure and empirical knowledge. Pure or a priori knowledge is both universal and necessary.
If we distinguish between analytic and synthetic judgments on the grounds that in analytic
judgments the predicate is contained in the subject and in synthetic judgments it is not, we discover
that there is a class of a priori synthetic judgments. But how is this possible? How do we make the
synthesis a priori, when the predicate is not contained in the subject?
There are three classes of synthetic a priori judgments.
1. Mathematics. Mathematical judgments such as 7+5=12 could not be analytic because the
concept of the sum of 7 and 5 contains nothing except the union of two numbers into one and no
analysis of the concept will show that the result must be 12. So, arithmetical judgments are both
synthetic and a a priori.
2. Natural science, namely physics, contains a priori synthetic judgments such as the law of
conservation of matter.
3. Metaphysics, although it is a mess, also contains such judgments.
So we are left with the question, how are these judgments possible? If you believe Hume,
they are impossible. But Hume is mistaken. The idea of a special science which gives a critique of
pure reason, will supply the principles of a priori knowledge. Such knowledge is transcendental
because it is not occupied with objects but with how it is possible for us to know about objects,
insofar as this mode of knowledge is to be possible a priori.
We need first a doctrine of the elements of pure reason and second a doctrine of the method.
And there are two sources of human knowledge, sensibility and understanding, and I guess by those
he means something like perception and rationality, or sensation and rationality. Through
sensibility objects are given to us, that is we can, for example, see them; but through the second,
rationality, we can think about them.
Transcendental Aesthetic
This section begins with the introduction of a series of fundamental notions which are very
poorly defined and I will come back to those later. The basic argument of the section concerning
space is that space cannot be an empirical concept derived from outer experience because in order
that a certain sensation be referred to something outside of me, I already have to have a
representation of space. For example I cannot get the idea of space from looking at the desk in front
of me, because in order that I can think that this sensation is a desk which is outside of me, I
already have to have the idea of space. [Notice that he is assuming, with Hume that the sensation
and the empirical desk are the same thing.] It follows that space is a necessary, a priori
representation that underlies all outer intuitions. It cannot be an empirical intuition. Therefore,
geometry can deal with synthetic a priori propositions. Space is empirically real, but
transcendentally ideal. It is nothing at all, except the condition of possibility of there being outer
experiences. But what we experience in space is never the thing in itself. We should not think that
space is something like a secondary quality. Space as a condition of outer objects necessarily
belongs to the world of appearance or intuition.
Time
Time also is not an empirical concept that is derived from experience, because only
presupposing time can we represent to ourselves a number of things as existing simultaneously or
successively. In order that I can think that I had breakfast before I had lunch, succession, or that I
read the paper while having breakfast, simultaneity, I must already have the concept of time. Time
is a rule under which experience is possible. It is a pure form of sensible intuition, the subjective
condition under which alone intuition can take place at all, but it is not a thing in itself.
Space
Just as time is the a priori subjective condition under which any representation at all can
take place, so space is the condition under which intuitions of our outer senses. Time is the pure
form of inner sense, and space if the pure form of outer sense. All objects of the senses, i.e., outer
appearances, such as mountains, trees and desks, are in space. But any appearance at all must be in
time. Thus pains, tickles and itches are in time as much as mountains, desks and trees, but they are
not outer experiences. It is important to keep emphasizing that inner and outer here do not refer to
what exists inside and outside my experiences, because both inner and outer sensations are inside
my experiences. That is what makes Kant an idealist.
This section, The Transcendental Aesthetic, makes clear a whole lot of things that weren't
clear in the preface and in the introduction. Specifically: the ordinary distinction we make between
reality and appearance, between for example, the rainbow which is just an appearance and the
raindrops which are supposed to be real, or between primary qualities which are supposed to be real
and secondary qualities which are supposed to be subjective, is for Kant is at best misleading. All
of these qualities, primary and secondary, raindrops and rainbows, are equally subjective
appearances. None of them are things in themselves. Now, within the realm of subjective
appearances, that is to say, within the realm of empirical reality, we can distinguish the real from
the illusory, but we should not think that the real in this empirical sense has anything to do with
things in themselves. All we can ever know of objects are their appearances. For us, the outer
appearance and the object of the senses, are exactly the same thing. So, the desk I see in front me
has a series of properties, but all of them, properties desk and all, is just appearance. Now the only
reason synthetic a priori judgments are possible, is precisely because the understanding can set
conditions on our being able to have the sorts of appearances that we have. So, in order that we
can even have the sorts of sensible intuitions we have of desks, raindrops, etc., we have to have the
a priori categories of space and time. Furthermore, what applies to so-called material objects also
applies to the inner self. What we think of normally as our soul, or self, or our mind, that too is just
appearance. My mind as far as I can be aware of it is just as much an appearance as is the table or
the chair in front of me. So, there is an empirical reality which consists entirely in appearances.
These appearances are subjective events in the mind but there is also the mind which consists of a
series of appearances. Now, all of these appearances must have a ground, and that ground is the
thing in itself, but of the thing in itself we can know nothing. We can't even represent the thing in
itself to ourselves.
Notice how much Kant departs from Descartes here. For Descartes, we know the mind as it
is in itself by a direct intuition, the cogito; but we know external objects only indirectly by
perceiving their representations in us. For Kant our representations of ourselves is just as remote
from the thing in itself as is our representations of the desk.
Question for Kant: How is publicity possible? How is it possible that you and I can see the
same appearance? Or maybe we can't?
Transcendental Logic
The elements of all of our knowledge are constituted by intuition and concepts. Both may
be either pure or empirical, and both are necessary. Concepts without intuitions are empty.
Intuitions without concepts are blind. This is because the understanding can intuit nothing and the
senses can think nothing. Only through their union is knowledge possible. However, the rules of
sensation are different from the rules of understanding, and the rules of sensation are the
transcendental aesthetic, which we just explained; and now we are going to explain transcendental
logic -- that is the rules of the understanding.
Not every kind of a priori knowledge is transcendental. Transcendental knowledge
concerns only how certain representations, whether intuitions or concepts can be employed or are
possible a priori. Transcendental knowledge, then, is about the a priori possibility of knowledge.
Transcendental logic concerns laws of understanding as they relate a priori to objects. The part of
transcendental logic that deals with the possibilities of knowledge is transcendental analytic.
Transcendental dialectic deals with sophistries and illusions. Transcendental analytic concerns
both the concepts and the principles of pure understanding.
Kant says judgment is the representation of a representation. What I think he means by that
is in every judgment the predicate concept is applied to the subject concept and therefore judgment
is always second order. He says that in every judgment there is a concept that holds of many
representations and among them of a given representation that is immediately related to an object.
(He gives the example: All bodies are divisible.) [This looks like a use-mention confusion. When I
say bodies are divisible I am not saying my concept of bodies is divisible but that bodies
themselves are divisible. He thinks it is not a confusion because of his idealism: all I can talk about
are my representations, so when I say bodies are divisible I am literally saying my representations
are divisible, because the only bodies I can ever have access to or talk about are representations. So
it is not a use-mention confusion.]
He thinks it follows from this that the functions of understanding can be discovered if we
can give an exhaustive statement of the functions of unity in judgments. The reason is that all
judgments are functions of unity among our representations. All knowledge requires synthesis
because synthesis is a matter of putting different representations together and of grasping what is
manifold in them in one act of knowledge. The conditions necessary for knowledge of an object
are first, the manifold of pure intuition, second, the synthesis of the manifold by means of the
imagination; third, the concepts that give unity to the pure synthesis. And these consist solely in
the representation of this necessary synthetic unity. According to the tables, there are the same
number of pure concepts of the understanding as there are types of judgments: quality, quantity,
manner and relation. There are three subcategories under each of these, and with Aristotle, we will
call the whole shebang "categories".
Transcendental Deduction
Some concepts have an a priori use, independently of all experience. The validity of these
concepts requires a transcendental deduction which will explain how they can relate a priori to
actual objects. Empirical deduction relies on experience, transcendental deduction is a priori. We
already have a transcendental deduction of the concepts of space and time. Unlike space and time,
the categories of understanding do not represent the conditions under which objects are given in
intuition. A question now is, how subjective conditions of thought can have objective validity --
that is can furnish conditions for the possibility of knowledge of objects. Causation, for example,
must be a formal condition of sensibility lying a priori in the mind. Causation, as Hume showed,
could never be grounded empirically.
There are only two ways in which synthetic representation and objects can make
connection. Either the object makes the representation, or the representation makes the object
possible. Now, in the first case, the relation is just empirical. I see a desk in front of me, for
example. There are lots of examples of this, but it does not show any necessary condition on the
possibility of knowledge. It is only in the second case that we get a priori conditions on the
possibility of knowledge. Now, the objective validity of the categories as a priori concepts rests
entirely on the fact that as far as the form of our thought is concerned, they are necessary conditions
for the possibility of experience. Experience is only possible as far as the form of our thought is
concerned, given the categories. Only through the categories can any object whatsoever of
experience be thought. So the whole transcendental deduction has to do with the a priori conditions
on the possibility of experience, where the experience is a matter both of the intuition and the
thought.
Hume was correct in seeing that you couldn't get beyond the limits of experience if you
tried to ground the concept of causation. But what he didn't see is that you might ground the
concept of causation in the nature of the understanding itself.
A summary of Transcendental Deduction A: Transcendental Deduction of the Categories A
The categories are pure a priori modes of knowledge, and they are the necessary conditions
of our having conscious knowledge in experience. This argument is designed to explain how they
are possible. There are really two halves to the argument. The first has to do with the three stages
of experience and knowledge, and these three stages are sense, imagination and apperception, or as
he sometimes puts it, perception, imagination and recognition.
The second principle is the Transcendental Unity of apperception. The two extremes are
sensibility and understanding. But in order that they connected with each other, the connection has
to be through the imagination. Now, how does this come about? Well, when I look at a chair, if I
did not have any imagination, all I would have is immediate transitory experiences. But, I can
relate any of these transitory experiences to the present and the past by my imagination. This
exercise of the imagination makes it possible for me to recognize this object and assimilate it under
the concept "chair". So, the sensation on one hand and the understanding on the other have to
mediated by the imagination. But now, furthermore, all of these have to be united in a single
consciousness or there would be no experience at all. This unchanging "I" of pure apperception has
to go along with all of our representations in order that I should be conscious of them as part of one
and the same experience. So, all empirical consciousness must be part of an overriding
transcendental self-consciousness. But now, this apperception has to be added to the imagination in
order that the whole system can function in a conceptual or epistemic or intellectual fashion.
Concepts always belong to the understanding, but they can only be brought into play through the
relation of the manifold to this total unity of apperception. The imagination makes the connection
between the given data of sensation and the conceptual apparatus of the understanding. But in
order that the connection should ever come about, there has to be this transcendental unity whereby
all consciousness are united in a single self-consciousness.
Just to make sure you have the hang of this. I will go through it again. The raw sensations
of the chair do not yet give the understanding that there is a chair there. In order to get that the
sensations have to be connected by the imagination. It takes imagination to unite past, present and
future raw experiences under the concept "chair". But in order that I should achieve this unity of
sensation, imagination, and understanding, the whole sequence has to occur as part of the unified
conscious field. All of my specific consciousnesses, of the chair, the table, the sound of the water
outside have to come to me as part of single unified, self consciousness.
It is at this last point that we introduce the categories. The categories are categories of the
understanding without which the transcendental unity and actual experience would be impossible.
So, it turns out that the understanding really is a faculty of rules. Sensibility gives us the forms of
intuition but understanding gives us rules. Crazy as it sounds, all empirical laws are only special
determinations of the pure laws of understanding, and thus the understanding is the source of laws
of nature, and all of this follows from the fact that we have no knowledge of things in themselves.
In order that we can deal with appearances, we have to have a priori a certain set of
concepts. Those concepts are the categories.
Commonsense Summary of the Transcendental Deduction A
In commonsense terms, the argument is as follows: there are only two ideas in A of the
Transcendental Deduction. The first is that experience is tripartite: there is a sensory component,
an understanding component and these are put together by the imagination in his technical sense of
imagination. So, the way it works is this. If I see that that is a chair over there, then there is a part
that is raw sensation. But in order that the raw sensation should connect with categories like
"chair", I have to be able to associate different raw sensation as the same object. I have to be able
to remember and to project into the future. And that is what he calls "imagination". So, that in
order that I can put it all together under the category "chair" which is an act of the understanding, I
have to connect the raw sensation with the understanding by way of imagination. That's one idea.
Now, the second idea is that in order to have an experience, it has to be part of a unified set
of experiences. There is a unity of consciousness. But, and this is what is important, Kant sees this
unity of consciousness as implying self-consciousness. In order that I can be conscious of the
chair, I have to be conscious of the "I" as the consciousness who is having all of these particular
experiences as part of the single unified consciousness. That is what he calls the "Transcendental
Unity of Apperception." Now, the way these come together in the deduction of the categories is that
the categories are categories of the understanding and there are conditions under which the
transcendental unity of apperception in the third part of the tripartite character of experience can
actually experience as objects. All concepts for Kant are rules and these are general rules.
The Transcendental Deduction B
Categories are concepts which prescribe laws a priori to all appearances. Since these
appearances are nature, the categories prescribe laws to nature itself. The only way that this is
possible, is that nature consists in appearances and these appearances only exist in us and
consequently must agree with our understanding in its a priori form. Now, the way in which the
understanding explains the categories has to do with the combining of the manifold of experience
into a unified consciousness, into the transcendental unity of apperception. The appearances exist
only in us, and consequently, the laws of the appearances exist only in us, in so far as we have
understanding. We cannot think an object except through the categories and we cannot know an
object except through intuitions which correspond to the categories.
The key device in the Transcendental Deduction of the Categories is the a priori synthetic
unity of apperception, the unity of consciousness. The unity of experience is not given by
experience itself, but it is produced by the faculty of the understanding. The categories are
necessary for this production.
This unity is marked by the fact that "I think" can accompany all of my representations. My
thought is composed of my representations and my thought in the form of "I think" can always
accompany my representation. Hence, my representations must conform to the condition that they
can all stand combined together in one universal self-consciousness. Only in so far as I can unite a
manifold of representations in one consciousness, is it possible for me to represent to myself the
identity of my consciousness throughout these representations. Now, from this principle of the
unity of apperception, it follows that the categories are necessary. Just as the supreme principle of
the possibility of all intuition and sensation was given by space and time, so the supreme principle
of the possibility of experience in its relation to understanding is that the manifold of intuition
should be subject to this original synthetic unity whereby all of my experiences are united into one
consciousness. The unity of consciousness is a condition for all objective knowledge because it is
not just a condition that I have to have in order to know an object but it is a condition under which
intuition has to meet in order to become an object for me. The categories make possible objective
knowledge in this sense. My knowledge can't just be of the form "Bodies feel heavy to me", that is,
if I support a body, I feel an impression of weight. But rather, I must be able to universalize and
objectivize this claim in the form "Bodies have weight", or "Bodies are heavy". All sensible
intuitions, that is, all experiences of the senses are subject to the categories as conditions under
which alone their manifold can come together in one consciousness. That is the basic idea of this
whole deduction: We could not have objective knowledge without unified consciousness, but the
categories are conditions of that unified consciousness.
Digression
The I that thinks and the I that intuits itself are one in the same and they are both appearance
and not things in themselves. We can intuit ourselves only as we are inwardly affected by
ourselves. Hence, even of ourselves, we have no knowledge of things in themselves. I have no
knowledge of myself as I really am, but really as I merely appear to myself. Categories make
experiences possible.
The only plausible example, at least the only plausible example that Kant gives of how the
Transcendental Deduction of the categories is supposed to work, is the example of water and ice. If
I experience or see water turning to ice or ice turning to water, I can only have this experience
under the aspect of causation, that is the category of causation is essential to the form of perception
that I have of this phenomenon.
Transcendental Doctrine of Judgement
We now need to show how the pure concepts of the understanding derived in the
Transcendental Deduction can be applicable to empirical appearances. In order that the categories
be applicable to the appearances, there must be some mediating representation. It has to be pure,
that is free of empirical content, but at the same time, it is got to be sensible. That is, it is got to fall
under the aspect of sensation. This is the Transcendental Schema. In order to understand this
schema, we have to use the notion of imagination, and indeed it is schemata that underlie our pure
sensible concepts. The only way that we can get the pure concepts of understanding to relate to
empirical objects, that is to say, appearances is by way of schemata.
The highest principle of analytic judgments is the principle of self-contradiction. That is, it
cannot be both the case that P and not P. The highest principle of all synthetic judgments is a
bigger deal. Experience rests on the synthetic unity of apperceptions. If there were not such
synthetic unity, then experience would just be a random rhapsody of perceptions and it would not
fit according to the rules. So, the highest principle of all synthetic judgments is that every object
requires the synthetic unity of the manifold of intuition in a possible experience in order that it can
exist at all. So, you get these three features in order to have synthetic A priori judgments. First, the
formal conditions, a priori of having intuitions. Second, the synthesis of imagination, and third, the
necessary unity of this synthesis in transcendental apperception. And we relate these three to the
possibility of empirical knowledge in general. The possibility of experience is the possibility of
objects of experience and that is why they have object validity as synthetic a priori judgments.
The first analogy
The amount of substance in the universe is constant. Any change in appearances doesn't
change the constancy in the amount of substance. The proof is that change requires time but time
itself doesn't change, therefore there must be found in the objects of perception appearances the
substratum that represents time in general. [This is a dreadful argument. Almost as bad as all those
arguments that go back to Aristotle. The claim that in order that anything can change, something
must remain constant.]
The second Analogy
All change must take place in accordance with the law of causation. Experience itself is
identical with empirical knowledge of appearances. But experience is only possible if the sequence
of appearances is subject to the law of causation.
We saw earlier that all change is merely alteration. The underlying substance remains the
same. Now, with that as a reminder, we can prove that these alterations must take place in
conformity with the law of the connection of cause and effect. When we see appearances, we
connect them together in time, but we cannot perceive any such connection, it is only because of
the imagination that we can connect different states in time. In order that this relation be objective
they must be related by cause and effect. So, experience as such, that is empirical knowledge of
appearances is only possible insofar as we bring the succession of appearances under the law of
causality.
For example, when I see a house I have a manifold of appearances which occur in time
successively. But the house isn't successive. Now the house isn't a thing in itself but only an
appearance. So how do I connect the successive manifold of appearances with the appearance of
the house itself which is nonetheless is nothing in itself? So, the appearance which is the real house
can be distinguished from my appearances only if the real house appearance stands under a rule that
distinguishes it from mere appearances and necessitates a particular mode of appearance in the
manifold. The objective succession of appearances has to be distinguished from the subjective. The
distinction is that the objective falls under a rule of cause and effect. This is because in the
objective sequence there has to be a necessary relation between the preceding and the succeeding.
The order has to be irreversible. In order for any event to actually happen, it has to happen
according to a rule.
This differs from the standard view according to which we derive the rule from the
succession of our appearances. Kant is arguing, on the contrary, the sequence of experiences is
possible only because of the rule. The rule is the condition of the synthetic unity of appearances in
time. The rule is the a priori condition of possibility of the experiences. Objectivity is possible
only because the representations are necessitated in a certain time order. When I perceive
something happen, I have to perceive it as being preceded by something else. Two things follow
from this: first I can't reverse the series, and secondly, the sequence is necessary. Thus, the
principle of sufficient reason is the ground of all possible experience, and thus the ground of all
objective knowledge of appearances as regard their relation in time.
We can summarize this by saying that the relation of cause and effect is the condition of the
objective validity of empirical judgments at least as far as the series of perceptions goes. The
causal law is the condition of possibility of experiences.
It is, by the way, not an objection to this to point out that sometimes cause and effect are
simultaneous. The point is a point about the order of time not about the lapse of time. The lapse of
time may be zero and yet the order is necessary.
Causality leads to the concept of action, and that to the concept of force, and that to the
concept of substance. All four concepts hang together as part of the world of appearance.
We can summarize this whole second analogy by saying that time is the condition of
possibility of sequential experiences but causation is the condition of the possibility of the objective
ordering of apperceptions and that objective ordering is just the time ordering. So time presupposes
causation.
Third analogy
Substances insofar as they are perceived to co-exist in space are in community or
reciprocity. In order that different substances should co-exist simultaneously they have to stand in
relations of mutual interaction.
All three analogies are about how the unity of nature as regards time connects to the unity
of apperception. All appearances lie and must lie in one nature because without this a priori unity,
no unity of experience, and therefore no determination of objects, would be possible.
The refutation of idealism
The refutation has to show that we actually have experiences of the external world and not
merely imaginations. This will be shown by showing that our inner experiences are possible only
on the assumption of outer experiences. The consciousness of my own existence proves the
existence of objects in space outside me.
My consciousness of my own existence proves the existence of objects in space outside me.
This is proven as follows: I am conscious of my own existence as having a determinate position in
time. But all determination in time presupposes something permanent in perception. This
permanent can't be something in me, because it is only relative to that that I can have my
determinate position in time. So, it is only if there is a thing outside me and not just a
representation of a thing outside me, but a thing outside me, that my determinate position in time is
possible. It is possible only through the existence of actual things that I perceive outside me. So, in
other words, the consciousness of my existence is at the same time an immediate consciousness of
existence of material objects, or at least of objects outside me.
It doesn't follow, of course, that all my representations of outer things are adequate. Some
of them may be illusory or imaginary.
The bottom line on the postulates of empirical thought, is that all principles of the pure
understanding are nothing more than a priori conditions of the possibility of experience. The
possibility of a priori synthetic propositions rests entirely on this relation.
Phenomena or Noumena: The distinction of all objects into phenomena or noumena.
The logical employment of concepts is possible only with regard to appearances. A
transcendental employment of concepts to things in themselves is impossible. The proof is this: an
object can be given to a concept only in intuition. So all concepts, and thereby all principles, even
a priori concepts and principles, have to relate to empirical intuitions, that is to the data for a
possible experience. Apart from this relation, they have no objective validity.
The concepts and principles are a priori but their employment and their relations to
professed objects can only be an experience. So principles such as between two points there can
only be one straight line, or space has three dimensions, though they are a priori principles can only
be applied to empirical reality by way of experience. And this is true even though they contain the
formal conditions of the possibility of experience.
So, it follows that the pure concepts of the understanding never admit of transcendental, but
only of empirical use. And the same goes for the principles. They are never applicable to things or
things in general without regard to the mode in which we are able to intuit them. That is, they have
to be applicable to the objects of the senses. So all that understanding can achieve a priori is to
anticipate the form of possible experience in general. You could put the matter like this: the pure
categories apart from the fact they are formal conditions of sensibility, have only got a
transcendental meaning. Still, they can't be used transcendentally. That's impossible. Now they are
pure categories, so you can't employ them empirically, but we just saw that you can't employ them
transcendentally, they are nothing but the forms of thought which contain the merely logical faculty
of uniting a priori, the manifold given in intuition.
The bottom line of this whole discussion is that the most the understanding can achieve a
priori is to set the form of possible experiences. Now, since only appearances can be the object of
experiences, the understanding can never go beyond the limits of sensibility within which objects
can be given to us. They can't go beyond experience.
The senses represent something to us only as it appears. The senses have access only to
appearances or phenomena. If this is so, then this "something" must also in itself be a thing, and
thus an object of a non-sensible intuition, that is of the understanding. It is through the
understanding, objects would be represented as they really are, and not just as they appear. But we
only have access to objects through sensation. We have no access through the understanding. So
this "something" the thing in itself, equals x of which we know nothing and given our present
subjective constitution we can know nothing whatever. Understanding, employing the a prior
categories, combines the manifold of sensation into the concept of an object; and we have to
postulate this transcendental object, we cannot separate it from the sense data, because nothing
would then be left through which it might be thought. Sensibility is limited by the understanding in
such a way that it can have nothing to do with things in themselves but only with the mode in
which they appear to us and this is due to our subjective constitution, the way we are made. Still,
something must correspond to appearances which is not itself an appearance. This is the thing in
itself or the transcendental object. Thus in addition to phenomena there are noumena.
We can say, but only in a negative sense, that objects can be divided into phenomena and
noumena; a world of the senses and a world of the understanding. There are concepts which are
sensible and those which are intellectual. However we cannot say any of this in the material mode
in a positive sense, because we have no access at all to noumena, to a world of the understanding.
Rather we have a set of categories given to us a priori which enable us to organize the manifold of
sense data under the concept of object and under the other concepts provided to us by the a priori
categories. In addition to the empirical employment of the understanding there is no possible
transcendental employment, no purely intellectual employment. The objects of pure understanding
will always remain unknown to us. So the combined faculties of understanding and sensibility can
determine objects only when they are employed in conjunction. If you separate them you have
intuitions without concepts or concepts without intuitions. Neither can be made to apply to any
object.
Concepts of Reflection: The Transcendental Dialectic
The dialectic is a logic of illusion. And the study of the dialectic is going to be about
exposing the illusion of transcendent judgments. Pure reason is the seat of transcendental illusions.
Now, in order to make that clear we have to make clear the distinction between understanding and
reason. Understanding is a faculty that secures the unity of appearances by means of rules; reason
is a faculty that secures the unity of the rules of understanding under principles. Now, because of
this reason can never apply directly to experience or to any object but only to understanding. It
does that in order to give to the manifold knowledge of understanding an a priori unity by means of
concepts. That is called the unity of reason. The way these illusions happen is that we mistake the
subjective necessity of the connection of our concepts for an objective necessity in the
determination of things in themselves. We mistake in short, features of concepts with features of
an underlying reality in itself. We cannot prevent this illusion. It is unavoidable. But at least we
can expose it.
Now, in order to get on with this we have to make clear the nature of ideas in general. We
call the pure concepts of understanding categories, so we need a new name for the concepts of pure
reason, we'll call them "transcendental ideas". Of course, "ideas" is used in a lot of different ways
in philosophy and Kant digresses to talk about Plato for several pages here. But the basic account
is this: There is a taxonomy that works as follows. The highest genus is a representation. Now,
representations with consciousness are called perceptions. And a perception that relates solely to
the subject (the self) is a sensation. So, I guess, seeing a cow is a perception, but feeling a pain is a
sensation. An objective perception is knowledge. That's if I really do see the cow, and knowledge
can be either an intuition or a concept. Now an intuition relates immediately to the object and is
single. So, I see that very cow. Concepts refer to the cow immediately by means of a feature that
several things have in common. That's when I use the word "cow" to describe it. And the concept
can be either an empirical or a pure concept. Pure concept that has its origin in the understanding
alone is called a notion. Now, a concept that you get from notions but transcends the possibility of
experience is an idea or a concept of reason. Now that is the definition of idea for Kant's purposes.
So, on this definition for example, it is ridiculous to talk about the representation of the color red as
an idea. That's not the way the word is being used at all.
In this account of ideas, an idea is a necessary concept of reason to which no corresponding
object can be given in sense experience. So the concepts of reason are transcendental ideas.
Concepts of pure reason. These are not arbitrarily invented but are imposed by the nature of reason
itself. But they are transcendent in that they overstep the limits of all experience. So, for example,
the absolute whole of all appearances is only an idea because we can never represent it as an image.
Now, because these ideas are only ideas it doesn't mean that they are superfluous or useless. On the
contrary, the ideas of practical reason are very valuable. But we are going to leave them on one side
and consider only speculative or theoretical reason in its transcendental employment.
Reason is the faculty of inferring. So, reason arrives at knowledge by means of acts of
understanding.
The Transcendental Ideas
By the notion Transcendental idea, Kant means a necessary concept of reason to which no
corresponding object can be given in sense experience. These ideas are not arbitrarily invented but
are imposed by the very nature of reason itself. So, for example, the idea of the totality of all
experiences is only an idea. In the system of transcendental ideas, there are three kinds. First the
idea of the unconditioned unity of the thinking subject. Second, the absolute unity of the series of
conditions of appearance, and third, the absolute unity of the condition of all objects of thought in
general. So, on this account, pure reason furnishes the idea for a transcendental doctrine of the
soul, for a transcendental science of the whole world and finally for a transcendental knowledge of
God. Pure reason never relates directly to object but to the concepts which the understanding
frames in regard to objects.
The dialectical syllogisms of pure reason are not rational but pseudo-rational. There are
only three kinds of dialectical syllogism. First, what Kant calls the "parallogism". In this we
conclude from the transcendental concept of the subject, the absolute unity of the subject itself of
which we have no concept whatever. The second kind Kant calls the "antinomy of pure reason"
and this has to do with the fact that the concept of the unconditioned synthetic unity of the whole
series of conditions for any given appearance is self-contradictory. The third kind he calls the
"ideal" of pure reason and this is a pseudo rational inference from the totality of conditions under
which objects in general have to be thought to the absolute synthetic unity of all conditions of the
possibility of things in general. We will consider these in order.
The Parallogisms of Pure Reason, Edition A: The First Parallogism
The pseudo-rational processes of the parallogisms depend on the application of the "I think"
in a way that is supposed to be independent of experience. Thus, we arrive at the following
doctrine: 1) the soul as substance, 2) it is simple, 3) it has unity, 4) it is in relation to possible
objects in space.
These four relations give us four parallogisms of transcendental psychology. 1. We do not
and cannot have any knowledge whatsoever of any substance called the soul. Consciousness,
indeed, makes representations to be thoughts and therefore in consciousness all our perceptions are
found. Consciousness is the transcendental subject of perception but beyond its purely logical
meaning of "I" we have no knowledge of any subject in itself. So, we cannot logically conclude to
the immortality of the soul.
The Second Parallogism
There is a powerful argument for the unity of the thinking subject and that is in order to
constitute a thought, the many representations must be contained in the absolute unity of the
subject. But, we cannot derive this from experience and it is not analytic. Nor is the simplicity of
myself as a soul inferred from the proposition, "I think" because it is already involved in every
thought. The proposition, "I am simple" has to be regarded as an immediate expression of
apperception. The proposition that I am simple really doesn't mean anything else than that in the I
think, the "I" doesn't contain any manifold. This claim could only be useful if we could distinguish
this thinking subject from all matter. But simplicity isn't enough to distinguish the soul from
matter. If we consider matter, as we ought to, as mere appearance. Matter is not a thing in itself.
Matter is just outer appearance, the substratum of which can't be known.
So, the whole of rational psychology is involved in the collapse of its main support. We
can't find the simple nature of the soul in any experience, so there is no way of attaining to it as an
objectively valid concept.
Third Parallogism
The third parallogism of personality. If I view myself from the standpoint of another
person, he will not be able to judge me as I judge myself as having an identity of consciousness
through time. The identity of consciousness of myself at different times is just a formal condition
of my thoughts and their coherence and it does not prove the numerical identity of my subject.
Fourth Parallogism of Ideality
Descartes attempts to justify the inference from the perception of my inner states to an
external world as cause is always doubtful because any effect may be due to more than one cause.
So it is always going to be doubtful whether the cause of my inner states is internal or external.
I will call "idealists" therefore, not just those who deny the existence of external objects, but
those who refuse to admit that the existence of external objects is known through immediate
perception and that therefore, we can never be completely certain about their reality. Now, in order
to understand this we have to distinguish between this kind of idealism, empirical idealism, and
transcendental idealism. Transcendental idealism is the view that appearances are representations
only, not things in themselves, and time and space are sensible forms of intuition, not
determinations given as existing by themselves, nor conditions of objects viewed as things in
themselves. This is opposed to transcendental realism that thinks of space and time as something
given in themselves independent of our sensibility.
The transcendental idealist may be an empirical realist, for he is called a dualist. For the
transcendental idealist who is an empirical realist, matter is a species of representation. Material
objects are external not because they stand in relation to objects which are in themselves external,
but because they relate perceptions to the space in which all things are external to one another even
though the whole of space itself is in us.
I am conscious of my representations, therefore, both I and the representations exist. But
external objects are just representations. They are just representations because they are mere
appearances. So, external things exist as well as I myself. The representation of myself as a
thinking subject belongs to inner sense, while the representations of outer things belong also to
outer sense. I do not need to infer the existence of the house or the table any more than I need to
infer my own existence, because in both case the objects are nothing but representations and the
immediate perception of each is a sufficient proof of their reality.
Descartes was a transcendental realist and an empirical idealist. Kant was a transcendental
idealist, and therefore an empirical realist. Because Descartes' view regards the object of outer
sense as something distinct from the senses themselves. Appearances are representations in us,
hence we are immediately consciousness but objects just are appearances. (These pages are crucial,
pages 345, 346, 347).
We call tables and chairs external objects because they are represented as being in space as
opposed to pains which are inner objects because they are represented as being only in time, but the
point, and the whole point is this: neither space nor time exist except in us. So the expression
"outside us" is ambiguous. It can mean either the thing in itself, or it can refer to outer
appearances. Outer appearances are things found in space.
All outer perception yields immediate proof of something real in space, so empirical realism
is beyond question.
The distinction between imagination and reality, then, is this: whatever is connected with a
perception according to empirical laws is actual.
You can know a lot about the real world a priori but you can't know anything about the self
a priori. You can't know it from the concept of a thinking being. The reason is this. Both are
appearances, but the appearance of outer sense has something fixed or permanent that supplies a
substratum but when it comes to "I" it is the mere form of consciousness and it is not an intuition.
Thus the whole of rational psychology fails.
If you remove the thinking subject, the whole physical world vanishes. The physical world
is nothing but the appearance in our sensibility of our subject and a mode of its representations.
(p.354). Matter is not a kind of substance distinct from the soul, it is just the distinctive nature of
certain appearances of objects which are in themselves unknown to us. The representations of
these we call "outer" as compared with those that we count as belonging to "inner" sense, but like
everything, all of these representations belong only to the thinking subject.
Now, we are deluded about this because we think that these because these representations
represent objects in space, they represent something apart from us, they appear to hover outside us.
But the very space in which they are intuited is nothing but a representation.
There are only three solutions to the mind-body problem, and they are hopeless. One is
physical influence, the second is predetermined harmony, and the third is supernatural intervention.
Now all of these are mistaken and they all have a common origin. Namely, that the assumption that
matter as such is not an appearance, but is the object in itself as it exists independently of all
sensibility. This is wrong. What we should say is that matter consists entirely of appearances, that
is representations in the mind. So, the real question about the causal relations between the thinking
and the extended comes down to this. How in a thinking subject can outer intuition be possible?
But that question no human being could possibly answer. We have no way of knowing on what the
outer appearances depend.
There is a general pattern to these illusions. It consists in treating the subject of condition of
thinking as being knowledge of the object.
Reason is the faculty of principles. It mistakenly represents several things as unconditional
conditions of the soul. Specifically, 1) it is self-subsistent, 2) simple, 3) unified, and 4) aware of
itself in a way different from its awareness of other things merely as its representations. All of
these are parallogisms and thus illusions of pure reason.
The Antinomy of Pure Reason
Just as the parallogism of pure reason gave us a lot of illusions about psychology, so the
antinomy of pure reason will exhibit to us the transcendental principles of a pretended rational
cosmology that is a theory of absolutely everything. Now, when we think about the system of
cosmological ideas, there are two things to keep in mind: First, these pure and transcendental
concepts can only accompany understanding. Reason by itself never generates any concepts. The
most reason could do is to free a concept of the understanding from the limitations of possible
experiences and so, to try to extended beyond the limits of the empirical. The illusion that
generates this move, is that if the conditioned is given, then the entire sum of conditions and hence,
the absolutely unconditioned, is also given. So, the transcendental ideas in the first place are just
categories extended to the unconditioned.
On the second place, not all categories are fitted to this employment, but only those in
which the synthesis constitutes a series of conditions subordinate to one another and generative of
some given conditioned phenomenon. For example, we necessarily think that time has completely
elapsed up to a given moment, and as being itself given in this completed form. So we think any
element time t (?) is conditioned by t - n but not as conditioned by t + n. Now, call the synthesis of
the series that goes backwards, the regressive synthesis, and the series that goes forward, the
progressive synthesis. The cosmological ideas deal with the totality of the regressive synthesis
preceding in antecedence not in consequences. Now, because time is ordered it has both a
progressive and regressive synthesis, but that is not true of space. However, even for space, the
measuring of the space is a synthesis of a series of conditions. But in space it looks like regress and
progress are the same, because one part of space is not given through the other parts but only
limited by them. We have to consider each space insofar as it is limited as being also conditioned
in that it presupposes another space as a condition of its limits and so on.
Second, reality and space, that is matter, is a condition. Its internal conditions are its parts
and of course, the parts have parts of parts and those are even remote conditions.
Third, the idea of substance is not a transcendental idea because it doesn't involve a notion
of conditions.
Fourth, the same goes for the modal notions of possible, actual and necessary. They do not
lead to any series of conditions.
So, if you pick out those categories that necessarily lead to a theory in the synthesis of the
manifold, there are four of these cosmological ideas.
1. Absolute completeness of a composition of the given whole of all appearances.
2. Completeness in the division.
3. Absolute completeness in the origination of an appearance, and absolute completeness.
4. Absolute completeness as regard to dependence of existence of the changeable in the
field of appearance.
Now, we are going to get to the antimonies of pure reason. Now what these are a set of
opposed propositions, a thesis and an antithesis, and these are illusions that are natural to us---
illusions generated necessarily by reasons. Now these dialectical illusions come not from the unity
of the understanding but from the unity of reason in mere ideas. The unity of reason involves a
synthesis according to rules, but it has to conform to the understanding. Now, when it is adequate
to reason, the unity is too big for the understanding but when it is suited to the understanding it is
too small for reason, so you get a conflict, and you can't avoid the conflict.
In order to understand all this stuff, we are going to use the skeptical method, and that
consists in trying to discover the point of misunderstanding that underlies these disputes.
So, here we go. The first antimony is this. Thesis: the world has a beginning in time and is
limited to space. Antithesis: the world has no beginning and no limits in space. It is infinite in
both ways. The problem with both thesis and antithesis is that you can get very powerful
arguments for each view and very powerful arguments against each view.
Second antinomy. Thesis: every composite substance is made up of simple parts and
nothing really exists except the simple parts and what is composed of simple parts. Antithesis: no
composite thing is made up of simple parts and in fact, there doesn't exist anything in the world
which is absolutely simple.
Third antinomy. Thesis: the causality of the laws of nature is not the only kind of causality
needed to explain appearances in the world. We also need the causality of freedom. Antithesis:
there is no freedom. Everything in the world takes place solely in accordance with the laws of
nature.
Fourth antinomy. Thesis: The world contains a necessary being. Antithesis: there is no
necessary being in the world, nor does such a thing exist outside the world as its cause.
Reason is necessarily led to these antinomies because it tries to free all the conditions things
which can never be determined except as conditioned. It tries to treat the conditioned as if it were
unconditioned and that is where it gets into these big cosmological ideas. It's the usual thing where
reason tries to extend its domain beyond the limits of experience. So, in these cases reason ought to
consider whether or not these arguments didn't come out of the misunderstanding. All the theses,
as opposed to the antitheses presupposed not only the empirical mode of explanation but also that
everything has an intelligible beginning. I called this the dogmatism of pure reason. That the
world has a beginning, that my thinking self is simple and therefore indestructible, that it is free and
raised above the compulsion of nature, and finally that everything in the world is due to a
primordial being---these are the foundation stones of morality and religion. Now, the antithesis
robs us of all these supports, or at least it seems that it does.
So, we have a speculative interest on the side of the thesis because you can grasp the
arguments for the thesis completely a priori, but that is not so with the antithesis.
Third, the thesis also has the advantage of popularity, but on the empiricist side, the
antithesis doesn't have the practical interest and certainly is not going to seem very popular because
you wind up apparently denying the existence of God and freedom. Though empiricism does have
an advantage in the speculative interest in reason because if you follow the empiricist precepts, then
understanding is always on its own proper ground namely, genuinely possible experiences and
empiricism wants to investigate the laws of experiences. Now, if the empiricists were just satisfied
with this. If he were just saying, "look, do not extend reason beyond the limits of experience" then
he would just be urging modesty. But the problem is, empiricism frequently becomes dogmatic
and it confidently denies what lies beyond the sphere of its intuitive knowledge. But that is just as
immodest as the anti-empiricist. And it is actually a bad idea because it has bad practical
consequences. So this antithesis between empiricism and its enemies is really the same as the
opposition between epicureanism and platonism.
In the end, each kind of philosophy, the empiricist and the platonist says more than it
knows. The anti-empiricist tries to prove the existence of God, freedom and immortality, and the
empiricist, which is in its way just as bad, tries to deny the existence of God, freedom and
immortality.
Empiricism is always going to be unpopular and it is never going to get influential with the
mass of people.
Section IV: The necessity of the solution to the transcendental problems of pure reason
Now, since in these question we are dealing with these things as objects of possible
experiences, not as a thing in itself, we have to find a solution to our problems in the idea. We are
not asking about the constitution of any object in itself, nor are we asking about what can be given
in any experience. Our only question is to what lies in the idea. Now, since the idea is invented by
reason, it is the creature of reason, reason can't just walk away from it. All these questions are
really referring to an object that exists only in our thoughts, namely the absolutely unconditioned
totality of the synthesis of appearances. That is what all these antinomies are about. As long as we
think that there is an object that corresponds to the idea, the problem has no solution. Now, there is
no way we can get empirical knowledge of the complete synthesis and consciousness of any
absolute totality of our intuition. And of course, we are never going to get any answers to these
questions from the thing in itself.
So far, the antinomies of pure reason can be summarized as follows: reason has a natural
urge to generalize about totalities, the totality of the sequence of cause and effect, the totality of
events in time, the totality of space, etc. The problem is, that all of these urges to generalize are
meaningless because we can never have any experiences corresponding to these totalities. They
naturally ? antimonies because on the one hand, we want to give some platonic interpretation of
the totalities that leads us to the conclusion that God exists (there is a first cause), that the Universe
has a beginning in time etc., and all these are the various theses. Now, in response to these, the
empiricist's antithesis tries to point out that we cannot draw these conclusions, however the
empiricist in the end becomes just as dogmatic as the platonist when he says that God does not exist
or we do not have freedom.
Section V
We should examine the transcendental ideas critically because both sides of the antinomies,
the thesis and the antithesis, are nonsense. This is because they rest on a false presupposition. In
every case, when I try to think of the unconditioned, the result is either going to be too big or too
little for the understanding. For example, if the world has no beginning, then it is too big for our
concepts. If, on the other hand, if the world has a beginning, then it is going too small for the
concept of the understanding. Now, this same is true of the other cases. If space is infinite, then it
is too large; if it is finite, then it is too small. The same goes for the idea of infinitely many parts,
and with the infinite sequences of cause and effect, and fourth and finally, if we admit an absolutely
necessary being and set it at a time infinitely remote from any given point in time, then such an
existence is too large for our empirical concepts. But otherwise, it is going to be too small.
Now, the reason that we are getting this sort of nonsense in every case is that a possible
experience is the only thing that can give reality to our concepts. In the absence of a possible
experience, these concepts are mere ideas, they have no truth, they are without any relation to any
real objects.
Section VI
Now, the way we are going to solve all this stuff is by transcendental idealism. Space, time
and causation and all objects are nothing but appearances in the mind, nothing but representations,
and they cannot exist outside our mind. Even our awareness of our own mind is just an appearance.
It is not the self as it is in itself. It is not a transcendental subject.
All the same we can distinguish truth from reality. We can distinguish real things from
dreams because dreams and genuine appearances cohere truly and completely in one experience in
accordance with empirical laws. Real things are real because they stand in an empirical connection
with empirical laws and with my actual consciousness. But to repeat, that does not mean that they
are things in themselves. Our sensibility is just a receptivity. Now, the non-sensible cause of these
representations is completely unknown to us. [This is the first time that Kant has said that
appearances are caused by things in themselves. Antinomies, p.441. On his own terminology, he
is not allowed to say that. But he repeats the mistake twice on page 441]. Now, since objects are
only appearances, what does it mean to talk about real things, for example, in the distant past?
Here's the answer: they are objects for me in real and past time only insofar as I represent to myself
that a regressive series of possible perceptions in accordance with empirical laws leads us to a past
time series as a condition of a present time. But the series cannot be represented as actual in itself,
but only as a connection to a possible experience. So, the whole idea of talking about a past that
existed before I did means nothing except that there is a possibility of extending the chain of
experience from the present perception back to the conditions that determine this perception in
respect to time. All it means to talk about objects existing before I existed is that if I project the
sequence of actual experiences beyond the actual to the merely possible, then I get to the idea that
there are possible experiences that I have never actually had. Say that they exist prior to my
experiences just to say that if I project this thing from my present perception back into the past, I
eventually get to the part of the experience to which the possible experiences belong. Now, what
the causes of that could be, I can't say, because that would things in themselves. It is necessarily
unknown to me.
Section VII, the solution of the cosmological conflict
Now, all these antinomies rest on a dialectical argument; if the conditioned is given, then
the entire series of all the conditions is likewise given. So, if the object of the senses is given to
me, for example, the chair I see in front of me, then all of the entire series of its conditions is given.
Then that leads us to postulate an absolute totality, an absolute totality of space and time, of cause,
etc., and that is where we go wrong. As far as the understanding is concerned, it is an analytic truth
that if there is a conditioned, then the entire series of its conditions is already implied. The problem
is when you get to appearances. You cannot say in that same sense that if the condition of your
appearance is given, then all of its conditions as appearances are likewise given. Appearances are
just an empirical synthesis in space and time and they are given only in the synthesis. So, it doesn't
follow that if the conditioned in the field of appearance is given that the synthesis that constitutes
its empirical condition is presupposed, is also given.
The result is that there is no way to settle these issues. Each side can refute the other
decisively, but that is only evidence that they are quarrelling about nothing. That the
transcendental illusion is mocking with a reality where none is to be found. So the key to the
dialectic is this: the apparent inconsistency of asserting a proposition and its negation is only an
illusion. In each case, there is a presupposition, and the presupposition has to do with things in
themselves. He gives the following analogy which is a pretty good one: Suppose somebody says
that all bodies are either good smelling or not good smelling? Well, those aren't the only
possibilities. Each of those has the presupposition that bodies have a smell. But some bodies may
have no smell at all, in which case, you do not have a genuine contradiction. Either this body
smells good, or it doesn't smell good. Now, similarly, if we say that the world is either finite or
infinite, both statements might be false. Because we might be regarding the world as itself
determined in its magnitude, but in that case, we may be presupposing about how things are
actually in themselves and we cannot do that. So, where you have a dialectical opposition between
two such judgments, they are not genuine contradictories, because they both may be false. Since
the world does not exist in itself independent of the regressive series of my representations, it exists
in itself neither as an infinite whole nor as a finite whole. It exists only in empirical regress in the
series of appearances. It is not something in itself.
The world is not an unconditioned whole, and hence it is neither infinite nor finite. This
solution applies to all the other antinomies as well. The idea of an absolute totality applies only to
things in themselves, not to appearances. But we cannot say anything about things in themselves.
Another way to see this whole bunch of issues is to see that the antinomies are an indirect proof of
the transcendental ideality of appearances. If the world as a whole existed in itself, then it would
either be finite or infinite. Both alternatives are absurd, so the world as a whole does not exist as a
thing in itself. Skipping from page 449 to page 467.
The possibility of Causality through freedom in accordance with Universal laws of nature
In this section we are going to prove not that freewill is real; in fact we are not even going
to prove that it is possible. Rather, what we are going to show is that the antinomy rests on an
illusion and that causality through freedom is not incompatible with natural determination. Now,
here is how the proof goes.
Anything which is an object of the senses but is not itself is an appearance, I call
"intelligible." Now, man is part of the world of appearances, but man has a faculty which is not an
object of sensible intuition and that is reason. Now reason can have a kind of causality in its
actions which is intelligible but also sensible in its effects. Now, as part of the world of
appearances, our actions are completely determined. Every feature is determined. But I, as an
acting subject, in my intelligible character am not in time. Time is only a condition on
appearances, not of things in themselves. But then if reason as an intelligible phenomenon can
cause action it does not do it in time. Therefore, in its intelligible character, the human subject has
to be considered to be free from all influences of sensibility and hence from all determinations.
Insofar as it is a noumenon and not a phenomenon, nothing actually happens to it. The active being
is independent of and free from all such necessity. So where all events in nature are determined as
part of the order of nature, and in fact, they can only be part of the order of nature if they were
determined, nonetheless, one and the same event can be in one aspect merely an effect of nature,
and in another aspect, an effect due to freedom. How is this possible?
Well, granted that effects are appearances, and causes are appearances, is it necessary that
the causality of their cause should be exclusively empirical? Maybe not. Nothing rules out the
possibility that it could be the effect of a causality that is not empirical but intelligible. Now, man
knows the rest of nature solely through the senses, but he knows himself also through pure
apperception. So, man is to himself on one hand, phenomenon, but on the other hand, with regard
to certain faculties such as reason, he is an intelligible object. Now, we know that reason has
causality and that is evident from the categorical imperatives where we use the word "ought.", not
to say how things actually are, but how they ought to be. So it is at least possible for reason to have
causality with respect to appearances. So far as we regard the empirical character of action, there is
no freedom. But when we consider actions in relation to reason, and reason insofar as it is itself a
cause producing action, then we find a conception of order and rule altogether from the order of
nature. Because in the order of nature there is only what has happened, but in the order of reason
there is what ought to happen. So, reason can both have causality and be free, and thus the
condition of a successive series of events may be itself empirically unconditioned. Reason acts
freely because it is not located in time. Reason in its causality is not subject to any such conditions
which are conditions on appearances in time. So both freedom and natural necessity can exist
without conflict in one and the same action.
(This whole account seems totally inconsistent with what Kant says about things in
themselves. How can we talk about the noumenal thing in itself "reason" being a cause of
anything? Causes are part of the world of appearances.)
The Idea of a Necessary Being
The antinomy of a necessary being might be resolved by pointing out that though the
sensible world contains only appearances and these are always conditioned, there might also be a
necessary being as part of the intelligible ground of these appearances, but not part of the world of
appearances. So, it can be both true that there is a necessary being and in the intelligible world and
everything in the empirical world is contingent.
The only way we could ever talk about such a necessary being is by analogy. We have no
knowledge of intelligible things. But we can use our knowledge of the world of appearances to talk
about such things by analogy. (JS compare with Aquinas and his theory of analogical predication.)
In the next section we will talk about a necessary being.
Dialectic. Book 2, Ch 3, The Ideal of Pure Reason
Kant begins this section with the summary of the relationship between concepts, etc. You
can't represent an object like an object like a chair or a house with just the pure concepts of the
understanding, you have got to have understanding as well. But when you stick the concepts of the
understanding onto appearances, then you get concrete objects. Now, when you get to ideas, in his
technical sense of idea, they are further removed from reality than are the categories of the
understanding because no appearance at all can be found in which they can be represented in a
concrete case. Now, even worse off than ideas are ideals, because the ideal is sort of an abstract,
but still concrete thing. So, for example, humanity is an ideal---the idea of a perfect human. Now,
reason contains not only ideas but also ideals. And these even have a certain practical power.
These ideals have a kind of practical power as regulative ideals. The theory is this: virtue is an
idea, but the wise man is an ideal. The idea gives us a rule but the ideal gives us an archetype.
Leibniz's principle that every particular thing is completely determinate doesn't mean that
for any given bunch of predicates, one must be true and its negation false. But rather, for any
possible predicates, one must be true and its negation false. However, logical negation isn't really a
type of concept. It's just a relation of another concept in a judgment. So, negation by itself is quite
insufficient to determine a concept in respect to its concept [Kant is dead wrong about this. Frege
understood this point much better than Kant. Frege saw that denial is not a different kind of speech
act from assertion]. Kant thinks that all true negations are nothing but limitations. However,
according to Kant, the idea of a completely determined thing, is the idea of a thing in itself. It is
therefore a transcendental ideal. And that serves as the basis for the complete determination that
necessarily belongs to everything that exists.
Now, here is the basic argument for a necessary being. If you think that anything exists,
then you have got to admit that there is something that exists necessarily. Because if the first thing
existed only contingently, then it has got to be contingent on some other existence, and so on until
you get to something that doesn't exist contingently but exists necessarily.
Now, this is the way human reason always works. It persuades itself of the existence of
some necessary being or other. Now, if you feel that you just have to come to a decision, then that
argument looks pretty good. But if you are willing to leave your mind open, then the argument
doesn't look so good. Actually, as far as the argument goes, we are entirely\ free to conclude that
any limited being whatever might be unconditionally necessary. So, the argument doesn't prove the
existence of a necessary being.
There are only three ways to try prove the existence of God by reason. These three ways
are the physical-theological, the cosmological and third, the ontological. There can be no others.
Impossibility of an Ontological Proof
A triangle necessarily has three interior angles. The problem is that there is no
contradiction in denying the existence of triangles altogether. Of course, if you say "if something is
a triange then it has got to have three angles," that is indeed necessary. But it doesn't get you
anywhere if you are trying to prove the existence of triangles. Now, the same thing holds of the
idea of a necessary being. If you reject its existence, then you reject the thing in itself with all its
predicates. And there is no contradiction in that. It is self-contradictory to deny that God is
omnipotent, but it is not self-contradictory to say that there is no God.
All existential propositions are synthetic. Now, existence is not a real predicate. It is not
something that you could add to the concept of a thing so it couldn't follow from the definition of
the concept. For example, the concept, God is omnipotent, contains two concepts, each of which
has its object---God and omnipotence. But if we say that God exists, we attach no new predicate to
the concept of God. The concept of a supreme being is a useful idea, but it is a mere idea. All
these efforts by Leibniz and Descartes, and so on, to try to prove the existence of God by the
ontological argument is just a waste of time.
Impossibility of a Cosmological Proof of the Existence of God
The whole project of trying to prove the existence of something corresponding to some
arbitrarily selected idea is really crazy from the start. The only reason we try to do it is that we
want to prove the existence of a necessary being in order to be able to make sense of ourselves as
contingent beings. We saw this already in the ontological proof, but in the cosmological proof it
gets even worse. Here is how it goes.
If anything exists, then an absolutely necessary being must also exist. Well, I, for example,
exist. Therefore, an absolutely necessary being exists. However, if it is an honest to John
necessary being, then it has to be determined by itself. So, it isn't just any old necessary being, it is
a supreme being.
Now, in a way, this argument is just a variation of the fallacy in the ontological proof.
Because once again, we are just trying to infer existence from a concept. Furthermore, if you try to
state this one in a syllogism, it looks like it is going to prove that every ens realissium is a
necessary being. And this really just is the ontological proof all over again.
Now, this cosmological argument is just a whole mess of dialectical assumptions, and by
now, we are all such experts at these that we can figure them out ourselves. But here are four of
them:
1. It is a transcendental principle to try to infer from a contingent to a transcendental cause.
It makes no sense at all.
2. You try to infer a first cause from the alleged impossibility of an infinite series of causes,
but we cannot justify this.
3. Reason here mistakenly supposes that it is given the completion of the concept of a
series, just on the grounds that we cannot imagine anything further.
4. Confusion between the logical possibility of a concept of all reality united into one, and
the transcendental possibility of such a reality.
Now, we might want to postulate the existence of some all sufficient being, but it is just
going too far to suppose that we could prove it.
There is an explanation for why we make all these mistakes. Why we have a dialectical
illusion in the transcendental proofs of the existence of a necessary being. The problem is this.
The concept of necessity is only to be found in our reason. It is a formal condition of thought and
we make the mistake of hypostatizing it as a material condition in the real world. The right way to
think of matters, would be to see that the ideal of a supreme being is a regulative principle of
reason. It tells us to look upon things in the world as if they came from some necessary cause. But
the mistake we make is to try take this regulative ideal for some honest to John fact.
Impossibility of a physico-theological proof
This last proof is the one that is the most fun, and is the most popular with the general
public. We should treat it with respect, but we have to face up to the fact that it isn't any good
either.
Once again, it has to rely on the ontological proof. Here is how it goes. First, whenever we
look at the world, we find clear signs of an order. We find signs of purpose. Everything seems to
make sense and to be made for a purpose. But somebody must have designed the world.
Somebody must have put all these purposes in the world. 3) Therefore, there exists a sublime and
wise cause. 4) We can infer that this cause is unified. The cause is namely God.
Now, what is wrong with this proof? Well, you can't just advance to the absolute totality by
going this empirical road. This proof can lead us to admire the greatness, wisdom, etc., of the
author of the world, but it can't get us any further than that. You just cannot prove existence that
way.
In the end this proof has to fall back on the cosmological proof and that is only a disguised
version of the ontological proof.
Critique of all theology based on speculative reason
The upshot of this whole discussion is something we already knew. Namely, transcendental
questions allow only transcendental answers. That is, the only kind of stuff that you can prove a
priori, the only synthetic a priori knowledge is that of formal conditions on possible experiences.
But the problem with all these proofs of the existence of God, is that they are trying substantive.
And something substantive that goes beyond empirical knowledge. Synthetic a priori knowledge
gives us a bunch of principles that apply only to objects of empirical knowledge, to appearances.
All of these attempts to prove the existence of God reduce to the ontological proof. All of
them commit the fallacy of trying to derive the existence of an object from the features of the
concept. But you can't derive the existence of any object just from the concept.
Now, in spite of all that, there is still useful employment for transcendental theology. But it
is purely negative. You cannot prove the existence of God by pure reason, but also, you can't
disprove the existence of God. Yet, but what you get out of this discussion is the ideal of God. It is
an ideal without a flaw. It completes and crowns the whole of human knowledge.
Appendix: The regulative use of the ideas of pure reason
Now, we have seen over and over that reason is never in any immediate relation to any
objects, but only to the understanding. So, reason can't actually create any concepts of objects, but
it just orders them. It gives them a unity. The upshot is that transcendental ideas never have any
constitutive employment. On the other hand, they do have an excellent and indispensable
regulative employment. Namely, they direct the understanding towards a certain goal. What is
peculiar about reason is that it tries to achieve some kind of systematization. It tries to exhibit the
connection of all the parts. For example, our ideals of pure earth, pure water, pure air, etc., are just
ideals. But it is very useful to be able to talk this way if you are going to do natural science.
Now, the systematic unity of the manifold knowledge of understanding as prescribed by

reason is just a logical principle. Its function is to assist the understanding by means of ideas. It
makes our knowledge more coherent.
An example of the operation of reason here is our ability to categorize things according to
genus and species. Just looking at things empirically wouldn't give you our categories of genus and
species, unless it were guided by a transcendental law of specification. Now, that law is a principle
of reason and it leads us always to seek for further differences. And indeed, we always suspect
further differences even when the senses can't perceive them.
So, to summarize all of this, reason prepares the field for the understanding, 1) through a
principle of the homogeneity of the manifold under higher genera. 2) through a principle of
variety of the homogeneous under lower species, and 3) in order to complete this systematic unity
of further law of affinity of all concepts. According to this law, we proceed from each species to
every other by gradually increasing diversity. These three laws are the principles of homogeneity,
specification, and continuity of forms. The last one results from the union of the first two. (This
whole discussion of genus and species is a mess. Kant doesn't understand the causal basis on which
well-motivated taxonomies are constructed).
Reason operates on the understanding in the same sense that the understanding operates on
sensibility. The job for reason is to make the acts of the understanding systematic. Just as the job
of the understanding is to connect a manifold of appearances by means of concepts, and to bring
that manifold under a bunch of empirical laws. However, without the schemata of sensibility, the
acts of the understanding are undetermined. Now, in the same way, the unity of reason is also
undetermined as regards the conditions under which and to the extent to which the understanding
ought to combine its concepts in some systematic fashion. Now, let's call the subjective principles
which arise from interest has that reason has, the maxims of reason. These do not depend on the
objective constitution of objects, but just on the subjective nature of reason. They are maxims of
speculative reason. Now, as long as you remember that these are just maxims, you do not get into
trouble. Now, for example, somebody might be interested in the manifoldness, but somebody else
in the unity. Now, each thinker believes that his judgment is arrived at through some special
insight into the object, but it is just really an attachment on, it is just a matter of different
attachment to subjective principles.
Final purpose of the natural dialectic of human reason
You can't have a transcendental deduction of the ideas of pure reason in the way that we had
a transcendental deduction of the concepts of the understanding, however, we have to have some
kind of deduction of these things if they are any good at all. (Page 551)
I want to make this clearer. What Kant is going to show is that the principles of pure reason
should be thought not as telling us how things are, but as if principles. So, in psychology we have
to act as if the mind were a simple substance, in cosmology--as if there were an infinite sequence of
appearances and in theology---as if there were a supreme intelligence.
Now, in this way if we assume these ideal states, we are not really extending our knowledge
beyond the objects of possible experiences, we are just extending the empirical unity of such
experiences. And we do that by means of the systematic unity for which the schema is provided by
the idea. But the idea (and remember these ideas are just regulative principles not constitutive
principles). Now, the conception of God that we get out of this is in a strict sense, deistic. We
have the right to assume the existence of God in a relative sense, but not absolutely.
The concepts of reality, substance, causality and even of necessary existence, may be used
to explain the possibility of empirical knowledge of objects, but they have no meaning at all as
applied to the universe itself. They cannot explain possibility of the universe itself. And these big
deal concepts really have no application to something other than the world of sense.
Now, here is the upshot. Pure reason is really occupied only with itself. It gives us only
regulative maxims. It does not tell us anything about the world of experience, much less about
things in themselves. So, the systematic unity demanded by pure reason is a mere idea. But that is
ok as long as we do not misunderstand these ideas. I will show how this works for all three
examples.
First. I treat the self as if it were a simple unity and this enables me to make better sense
out of my experiences and it is ok to do that as long as you do not suppose that there is any
meaning to claims such as the claim that the soul in itself is of a spiritual nature.
Second. We ought to treat the series of appearances as if it were infinite. But this is strictly
an as-if principle.
Third. We can treat the whole universe as if it were connected together by a single all-
embracing being as its supreme and all sufficient cause, provided that we restrict ourselves to the
regulative use of these principles, there is not going to be a problem. Even a mistake cannot do us
any serious harm, because, for example, we might expect a teleological connection in the universe
and we find only a mechanical or a physical connection. The big mistake is to take the regulative
principle of the systematic unity of nature as if it were a constitutive principle.
If we ask in transcendental theology whether there is anything distinct from the world which
is the ground of its order, the answer is, there undoubtedly is. The world is a sum of appearances
and there must be a ground of the appearances. But second, if we ask questions about this being is
it substance, necessary, etc., those questions are absolutely meaningless. But let's push the question
further. Can we assume an omnipotent Author of the world? The answer is, not only may we, we
have to, we must. This doesn't extend our knowledge beyond the field of possible experiences. All
we have done is presuppose a something, a transcendental object of which we really have no
concept whatever. Then the question can be asked, can we make use of this presupposition of a
supreme being. And the answer is, yes we can. But only on the condition that we do not care
whether or not the universe has been organized by divine wisdom or whether or not it is just
organized by nature. Because what justified us in adopting the idea of a supreme intelligence is
precisely this greatest possible systematic and purposive unity. So it would be circular to try to use
the one to show the other. In summary, pure reason gives us nothing but regulative principles.
All human knowledge begins with intuitions, precedes to concepts and ends with ideas. But
as far as knowledge is concerned, we can never go beyond possible experiences.

Unconditioned, So If We Thought of Our A Priori Knowledge As Concerning Things in Themselves, We

Uploaded by

Copyright:

Available Formats

Unconditioned, So If We Thought of Our A Priori Knowledge As Concerning Things in Themselves, We

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Unconditioned, So If We Thought of Our A Priori Knowledge As Concerning Things in Themselves, We

Uploaded by

Copyright:

Available Formats

Summary of the Critique of Pure Reason 1

Summary of the Critique of Pure Reason

Now, the systematic unity of the manifold knowledge of understanding as prescribed by

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.