Kenny Anthony. An Illustrated Brief History of Western Philosophy
Подождите немного. Документ загружается.
the critical philosophy of kant
284
principle of non-contradiction will not take us beyond the field of analytic pro-
positions: it is a necessary, but not a sufficient condition for the truth of synthetic
propositions.
In a synthetic judgement two non-identical concepts are put together. Kant
lists four groups of principles which underpin synthetic judgements: he gives them
technical terms, but we need not concern ourselves with these since they are more
confusing than helpful.
The first of these principles is that all experiences are extensive magnitudes.
Whatever we experience is extended – that is, has parts distinct from other parts
– either in space or in time. ‘All appearances,’ Kant says, ‘are experienced as aggre-
gates, as complexes of previously given parts.’ It is this, according to Kant, which
underpins geometrical axioms, such as the axiom that between two points only
one straight line is possible.
The second principle is that in all appearances the object of sensation has
intensive magnitude. For instance, if you feel a certain degree of heat, you are
aware that you could be feeling something hotter or less hot: what you are feeling
is a point on a scale which extends in both directions. Similarly, to see a colour is
to see something which is located on a spectrum. Kant calls this an ‘anticipation
of perception’, but the term is an unfortunate one: it is as if he is saying that
whenever you have a feeling, you can know a priori what feeling is going to come
next. But of course only experience could show that; as Kant says, ‘sensation is
just that element which cannot be anticipated’. When I have a sensation what is
known a priori is simply the logical possibility of similar sensations at other points
upon a common scale. To catch Kant’s sense a better word than ‘anticipation’
might be ‘projection’.
The third principle is this: experience is only possible if necessary connections
are to be found among our perceptions. There are two main stages on the way to
establishing this. (a) If I am to have experience at all I must have experience of an
objective realm: and this must contain enduring substances. (b) If I am to have
experience of an objective realm I must have experience of causally ordered inter-
acting substances. Each of these stages takes off from reflection on our awareness
of time: time considered first as duration, and then as succession.
First, Kant points out that time itself cannot be perceived. In the experience
of a moment, considered simply as an inner event, there is nothing to show
when the experience occurs, or whether it occurs before or after any other given
momentary experience. Our awareness of time, then, must be a relating of phenom-
ena to some permanent, substantial, substratum.
If there is to be such a thing as change (as opposed to mere disconnected
sequence) then there must be something which is first one thing and then
another. But this permanent element cannot be supplied by our experience,
which itself is in constant flux; it must therefore be supplied by something objec-
tive, which we may call ‘substance’. ‘All existence and all change in time have
AIBC16 22/03/2006, 11:06 AM284
the critical philosophy of kant
285
thus to be viewed as simply a mode of the existence of that which remains and
persists.’
There are a number of ambiguities in this argument and its conclusion. It is
not always clear what type of change is being talked about: does the argument
concern the coming to be and passing away of substances, or is it about alteration
in the properties of an enduring substance? Consequently, there is doubt as to
how much is proved by the argument: is the conclusion that there must be some
permanent things, or is it that there must be a single permanent thing? Kant
sometimes speaks as if substance must be something everlasting; but in order to
refute empiricist atomism it is sufficient to show that there must be at least some
objective entities with non-momentary duration.
The second stage of the argument is based on a simple, but profound, observa-
tion. If I look at a house, there will be a certain succession in my experiences:
first, perhaps, I look at the roof, then at the upper floors, then at the ground
floor, then at the basement. Equally, if I stand still and watch a ship moving
down a river I have a succession of different views: first of the ship upstream, then
of it downstream, and so on. What distinguishes between a merely subjective
succession of phenomena (the various glimpses of a house) and an objective
succession (the motion of the ship downstream)? In the one case, but not the
other, it would be possible for me to reverse the order of perceptions: and there
is no basis for making the distinction except some necessary causal regularity.
‘We never in experience attribute to an object the notion of succession ...and
distinguish it from the subjective succession of apprehension, unless when a rule
lies at the foundation.’
This shows that there is something fundamentally wrong with Hume’s idea that
we first perceive temporal succession between events, and then go on to regard
one as cause and the other as effect. Matters are the other way round: without rela-
tionships between cause and effect we cannot establish objective order in time.
Moreover, Kant says, even if temporal sequence could be established independently
of the cause–effect relation, bare temporal succession would be insufficient to
account for causality, because cause and effect may be simultaneous. A ball, laid
on a stuffed cushion, makes a hollow in the cushion as soon as it is laid on it, yet
the ball is the cause, the hollow the effect. We know this because every such ball
makes a dent, but not every such hollow contains a ball. The relation between
time and causation is more complicated than Hume imagined.
Having refuted empiricist atomism and countered Humean scepticism about
causal connections, Kant goes on to present his refutation of idealism. He has
in view a twofold target: the problematic idealism of Descartes (‘I exist’ is the
only indubitable empirical assertion), and the dogmatic idealism of Berkeley (an
external world is illusory). Common to both of these is the thesis that the inner
is better known than the outer, and that outer substances are inferred from inner
experiences.
AIBC16 22/03/2006, 11:06 AM285
the critical philosophy of kant
286
Kant’s argument against these assumptions goes as follows. I am aware of
changing mental states, and thus I am conscious of my existence in time: i.e. as
having experiences first at one time and then at another. But, as has just been
argued, the perception of change involves the perception of something per-
manent. But this something permanent is not myself: the unifying subject of
my experience is not itself an object of experience. Hence, only if I have outer
experience is it possible for me to make judgements about the past.
Kant’s analytic closes with an insistence on the limits of the competence of the
understanding. The categories cannot determine their own applicability, the prin-
ciples cannot establish their own truth. Understanding alone cannot establish that
there is any such thing as a substance, or that every change has a cause. All that
is established a priori, whether by the transcendental deduction of the categories,
or by the exposition of the system of the principles, is that if experience is to be
possible certain conditions must hold. But whether experience is possible cannot
be established in advance: the possibility of experience is shown only by the actual
occurrence of experience itself. Concepts must be applied only to objects of pos-
sible experience; they may not be applied to things in general and in themselves.
Unless we are presented in intuition with an object falling under a concept, the
concept is empty and pointless.
Kant observes that philosophers make a distinction between phenomena (ap-
pearances) and noumena (objects of thought), and divide the world into a world
of the senses and a world of the understanding. His own analytic has shown that
there cannot be a world of mere appearances, mere objects of sense which do not
fall under any categories or instantiate any rules. But we cannot conclude from
this that there is a non-sensible world which is discovered by the understanding
alone. Kant accepts that there are noumena in a negative sense: things which are
not objects of sensible awareness. But he denies that there are noumena in a
positive sense: things which are objects of a non-sensible awareness. The concept
of noumenon, rightly understood, is simply a limiting concept, whose function is
to set the limits of sensibility. To accept the existence of noumena as extra-
sensible objects which can be studied by the use of intellect alone is to enter a
realm of illusion. In his ‘transcendental dialectic’ Kant takes us on an exploratory
tour of this world of enchantment.
The Transcendental Dialectic:
The Paralogisms of Pure Reason
The analytic has set out the territory of pure understanding. That is an island
of truth. But it is ‘surrounded by a wide and stormy ocean, the native home
of illusion, where many a fog bank and many a swiftly melting iceberg give the
deceptive appearance of farther shores, deluding the adventurous seafarer ever
AIBC16 22/03/2006, 11:06 AM286
the critical philosophy of kant
287
Figure 34 Title page of Kant’s first Critique.
(Photo: akg-images)
anew with empty hopes, and engaging him in enterprises which he can never
abandon and yet is unable to carry to completion’.
So, with this rare piece of romantic rhetoric, Kant begins his task of setting out
the logic of illusion in the transcendental dialectic. He is not interested in con-
tingent and accidental errors, like optical illusions or logical fallacies: his targets
are much grander, namely a priori psychology, cosmology, and theology. All of
these attempt to employ the mind in exploring a world beyond the bounds of
experience, an enterprise of which illusion is the natural and inevitable result.
All our knowledge, Kant says, starts with the senses, proceeds from the senses
to the understanding, and ends with reason. Reason, like understanding, operates
AIBC16 22/03/2006, 11:06 AM287
the critical philosophy of kant
288
through concepts; but while the pure concepts of understanding were categories,
the concepts of pure reason are Ideas. The allusion to Plato is deliberate: Ideas,
for Kant, are necessary concepts of reason to which no object corresponds in
sense-experience.
The Ideas of pure reason are arrived at by taking a form of inference, and
seeking to absolutize it. In ordinary life we infer conclusions from premisses; the
conclusions are true if the premisses are true. But this seems to be only a condi-
tional truth, since the truth of the premisses themselves may be called in question.
Reason looks for something unconditioned, a basis which is absolute, that is to
say, derived from nothing other than itself. What is absolutely valid is valid
unconditionally, in all respects, without restriction.
Kant says that there are three Ideas of pure reason; each of them is arrived at by
taking a pattern of inference and striving to reach an absolute. One line of argu-
ment starts from subjective experience and concludes with the soul as permanent
substantial subject. Another line of argument starts from the causal relationships
between empirical objects and reaches the idea of the cosmos as a totality of causes
and effects, unconditioned because it contains all conditions. A third line of argu-
ment starts from the contingency of the objects of experience and leads to the
unconditioned necessity of a being of all beings, namely God. ‘Pure reason thus
furnishes the idea for a transcendental doctrine of the soul, for a transcendental
science of the world, and finally for a transcendental knowledge of God.’
Let us consider first the illusions of a priori or rational psychology. Whereas
empirical psychology deals with the soul as the object of inner sense, rational
psychology treats of the soul as the subject of judgement. Rational psychology,
Kant says, ‘professes to be a science built upon the single proposition “I think”’.
It studies the transcendental subject of thinking, ‘The I or he or it (the thing)
which thinks’ is an unknown X, the transcendental subject of the thoughts.
The ‘I think’ which is the text of rational psychology is the expression of the
self-consciousness inseparable from thought. But how do we know that every-
thing which thinks is self-conscious? Answer: self-consciousness is necessary to
think of thinking, and in advance of experience we attribute to things those
properties which are conditions of our thinking of them.
Kant lists four fallacies – he calls them ‘paralogisms’ or bogus syllogisms – into
which we are led by our drive to transcend the limits of merely empirical psycho-
logy. In the first paralogism we proceed from the premiss ‘Necessarily, the subject
of thought is a subject’ to the conclusion ‘The subject of thought is necessarily
a subject’. In the second, we pass from ‘The ego cannot be divided into parts’
to ‘The ego is a simple substance’. In the third we move from ‘Whenever I am
conscious, it is the same I who am conscious’ to ‘Whenever I am conscious, I am
conscious of the same I’. Finally, in the fourth, we argue from the truth of ‘I can
think of myself apart from every other thing, including my body’ to the conclu-
sion ‘Apart from every other thing including my body, I can think of myself ’.
AIBC16 22/03/2006, 11:06 AM288
the critical philosophy of kant
289
In each paralogism a harmless analytical proposition is converted into a con-
tentious synthetic a priori proposition. Taken together, the paralogisms add up
to the claim that the self is an immaterial, incorruptible, personal, immortal
entity. This is the delusion of a priori psychology.
The Transcendental Dialectic:
The Antinomies of Pure Reason
We turn next to a priori cosmology. Here Kant presents us with a set of antinomies.
An antinomy is a pair of contrasting arguments which lead to contradictory
conclusions (a thesis and an antithesis). Kant constructed a set of these to show
that any attempt by reason to form ‘cosmical concepts’, that is to say, notions of
the world as a whole, was bound to lead to irresoluble contradiction.
The first antinomy has as thesis ‘The world has a beginning in time, and is also
limited as regards space’ and as antithesis ‘The world has no beginning, and no
limits in space; it is infinite as regards both time and space’.
The two propositions ‘the world has a beginning in time’ and ‘the world has
no beginning’ have had, as we have seen, a long history in the works of philo-
sophers. Aristotle thought the second could be proved. Augustine thought the
first could be proved. Aquinas thought neither proposition could be proved.
Kant now suggests that both propositions could be proved. That does not mean,
of course, that two contradictories are both true; it is intended to show that
reason has no right to talk at all about ‘the world’ as a whole.
The argument for the thesis starts from the definition of an infinite series as
one that can never be completed, and concludes that it cannot be the case that
an infinite world-series has passed away. But the argument is inconclusive. It is
true that any infinite discrete series must be open at one end: no such series can
be ‘completed’ in the sense of having two termini. But why may it not have an
end in one direction, while going on for ever in the other? Elapsed time would
then be ‘completed’ by having a terminus at the present, while reaching forever
backward.
The argument from the antithesis goes thus. If the world had a beginning,
then there was a time when the world did not exist. Any moment of this ‘void
time’ is exactly like any other. Hence there can be no answer to the question
‘why did the world begin when it did?’ The believer in a temporally finite world
can agree that it is not possible to locate the beginning of the world from outside
(at such and such a point in ‘void time’), while maintaining that one can locate it
from within (so many time-units prior to now).
Neither of Kant’s arguments is watertight, nor the parallel arguments which he
offers for and against the finiteness of the world in space. Altogether, the first
antinomy seems ineffective in establishing the impotence of reason.
AIBC16 22/03/2006, 11:06 AM289
the critical philosophy of kant
290
Kant presents four antinomies in all. The second concerns simplicity and com-
plexity; the third concerns freedom and causality; the fourth concerns necessity
and contingency. In each of the antinomies, the antithesis affirms that a certain
series continues for ever, the thesis that the same series comes to a full stop. Thus:
First: the series of items next to each other in space and in time comes to an
end (thesis) / goes on for ever (antithesis)
Second: the series of items which are parts of others comes to an end (thesis) /
goes on for ever (antithesis)
Third: the series of items caused by another ends in a free, naturally uncaused,
event (thesis) / goes on for ever (antithesis)
Fourth: the series of items contingent upon another goes on for ever (antithesis)
/ ends with an absolutely necessary being (thesis)
Each of the italicized relationships is regarded by Kant as a form of being condi-
tioned by something else: so that each of these series is a series of conditions, and
each argument concludes to an unconditioned absolute.
Kant thinks that both sides to each antinomy are in error: the thesis is the error
of dogmatism, the antithesis the error of empiricism. What the antinomy brings
out, he maintains, is the mismatch between the scope of empirical inquiry and the
pretensions of the rational ideal. The thesis always represents the world as smaller
than thought: we can think beyond it. The antithesis always represents the world
as larger than thought: we cannot think to the end of it. ‘In all cases the cosmical
idea is either too large or too small for the empirical regress.’ We must match
thought and the world by trimming our cosmic idea to fit the empirical inquiry.
The root error common to both the dogmatic thesis and the empiricist anti-
thesis is the idea of the cosmic whole. In each case a task set (e.g. to trace the
causal antecedents of an event) is confused with a task completed (e.g. a survey of
the totality of causes). The world as a whole could never be given in experience
and so ‘the world as a whole’ is a pseudo-concept. Hence it is not the case that
the world is finite and not the case that the world is infinite.
The third antinomy differs from the previous two. In the first two antinomies
both the thesis and the antithesis were rejected as false. But when Kant comes to
the third antinomy he seeks to show that, properly interpreted, both thesis and
antithesis are true. The thesis argues that natural causality is not sufficient to explain
the phenomena of the world; in addition to determining causes we must take
acount of freedom and spontaneity. The antithesis argues that to postulate tran-
scendental freedom is to resign oneself to blind lawlessness, since the intrusion of
an undetermined cause would disrupt the whole explanatory system of nature.
AIBC16 22/03/2006, 11:06 AM290
the critical philosophy of kant
291
Kant’s treatment of the third antinomy takes its place among the many attempts
which have been made by philosophers to reconcile freedom and determinism.
Determinists believe that every event has a cause, in the sense of a sufficient
antecedent condition. There are two kinds of determinists: hard determinists,
who believe that freedom is incompatible with determinism, and is therefore an
illusion; and soft determinists, who believe that freedom and determinism are
compatible, and can therefore accept that human freedom is genuine. Kant is a
soft determinist: he seeks to show that freedom properly understood is compat-
ible with determinism properly understood. An event may be both determined by
nature and grounded in freedom.
The human will, Kant says, is sensuous but free: that is to say, it is affected
by passion but it is not necessitated by passion. ‘There is in man a power of
self-determination, independently of any coercion through sensuous impulses.’
But the exercise of this power of self-determination has two aspects, sensible
(perceptible in experience) and intelligible (graspable only by the intellect). Our
free agency is the intelligible cause of sensible effects; and these sensible phenom-
ena are also part of an unbroken series in accordance with unchangeable laws.
To reconcile human freedom with deterministic nature, Kant claims that nature
operates in time, whereas the human will, as noumenon rather than phenomenon,
is outside time.
Many soft determinists have argued that freedom and determinism are compat-
ible because our actions, while determined, are determined by mental events in
our own minds; and an action is free, it is claimed, if it is determined by inner
rather than outer causes. Kant does seem to have believed in this kind of psycho-
logical determinism, but his reconciliation between freedom and nature does not
depend on defining free action as action that is psychologically determined. He
believed, surely correctly, that causal explanation (‘I knocked him over because I
was pushed’) and explanation by reasons (‘I knocked him over to teach him a
lesson’) are radically different types of explanation, and that the one is irreducible
to the other. But since the reconciliation he offers takes place not at the level of
experience but at the level of the noumenon, the thing in itself, his reconciling
project is fatally infected with the obscurity which attends those concepts.
The Transcendental Dialectic:
The Critique of Natural Theology
In the fourth antinomy Kant considers arguments for and against the existence
of a necessary being. He there leaves open the question whether a necessary being
is to be found in the world itself, or outside the world as its cause. It is in the
chapter on the Ideal of Pure Reason that he turns to consider the concept of
God, the object of transcendental theology.
AIBC16 22/03/2006, 11:06 AM291
the critical philosophy of kant
292
According to Kant all arguments to establish the existence of God must fall
into one of three classes. There are ontological arguments, which take their start
from the a priori concept of a supreme being; there are cosmological arguments,
which derive from the nature of the empirical world in general; and there are
physico-theological proofs, which start from particular natural phenomena.
In Kant’s rational theology a very special role is assigned to the ontological
argument. He claims that the cosmological argument is only the ontological
argument in disguise, and he argues that the physico-theological argument by
itself will lead us only to a designer, not to a genuine creator of the universe.
Hence the importance of his influential critique of the ontological argument.
What is meant by calling God an absolutely necessary being? Some philo-
sophers have defined a necessary being as one which exists in all possible worlds.
If we define God in this way, then surely he exists. Our world is one possible
world, otherwise it would not be actual; so if God exists in every possible world,
he must exist in ours.
But is it legitimate to build existence – even possible existence – into the
definition of something in this way? Kant thinks not. ‘There is already a contradic-
tion in introducing the concept of existence – no matter under what title it may
be disguised – into the concept of a thing.’ The ontological argument seeks to
make the statement of God’s existence an analytic proposition. If a proposition is
analytic, then the predicate is part of the subject and cannot be denied of it.
Taking the example of ‘A triangle has three angles’, Kant remarks:
To posit a triangle and yet to reject its three angles is self-contradictory; but there is
no contradiction in rejecting the triangle together with its three angles. The same
holds true of the concept of an absolutely necessary being. If its existence is rejected,
we reject the thing itself with all its predicate; and no question of contradiction can
then arise.
But why is Kant so sure that all existential propositions are synthetic? We can
argue from concepts to non-existence: it is because we grasp the concepts ‘square’
and ‘circle’ that we know there are no square circles. Why cannot we argue sim-
ilarly from concepts to existence? If ‘There are no unmarried bachelors’ is analytic,
why not ‘There is a necessary being’?
Kant’s principal argument is that being is not a predicate, but a copula, a simple
link between predicate and subject. If we say ‘God is’ or ‘There is a God’, Kant
says, ‘we attach no new predicate to the concept of God, but only posit the
subject in itself with all its predicates’. In fact, existential propositions do not
always ‘posit’, as Kant implies, because they may occur as sub-clauses in a larger
sentence. Someone who says ‘If there is a God, sinners will be punished’ does not
posit God’s existence. None the less we may agree with Kant that ‘exist’ cannot
be treated as a straightforward first-order predicate.
AIBC16 22/03/2006, 11:06 AM292
the critical philosophy of kant
293
Modern logicians, like Abelard in the twelfth century, rephrase statements of
existence so that ‘is’ does not even look like a predicate. ‘God exists’ is formulated
as ‘Something is God’. This clarifies, but does not settle, the issues surrounding
the ontological argument. For the problems about arguing from possibility to
actuality return as questions about what counts as ‘something’: are we including
in our consideration possible as well as actual objects?
Kant’s principal point remains, and it is similar to a point we have seen made
by Hume. ‘By however many predicates we may think a thing – even if we com-
pletely determine it – we do not make the least addition to the thing when we
further declare that this thing is. Otherwise, it would not be exactly the same
thing that exists, but something more than we had thought in the concept; and
we could not, therefore, say that the exact object of my concept exists.’ In other
words, whether there is something in reality corresponding to my concept cannot
itself be part of my concept. A concept has to be determined prior to being
compared to reality, otherwise we would not know which concept was being
compared and found to correspond, or not correspond, to reality. That there is a
God cannot be part of what we mean by ‘God’; hence ‘There is a God’ cannot be
an analyic proposition, and the ontological argument must fail.
Kant was wrong to think that the failure of the ontological argument implied
that all arguments for the existence of God collapsed. What his criticism does
show is that there is an incoherence in the notion of a being whose essence
implies its existence. But a cosmological argument need not purport to show the
existence of such a being, but only of one which is uncaused, unchanging, and
everlasting, in contrast to the caused, variable, and contingent items in the world
of experience.
Kant in fact has a criticism of the cosmological argument which is independ-
ent of his rebuttal of the ontological argument. All forms of the cosmological
argument seek to show that a series of contingent causes, however long, can be
completed only by a necessary cause. But we are faced with a dilemma if we ask
whether the necessary cause is, or is not, part of the chain of causes.
If it is part of the chain, then we can raise in its case, as in the case of the other
members of the chain, the question why it exists. But we cannot imagine a
supreme being saying to itself ‘I am from eternity to eternity, and outside me
there is nothing save what is through my will, but whence then am I?’ On the
other hand, if the necessary being is not part of the chain of causation, how can
it be its first member and account for all the other links which end with the
existence of myself ?
The argument for God’s existence which is most gently treated by Kant is the
physico-theological proof; it must always, he said, be mentioned with respect. His
aim is not to diminish its authority, but to limit the scope of its conclusion. The
proof argues that everywhere in the world we find signs of order, in accordance
with a determinate purpose, carried out with great wisdom. This order is alien to
AIBC16 22/03/2006, 11:06 AM293