SPACE AND TIME

IN

THE PHILOSOPHIES OF KANT AND BERGSON

Clifford Wellington Webb

University of Toronto
1956

UNIVERSITY

OF

TORONTO

SCHOOL OF GRADUATE STUDIES

PROGRAMME OF THE FINAL ORAL EXAMINATION

FOR

THE

DEGREE

OF

DOCTOR

OF

PHILOSOPHY

CLIFFORD WELLINGTON WEBB

2:00 P.M.

AT

MONDAY. OCTOBER 29th, 44 HOSKIN AVENUE

1956

SPACE AND TIME

IN

THE PHILOSOPHIES OF KANT AND BERGSON

COMMITTEE
Professor Professor Professor Professor Professor Professor Professor Professor Professor
J.

IN

CHARGE

R. O'Donnell, Chai

F. H. Anderson M. Long A. S. P Woodhouse D. P. Dryer G. Edison T. A. Goudge J. M. Kelly G. B. Phelan

BIOGRAPHICAL
1925 1951 1952 1953-55
56

--Born, Prescott, Ontario University of Western Ontario --B. A. --M.A., University of Western Ontario --School of Graduate Studies, University of Toronto
.

1955-56 --Lecturer, Department of Philosophy, University of Toronto.

Space and Time

in the Philosophies of

Kant and Bergson

(Abstract)

This thesis undertakes a systematic investigation of the theories of space and time of Kant and Bergson. Its purpose is to exhibit the relation between these theories and to show that Bergson's theory may be regarded as a logical develop-

ment

of Kant's.

As an essential preliminary to the investigation, an extended discussion of certain problems concerning space and time is presented. Of central importance is the question of whether it is necessary to maike a fundamental metaphysical distinction between space and time. It is concluded that the evidence seems to suggest that a theory which distinguishes between space and time is more tenable than one which treats space and time as exactly analogous elements in a four- dimensional space-time manifold. In addition, the preliminary discussion deals with problems created by the distinction between space and time. Since Kant and Bergson distinguish time from space by recognizing the intrinsic uniqueness of the passage of time, both philosophers are faced with the problems resulting from this distinction.

The introductory discussion provides a framework in which the views of Kant and Bergson are examined. First, by vindicating the distinction between space and time, it indicates the line which a development of Kant's views may justifiably take. Secondly, it uncovers the nature of those difficulties which lead Bergson to a development of Kant's views. Thirdly, it investigates the relationship between space and time and the principle of individuation in a manner which provides two alternative possibilities for the interpretation of Kant's theory of space and time.
It is suggested that a theory of space and time may take two alternative positions with respect to the relationship between space and time and the principle of individuation and other related principles which are the foundation of unity in nature. A theory may take the position that there is a principle of individuation for space which accounts for events in time, or that there is a principle of individuation for time which accounts for objects in space. In this thesis, the former alternative is called "possibility A", the latter, "possibility B".

At this point, Kant's views are introduced, and it is argued that his theory of space and time approximates to possibility B, rather than possibility A. Kant's manner of distinguishing between space and time, amd his doctrine of inner and outer sense are examined. It is shown that he conceived of space and time as homogeneous media, and that the distinction between them rests on the point that time has the characteristic of passage, and is uniquely associated with inner experience, while space is not. Space and outer sense seem to be, for Kant, an abstract aspect of inner sense, the peculiar form of which is time.

pointed out that Kant's exclusion of space from the account of the scherather than possibility A. is argued that Kant's Copernican revolution and his justification of For Kant synthetic a priori knowledge are only possible if he follows possibility B. must show that future experience will be determined in general in accordance with the categories, and this is only possible if the categories, as rules of synthesis which make for the unity of nature, apply to time rather than space. Otherwise, the passage of time might bring about experience not determined in accordance with the categories. In addition, Kant's answer to Hume's scepticism, by the same token, would not be possible except according to a theory which accepted possibility B. Kant's relevsmt statements are examined, and the weight of evidence seems to suggest that his theory of space and time approximates to possibility B.
It

is

matism

of the categories is consistent with possibility B.

it

In addition,

Bergson's views are then introduced, and the striking similarity between Kant and Bergson in closely associating time and inner experience is pointed out. It is shown that Bergson follows Kant closely in holding that the unique character of time is revealed in inner experience which is in time alone. Kant's views on the impossibility of a science of psychology reveal that, for him, inner experience is quite unlike outer experience. Inner experience constitutes an area of appearances, which, although indubitable, are not subject to categorical determination, and hence do not represent possible experience in Kant's sense. Kant and Bergson are in agreement that knowledge appropriate to spatially related objects is inapplicable to the flow of inner experience.
This raises several problems which Kant does not attempt to solve, but which Bergson deals with at length. Kant's Copernican revolution requires that the categories, which arise independently in the nature of human thinking, should apply directly to time itself. Kant's doctrine of the Transcendental Imagination and the Schematism represents an attempt to mediate between the categories with their independent source, and the concrete flow of time. But how such mediation is possible

remains obscure. Bergson meets this problem by declaring it to be insoluble, a pseudo-problem stemming from a basic misconception of the nature of time.
Kant held that time was adequately representable in terms of concepts appropriate to space. But he was unable to render this doctrine consistent with his view that time is a unique form of sensibility. Bergson's development of Kant's theory consists in large part of showing that the time of inner experience cannot possibly be understood in ternas of concepts appropriate to outer sense. Kant had refused to answer the question of how human experience, which is characterized by two different modes of sensibility, namely, inner and outer sense, comes to have a unity. He gives no clear account of the unifying relation between inner and outer sense, and consequently no answer to the question of the relation between time as a unique form of the perpetual flux of inner experience, and the homogeneous time of outer sense. Bergson attempts to answer this question by showing that homogeneous time is a spurious, spatialized concept, and by an appeal to intuition. Spatial concepts falsify time, but the relationship between space and outer sense, and time and inner sense may be grasped by a metaphysical intuition which reveals how the concrete flow of time is broken up into discrete spatial parts. Bergson, as well as Kant, subscribes to possibility B, in holding that the intellect applies to the basic flow of time. But for Bergson, this application constitutes a falsification of the metaphysical reality of time.

opment

Bergson's view that time is ultimate reality represents a consistent develof Kant's theory of space and time. For Bergson's theory of time as a con-

Crete process of change, which is glimpsed in inner experience, takes advantage of the fact that, for Kant, the appearances and changes of inner experience are beyond the pale of cognition. Inner experience, for Kant, has a status exactly analogous to the status of things- in- themselves in being incapable of being known by means of the categories. Yet inner experience is indubitably real since it is actually experienced as a continuous flux, a point which Kant often stresses. Bergson's theory carries to its logical conclusion the point that beyond the sphere of conceptual determination. there can be no distinction between form and content. Thus time is not merely the
real form of inner sense, as
in the indivisible
is
it

is for

Kant, but is concrete change itself as revealed

succession of conscious states. implied in consciousness itself.

Time

is real,

in itself,

because

it

Bergson's close association of consciousness with the flow of time is related another aspect of Kant's thought. Kant distinguishes between inner sense and apperception, but fails to stress that inner sense is conscious inner sense. Bergson seizes upon this point, combining it with Kant's view that inner sense reveals a perpetual flux which is not cognizable. Kant's distinction between inner sense and apperception, thus represents the seed of Bergson's radical separation of the intellect and real time revealed in inner consciousness.
to

Bergson erects the apprehension of the change revealed in inner experience supreme metaphysical principle. He thus follows Kant's view that a metaphysics which would penetrate beyond appearance to reality itself must be intuitive. But whereas Kant thought this intuition would have to be an intellectual intuition, Bergson, stressing the consciousness of change in inner experience, argues that the intuition is non- intellectual. He rejects Kant's doctrine of judgment, and a logic of temporal process, holding that intellectual thinking is through-and-through a spatialization of a fundamentally non-spatial reality. Following this line of thought Bergson tries to show how matter and the intellect itself arise from the basic flow of duration (la duree). He thus denies the independent origin of the pure concepts of the understanding, and tries to show that logic itself is derivative rather than fundamental. Bergson's development of some aspects of Kant's thought thus ends in a view which would have been anathema to Kant.
into a

GRADUATE STUDIES
Major Subject:
Systematic Philosophy and Metaphysics:

Professor F. H. Anderson Professor T. A. Goudge

Minor Subjects M. A. University

: ,

of Western Ontario (Subject of thesis: philosophies of Hegel and Whitehead)

Process

in the

English Language and Literature:

Professor A. S. P. Woodhouse, Professor F. E. L. Priestley.

SPACE AND TIME

IN
TIffi

PHILOSOPHIES OF KANT AND BERGSCMf

BY

Clifford Wellington Webb

A thesis submitted in conformity with the requirements for the degree of

Doctor of Philosophy
in the

University of Toronto

1956

CONTENTS
CHAPTER ONE:

INTRODUCTION

CIIAPTEc i'uL:

Pi-iSLIKBIARY DISCUSSION OF TH5 DISTINCTION

BETWEEN SPACE AND TIME

1.

General Featiires of the Distinction between Space and Time

Similarities and Differences betxvean Space and Tiiiie
The Direction of Tine and the Problem of Individiiation

2.

3*

CHAPTSIt TimSE;

KAIjTS TiiliCRI OF SPACE PdiB THIS

1, 2

Kant's Distinction between Space and Time

Inner and Outer Sense

Time and the Unity of Nature

3.

CHAPTER

FOUxl:

BERG30N3 THEORY OF SPACE AND TIME

AI^D

ITS

RELATION TO THAT OF KANT

1,

Time and Inner Experience

The Cognitive Representation of Time

2,
3,

Primordial Time and the Spatializing Intellect

NOTES

BIBLIOGRAPHY

CHAPTER ONE
IKTRODUGTIOK
This thesis aims at a systematic investigation of
the theories of space and tine of Kant and Bergeon,

Its

purpose is to exhibit the relation betvreen these theories,

and to

shov;

that 3ergson*s views may be re.jarded as a

It should be added

logical development of those of Kant,

that this endeavour proceeds from a metaphysical standpoint.

The Critique of Pure Reason . regarded as embodying Kant^s

mature philosophy, is not looked upon as solely an episte-

mological treatise.

That this v;ork is concerned v;ith basic

episteraolo^ical questions, of cou-Tss, soes without saying.

That it does not also involve metaphysical principles and

commitments, is here denied.
Thus, for example, we take

the view with Gottfried Martin that.

The question that Kant is asking is not how can know space and time, but quite simply vjhat space and time are. This question has an ontological sense, . . 1

v/e

Kant's ovm words, indeed, suppoirt this view, for

in the Transcendental Aesthetic he asks, not "/hat knoiv-

ledge do we have of space and

til

tir.ie?*',

but "IJhat, then, are

space and time?".

It becomes apparent that Kant does not

merely tell us

hovj

space and time figure in human knowledge,

but ;ives us a view of the natures of sjjace and time,

1

themselves.

It is necessary to emphasize this because va-

rious theories of space and time may be presented in different donains of knowledge.
The question of what space

and time are is answered differently, for example, in physics

and psychology.

Kant offers an ansxver to this question,

however, not merely in the context of some particular subject matter, but in a v/ider, metaphysical sense.

For this reason, we consider it essential to approach

our subject by way of an extended preliminary discussion.
In this discussion we shall deal, first, with the general

features of the distinction between space and time (Gh.II,

Sec. 1); secondly, v;ith the question of their ultimate si-

milarities and differences

(Ch. II, Sec. 2); and thirdly,

with certain relations between space, time and change (Ch, II,
Sec. 3).
In connection with the first of these topics, an

indication of the chief desiderata of a metaphysical ex-

planation of the distinction between space and time is

intended.
One of the central points to be made in this

thesis is that Kant*s theory of space and time leaves certain problems unsolved.
That is, Kants explanation does

not deal vjith, indeed, does not even attempt to deal with
some of the questions that arise in connection with space

and time.

Yet Kant gives us a metaphysical theory of space

and time, and a metaphysical account of their distinction.

His vlev;s do not pertain merely to space and time as they

are dealt v/ith either in physics or psychology.

His doctrine

is, nevertheless, incoi:iplete as a metaphysical theory

Kant

explicitly recognizes that he leaves some questions unanswered. Our preliminary discussion of the general features
of the distinction between space and time attempts to clari-

fy the nature of these questions.

Bergson's views are here regarded as a development

of those of Kant in the sense that Bergson, while agreeing

with Kant on fundamental points, goes on to develop a theory

which attempts to deal with the questions which Kant left
unanswered.
Thus Bergson's viev;s differ radically in some

respects from those of Kant,

Kant's theory of space and

time is necessarily limited by the critical position which

he adopted.

Bergson continues on, beyond the point where

Kant stopped, to make statements of a kind which Kant, on

principle, would refuse to make.

Accepting the premise

that unansvjered metaphysical questions concerning the re-

lationship between space and time require answers, Bergson's doctrines may be regarded in this respect as a logical

development of Kant's.

One point, ho'.vever, needs to be made

clear in this connection.

This thesis is not specifically

an attempt to evaluate the merits of Bergson's development

of Kant's views,

No defence of either Kant or Bergson is

attempted.

It is hoped, hoxvever, that presentation of the

4
issues involved in the relation between the theories of these

two philosophers v/ill aid the reader in drav/ing his own conclusions.

With recpsct to the second set of topics which we

consider an essential preliminary to our subject, namely,
those concerning the similarities and differences betxveen
space and time, it is intended to present some of the im-

portant considerations which bear on the question of whether

or not a fundamental distinction must be made betv/een space

and time.

By a fundamental distinction

betvireen

space and

time, we mean a metaphysical distinction v;hich recognizes

that time is intrinsically different from space at least

in that it has the characteristic of passage.
It should

be added that v;hen it is said that time has the characteristic of passage, it is not meant that time itself changes or passes, v/hich is a rather loose way of expressing a

difficult point.

The question of a fundamental distinction

betvjeen space and time, the crux of xijhich turns on the pro-

blem of the passage of time, is discussed quite generally,

;\rithout

specific reference to the two philosophers under

examination in this thesis.

Nevertheless, the conclusions

derived from this discussion have an important bearing on

the main conclusions of the thesis.

Since both Kant and

Bergson do make such a fundamental distinction between space

and time, the validity of their doing so is an essential

point in judging in

i.'hat

direction a development of Kant's

If Kant was wrong in making
xi?ould

views may justifiably be made.

this distinction, a reconstruction of his philosophy

involve a rejection of the distinction.

If Kant was right,

however, the only fruitful line of development vjould be to-

ward a theory which likewise made a fundamental distinction

between space and time.
Recent philosophical literature has included notable

attempts to show that the analogies beUveen space and time,

vjhich
ii?ill

be discussed herein, are not merely superficial,

but may be pressed much further beyond the point ^vhere they
vjere

heretofore thought to break doxm.

xi;hich

It is argued by cer-

tain writers that differences

have traditionally been

supposed to obtain betvjeen space and time, do not, in fact,

hold, and that the supposition tliat there are such differences
is the result of insufficient analysis. In connection with

this, there is an attempt to show that one characteristic

in particular vrhich was formerly thought to mark a funda-

mental difference between space and time, viz., the characteristic of the passage of time, as opposed to the static

quality of space, is not a characteristic of time at all,

and does not constitute ground for a differentiation of
space and time.
In conjiinction with this view there has

been presented xvhat is termed the "manifold theory".

This

theory not only denies that time is fundamentally different

from space in that it has the characteristic of passage, but

confesses to bafflement as to what a distinction on this

ground might mean.

The denial of passage is logically asso-

ciated with the metaphysical theory of a four-dimensional

manifold of space -time and is the main point at issue in

the problem of whether tliere is a fundamental distinction

between space and time.

Upholders of the manifold theory

agree that their theory is adequate to account for some

distinction between space and time, but they are vague as

to what the distinction is.

Hence it is difficult to de-

cide whether they have some other metaphysical distinction

in mind or not, and still more difficult to imagine v;hat

the nature of this other distinction might be.

Ari^uiaents

advanced in favour of the manifold theory

are herein examined with a view to deterruining their validity.

It is maintained that this position has not been de-

monstrated to be true, and that there are good reasons for

rejecting it in favour of
x.he

viev;

that there is a fundamental

It is important to em-

distinction between space and time.

phasize this because it has been supposed that v;ith some

reconstruction Kant^s philosophy could be made to accomodate

a theory of space-ti:;ie. a reasonable supposition.

In one sense, of course, tliis is

The physical theory of space-time

may be accepted without rejecting a fundamental distinction

between 3pace and time.
As to the further considez'ation of

whether Kant's epistenology excludes the possibility of nonEuclidean geometries, v;hich a numbsr of scholars
dealt with,
dering.
liave

ably

this issue is other than the one

v;e

are consi-

Je must carefully distinguish the question of

v/hether a theory of space-time, such as has been found use-

ful in physical science, is compatible v;ith Kantian epistemology, from the question of
-.ifhether

a metaphysical theory

of space-time, v;hich denies passage, is compatible vjith

Plant's metaphysics of space and time.
Ifhat

we are considering

is the possibility of a development of Kant's views which

espouses a theory of space-time v;hich utterly obliterates

Kant's
v;ay

of distinguishing betv;een space and time,

3uch

a theory, incorporating some of Kant's critical principles

could, no doubt, be formulated.

V/hether it would be Kantian

in the sense of including Kant's metaphysical presuppositions

concerning space and time, hov;ever, is another question.

It is argued herein that a viev7 v/hich fails to make a fun-

damental distinction beU;een space and time,

v,'ill,

in im-

portant respects, be un -Kantian, a departure from Kant i*ather

than a development from him.

It is important to distinguish this issue concerning

the passage of time from problems relating to space and time as they are dealt with in physics.

Criticism of the philo-

sophical conception of space-time is not necessarily a criticise

of thy physicists' conception of space-tirae.

The

"manifold theory" xvhich is here discussed is not identical

with the physicists* conception of space -tine, although there

are similarities between the two.

For exai-iple, both theories

concur in treating time mathematically as a fourth dimension.

But whereas the physicist s treatment of space and time is

fashioned in conformity

'iifith

the end of achieving the greatest

theoretical generalizations concerning physical events, the

philosopher's conception knows no such limitations.
Thus,

the latter speaks not merely of physical events in spacetime, but of all events as being in space-time,
1 believe that the universe consists, without residue, of the spread of events in space-time, and that if vje thus accept realistically the four-dimensional fabric of juxtaposed actualities we can dispense v/ith all those dim non-factual cate:3ories Vihich have so bedevilled our rave: the potential, the subsistential, and the influential, the noumenal, the numinous, and the non-natv-ral, 4

The metaphysical position which denies the intrinsic \mique-

ness of time, and represents the universe in the above manner,

is, we shall suggest, without foundation.

But the fundamental distinction between space and

time is likewise important for our piorposes in that Bergson's

position leans heavily on it.

It is not too much to say

that the core of Bergson's philosophy lies in his interpre-

tation of time.

Bergson states that he follows Kant's doc-

trines concerning space, but disagrees with Kant concerning

tine,

Nevertheless, they agree that there is a basic dif-

ference between the two, and they agree on certain important

features of this distinction, notably, on points concerning

that aspect of ti-o, the characteristic of passage, which

distinguishes it radically from space.

Bercson^s develop-

ment of the position held by Kant is a development of v/hat

is implicit in a position explicitly and rigorously adopted

by Kant, namely, that there is a fundamental distinction

between space and tine,
ture of this distinction.

Bergson probes deeper into the naHe sharpens the kind of distinc-

tion made by Kant, and argues that if time is something quite

unlike space, it is not even conceivable in terms of space.

This is the point where Kant stopped,

\'ith some

hesitation,

Kant declares that time is legitimately represented in spatial

terms.
But how this is possible in view of the basic kind

of distinction which Kant makes between space and time, re-

mains unexplained,

fCants association of time v/ith inner

sense is at the core of the difficulty, and is, in addition,

one of a nuriber of rather striking similarities between Kant

and Bergson,
That the philosophies of Kant and Bergson have re-

markable similarities is a point made by A. D, Lindsay,

also points out the other element in Bergson s thought

He

which allows the doctrines of the two philosophers to be

considered systematically in a relationship of development,
namely, Bergson s critical attitude.

10
IjOW without su^esting' any comparison in iinportance between Kant and Bergson, there is this resemblance bet'.i/QtJii thei.:, that much of bhe in'^crest o-f Jer^son's work consists in his statement and exposition of antincruies to be found in present-day pliilosophy, that as the best road to the solution of these antinomies he errors a ii;;'.; stateirient of w.h& ^aslc or probleni of philosophy, and propounds a new method. Like Kant, his ;vork professes to be critical: to find the main source of difficulties in an uncriticissed assumption, 5

Ber.^son's doctrines concerning space and time in-

volve a revelation of certain uncriticized assumptions made

by Kant, one of

xi/hich

has already been mentioned, viz, the

It is the criticism of

legitimacy of spatialiaing time.

this assuraptioh, among others, that leads Bergson to go be-

yond Kant in his theory of space and time.

It may be said,

however, that Bergson might well have criticized other as-

sumptions than those he did.

He might, for example, if he

had been a philosopher of a different persuasion, have cri-

ticized the way in which Kant distinguishes between space

and tine and have modified Kant^3 views in the direction of
a manifold theory which excluded passage.

It may be that a

theory of space-time which excludes passage vjill prove to

be the raost defensible theory,
/e
su;:^i:/;est,

hov;ever, that

at present this is far from being obvious, and that cogent

reasons seem to point to the opposite conclusion.

At any

rate, Berg'son's criticism did not take this line of depar-

ture, and for this reason we do not here deal vjith that con-

tingency.

But more essentially, it is felt as vieU that such

a modification of Kant does not represent a logical developmert

11
of Kants theory of space and time, but rather a clean break

with it in the form of an entirely different view of space

and time.

For in one

v.'ay

at least, Kant vjas naking no un-

critical assumption in drawing a ftindamental distinction between space and tirue.

He shov/s himself to have been conscious

of the most important reasons for doing so, particularly in his exa:::ination of the relation betvjeen time and inner

sense,

Kants theory of space and tine is intimately bound

o'.fn

up v;ith his

conception of the critical philosophy as he

conceived it, and it is the relation of this position to

that of Bergsoa that we consider.
V/ith

regard to the que-

stion of the extent to v/hich the critical philosophy may be

modified so as to be compatible xvith a theory of space-time

v/hich denies the fundamental uniqueness of time, and still

retain the essentials of the critical position, we make no

commitment, except that v?hich
vje

have already indicated,

namely, that such a raodification will, in some respects, be

lin-Kantian.

The respects in which

x-ze

maintain it will be

so, pertain, ivithin the scope of this thesis only to the

theor-/ of space and time.
VJe

do not intend to su?sest that

all aspects of Kant^s thought, or even the essential aspects

of it, necessarily involve his fundamental distinction be-

tween space and time.

Nor do we wish to pass judgment on the relative merits

of Kant^s position as it stands in his writings or on the

possibility of a renovated Kantianism which xvould involve

12

the obsolescence of i;ant*3 distinction.

In connection v;ith

this, however, it nay be said that it is not entirely ob-

vious that because

modem physics requires

a theory of space-

tine, Kant's position, if it is to be defensible in the light

of the new physics, needs to be clianged in a loanner which

would obliterate his way of distinguishing between space and

time.
This is, itself, a difficult
i'-'o

a^id

extensive problem,

and like

others mentioned above, lies outside the parWe take Kant's position

ticular subject we wish to explore.

on space and time as he expressed it, and consider only the

developments from a position which retains that specific

character, that is, I'rom a position which involves a fun-

damental distinction between space and tine.

Thus, v;hen

vje

speak of logical development, we mean development of the

metaphysical implications of a theory.

The theory which we

are dealing with is one which includes the premise that time
is ultimately and uniquely different from space,
i\nd

the

reason v;hy we treat of development in this sense is that we feel that there are sound reasons for making this basic distinction.
Finally, it is advisable to
svun

up the similarities

between Kant and Bergson, vjhich constitute the main raisons

d'etre of this thesis.
The importance which both philosoti^ie is

phers attach to their views of space and

ting,
3ei.'gson's vievjs

worth no-

concerning che relationship betv;een

13

space and time lie at the base of his ..letaphysics,

Kants

Transcendental Aesthetic stands in the foreground of his

philosophy as an absolutely indispensable part of it.

The

position of neither philosopher can be understood without a

comprehension of his doctrine

of space and time.

But the resemblance between Kant and Ber.^ison need

not be lifiited to genei-alities. This is particularly true

of their vlevvs on space and time.
Both, as
v^e

have already

emphasized, make a fundamental distinction becvjeen space and time, and for both this distinction has a considerable,
if not decisive, bearing on
of their philosophies.
hovj

they work out other parts

In addition, both philosophers re-

gard inner experience as highly important in connection

with the apprehension of time.

Both hold that time cannot

be represented directly, but only through the mediation of

space.

They agree that space is uniquely associated v/ith

the logical functions of the understanding. Both call atten-

tion to the all-pervasiveness of time, in contradistinction

to spatiality.

And both are faced with che problems V7hich

It is these

the distinction between space and time poses.

problems which lead Bergson to a theory v;hich can be regarded as a developuent of Kant's views, and vjhich allow

him to draw out the implications of Kant's node of distingviishing between space and time in a manner which Kant,

himself did not envisage. In thic respect, only one further

u
point of clarification concerning our subject needs to be
iaade#

This is that while there undoubtedly are historical

connections and influences betxveen Kant and Bergson, our

treatment is meant to be systematic rather

tlian

historical.

The relations we examine between the two men raay or may not
be paralleled in point ox historical connection, For the 6 most part, they probably are not.

CMFTER TWO
PRELimNARY DISCUSSION OF
BST\\rSEN
TKi*

DISTIIIGTIGN

SPACE AND TIMS

1,

General Features of the Distinction between

J pace

and Tine

No distinction is norc familiar to corxion sense and

more difficult to ^ive a philosophic

accoi;int

of than the

distinction betv;een space and time.

v/e

In everyday experience

are not often confused as to vjhat is spatial and what is But in philosophy, sone questions which
vje

temporal.

can

ask about the world seem to demand that we give some account
of the separation between space and time, ana indicate
hov;,

or in what sense, we knovj them to be fundamentally different,

If, for example, we ask ourselves whether all things

exist in space and time, or whether some exist in one and

not the other, or in neither,

v;e

find it difficult to put

forth an ansvver v/ithout corairdtting ourselves to some view

concerning the different natures of space and time.
Ve may,

of covirse, in answering such questions reach the conclusion

that it is a mistake to regard space and time as fundamen-

tally different, and this will have an important bearing on

the kind of answers we give,
tliat it is

3ven if we take the position

futile to attempt to inquire into the "nature"

of such tilings, and concern ourselves simply with the v/ay

15

16 in which spaces and tines are neasured,

v^e

have not removed

the necessity of deciding in v/hat sense they are distinguishable.

For we do in fact measure them differently, and this

7

alone constitutes a recognition of a difference between them.

Once any basis of distinction is accepted, the quest-

ion arises as to whether this basis is adequate to the ex-

planation of all aspects of the distinction betvjeen space

and time, and whether it is compatible \vith the similarities

between them.
vsays in v;hich

We have then to take acco\mt of the different

space and time are viewed in the various spe-

cial sciences.
The different ways in which space is regarded in

mathematics and in psychology is well expressed by Cassirer.

If from the standpoint of netageometry, 3uclidean geometry appears as a mere beginning, as given material for further developments, nevertheless, from the standpoint of the critique of knov/led.c^e, it represents the end of a complicated series of intellectual operations. The psycholOi~ical investigations of the origin of the idea of space (including those which were undertaken with a purely sensationalistic tendency) have indirectly confirr;ied and clarified this. They shov; unmistakably that the space of our sense percep * tion is not identical v/ith the space of our /geometry . but is distJRi'Cuished from it in exactly the decisive constitutive properties . "Above" and "below", "riifnt" and "left" are here not equivalent directions, which can be exchanged with each other v;ithout change, but they remain qualitatively distinct and irreducible determinations, since totally different groups of organic sensations correspond to then. In geometrical space, on the contrary, all these oppositions are cancelled. 6

Similarly, the difference betv;een tine as it appears

in psychology and as it appears in physics, has been stressed

17

by Gunn,

vjho

states that "Neither psychology nor physics

tLnie

attempts to srasp the problem of the natiire of full significance J

in its

the one is merely concerned with our sub-

jective avjareness of time and the other confines itself large9

ly to considerations of measurement".

Gunn points out some

of the differences betv;een tine as it is dealt with in these

two contexts.

For instance, tine as perceived is alivays limited. We never perceive the xifhole of tine. It is also perceived as sensibly continuous, as having a certain directional quality; it is transitive and related in its content to the subject at the moment of experience. Only if the vjider implied temporal perspective and the tirae-span immediately experienced be apprehended as passing into one another can Time be grasped, and in this way it is grasped as a continuura. Time as conceived is unlimited in character, is regarded as infinitely divisible and mathematically continuous like an infinite series. Further, it is looked on as involving an objective order of beforehand-after, ivhich is not to be equated \;ith the past, present or future of a subject, . Conceptual time is also conceived to be a unity in spite of the difficulty of ascribing to it any principle of coherence. Perceptual time, hovjever, is rooted in experience and professions of unity are not to be made in regard There may on this level be many unrelated to it. times. 10

Until quite recent times the distinction between

space and tine, v/hatever it may turn out to involve on closer inspection, has been carried over from comraon sense into
the special sciences, in general, without much modification.

Indeed, to common sense the distinction seems to be so ob-

vious that the question of whether or not space and time are
distinct seems to be irrelevant. Were it not for the mer-

ging, in contemporary physics, of physical space and physical

It
time in a space-tir.ie continuum, and for a similar merging 11 of space and time in recent metaphysical theories, one might

rest content in the view that they are distinct, and direct
one's attention to grasping the peculiar nature of each.

They offer perplexities enoiogh when taken separately, Ue

have Minkowski's famous statement, however, and its repetition by nur;ierous popular expositors, to remind us that it

would be rash to assume uncritically that space and time at

least in the context of physics are ultimate and distinct.
The viev;s of space and time v/hich I v/ish to lay before you have sprung from the soil of experimental physics, and therein lies their strength. They are radical. Henceforth space by itself, and tine by itself, are doomed to fade away into mere shadows, and only a kind of imion of the two v;iiak preserve an independent reality, 12

Yet Minko\irski s prediction seems to have been pre

mature,

/mother eminent physicist expresses a more moderate

estimate of the effect of the space-time conception on the

physical distinction between space and time.

Although mathematical attempts to demonstrate the unity of space and time in a single four-dimensional world do not completely obliterate the difference betx-jeen distances and durations, they certainly reveal a much greater similarity betv;een the two notions than vias ever evident in pre-Einsteinian physics, 13
Thus the question arises as to whether space and time are ultimately or fundamentally distinct; and if so,
in what their distinction consists.
It is not merely that
It is

we do constantly distinguish them in everyday life.

also a question of

xi^hether

those aspects of experience which

19
we deslf^ate as space and tine are sinple and irreducible,
each incapable of further explanation in terms of something
else; or whether they admit of analysis into more basic ele-

ments in

tor/iis

of which their very difference in experience

In either case, it is apparent that we

would be explained.

are faced with a philosophic problem rather than a purely

physical or psychological problem.

Physics, at least in its

present form, says nothinc about how the space-tine continuum

is related to the psychological apprehension of space and

time; not does the science of psychology relate its views

to those of physics.
It can be agreed, of course, that the accounts of

the distinction between space and time given by the special

SRiences, the expositions, that is, of v;hat each is "known

as", cannot be ignored by the philosopher vjho wishes to know

whether they are ultimately distincjr or not.

For he is not

merely setting out to give one more particular account of

space and time, but is concerned to give an explanation which

will embody the results of the particular sciences, either

in the sense of shov/ing one or more of them to be In error,

and

hov/

such error comes about, or in the sense of incor-

porating scientific results into a more general theory.

Such a general theory of space and time may be either

monistic or dualistic depending on whether it takes the view

that space and time are not ultimately distinct, or the view

20
that they are.
In either case, it v/ill be necessary to take

into consideration the way in which space and time are dis-

tinguished in different sciences.

the greatest difficulty arises.

But it is just here that

If it is supposed that they

are ultimately distinct, it will not be enough merely to record this.

It vjill be necessary also to state in vjhat manner
v/hat

they are distinct, or in other words, in

consists.

the distinction
iii

Are both space and time to be placed

the cate-

gory of that which is immediately intiiited in sensible experience?

Are they to be regarded as atomic qualia exhibited

in experience^

Are they to be construed as different formal

wholes which are logically prior to spatial and temporal

relations?
Or may they be distinguished by placing one in

the realm of the purely conceptual, and the other as the con-

tent of a peculiar hind of non-intellectual intuition?

Are

they to be interpreted as different self -subsisting substance

Or
are they ultimate and irreducible kinds of relations?

These are some of the ways in which space and time may be

basically distinguished.

For all of them there are a number

of crucial problems which must be faced,

A metaphysical theory of space and time must take

account of the claim that experience

lias a

unity,

either
If it

in the sense of denying such unity or accepting it.

be agreed that human experience presupposes a unifying prin-

ciple, it follows that spatio-temporal aspects of that

21

experience must reflect that principle.

Triat

is, there must

be a uiuity, in some sense, ox space and tine.

How that unity-

is possible, considering the intrinsic difference betv/een

space and time, is a question of first importance.

It v/ill

not be a problem for a philosopher who repudiates any fun-

damental distinctioii

bet\\;een

space and

tiiiie.

But for one

who accepts a distinction it will be crucial,

V/hat,

then, are the problems which such a philoso-

pher must try to solve?

For purposes of illustration, let

us consider the viev; that space and time are distinct sub14 stances. This theory brin;';s certain difficulties into sharp

focus, for it places them in the context of the traditional

problem of how one substance can act upon, or be related to another substance.
The concept of substance as that which

exists in itself is peculiarly obdurate to relatedness.

Spinoza s addition to the traditional definition, namely,
that a substance is also that v/hich is conceived thro\igh
-15

itself,

clarifies the nature of these difficulties.

For

a substance qua substance,

(i.e. not qua accident), has,

coni;

according to its definition, nothing in

on with any other

substance, and hence two such substances cannot be mutually 16 understood. Kow hovj two such substances can liave any sort
of relation, capable of being known, is indeed a perplexing

question.

It is possible that they might be related in a

fashion which cannot be known, that is, in a fashion v;hich

22

cannot be understood, but must be grasped, if at all, through

means other than cognition,

(This is an important possibility

and has an obvious affinity to Bergson^s doctrine of meta-

physical intuition).
substances must have

The cnuc of the puzzle is that

soraothinr::

tvjo

in

corar^ion,

or must resemble

one anotlier in some v;ay, in orr'er to be mutually understood,

qua substance,
in common.

and they cannot, by definition, have anything

It ni:ht v/ell be said that it is the definition

of substance which is at faid.t, and should therefore be discarded, but since we are here concerned only
in the context of an illustration,
vie

i.'ith

substance

leave that objection

to one side.

For the importance of substance for our pre-

sent purposes lies not in any question of the usefulness of

substance for philosophy, but in the illustration of the

difficulties of cognizing relatedness.

Considered from this

point of view, the old problem concerninr the relations of

tv;o

substances, comes to be seen as the problem of knov;ing

two things which have nothing in common, and this applies,

not merely to substances, but to anything vjhatever that we
are acquainted with.
If we become acquainted with anything

v^hich literally has nothing in common

with anything else we

are acquainted with, the possibility of cognizing it is ruled

out.
V/e

can be said to knoxv it only if we consider it per-

missible to speak of knowledge by acquaintance in the manner

17 of Russell,

and this is a different sense of the terra "know-

ledge" from the one we are considering at the moment.

23

Knowledse by acquaintance of something vjhich has

nothing in common with anything else, hov;ever,

us very far in explaining the world, for
\:e

\7ill

not carry

cannot relate

such knowledge to anything else, except accidentally, or a

posteriori .

Whatever else it may be, such knov;ledge is not

the kind which metaphysicians have traditio?ally sought. About

all

vje

can do with an entity wliich has nothing in

coraraon

with anyching else is to desi,:piate it by a proper name.

might be said that in naming it
vie

It

endow it with one

coiiUion

property at least, viz., that of being in the class of things

which may be spoken of.

ever name
\-m

But this ignores the fact that whatar.

give it is not intrinsic to the thing, not

essential attribute v;ithout which the thing v/ould not exist,

but is merely accidental and unessential.

3y naming it

vre

do not relate it necessarily to anything else, for our name

does not imply that there is anything

coraraon beti^een

the en-

tity and something else.

It is only when we can form a con-

cept of a thing which contains something inAwith the concept

of another thing, that

tv/o.

v/e

can have rational knov;ledge of the

And in this case, of course, the very fact that we can

conceptualize a thing in teinns of comraon properties, means

that the thing is not unique.
It may well be asked v/hether there are any such ab-

solutely unique entities, or alternatively, v/hether everything with which we are acquainted is absolutely unique.
V/e

24

may put this question in other terms and ask v^hether we ever
do have concepts, in terms of common properties, of things

with which we

ai-e

acquainted, or whether our concepts, in

their (generality, always allow the intrinsic uniqueness of

things to escape,
thing
v;e

/uid

this amo\.mts to asking whether anyThis,

are acquainted with resembles anything else.

it sliould be uo'oed, is a point of a different order from

the truism that general concepts, in their generality, leave

certain aspects of the objects conceptualized unspecified.

If it be admitted that describing something in general terms

omits its particularity, it is an easy step to the posi-

tion that there is in the universe a principle ox the vndetermined, vmich remains over v/hen conceptualization has
done its
i-jork,

and might be suitably spoken of as "that which

If concepts leave a residue, This position,
v.'e

in itself is not a this*".

the residue must be conceptually undetermined.

interesting as it is in itself, is not the one

cerned with here,
\ihat vie

are con-

are concerned vnth is the possi-

bility that there may be entities which are absolutely

unique, and not amenable to any sort of conceptualization.

With reference to the majority of objects of experience, we

may feel sure that &he question ansvjers itself, for we undoubtedly are aware of numerous resemblances betv/een things.
If
v^e

relate this question to space and time, however, we

seem bo be less certain of the answer, and several possi-

bilities present themselves.

25 First, we may take

tlie

view that

oiu-

concepts of
This raises

space and tine are adequate to their objects.

the question of whether our concepts of space and time have

anything in coi^uon and whether they have, together, anything

in common v/ith anything else.

This view involves the question

v;e

of xvhether, in dealing with space and tine,

are dealing

with two unique entities, or whether

vje

are forming discur-

sive concepts of spatiality and temporality, that is, whe-

ther

v/e

are bringing under single concepts of space and tirae,

a plurality of spaces and times, respectively. For Kant,

of course,

tlic

process of conceptiialization alv;ays involves

uiic-er

the bringing of particulars

one concept,

lie

does not

use the torm "concept", except for the occasional slip, in

a manner which implies that we could have a concept of sone-

thing absolutely -unique,

/md he rejects the view that space

and time are discursive concepts.

Secondly,
v/e

may take the view, with

Kaiit,

that space

and time are not concepts.

This means that they are entities

We cannot conceive them in

which are absolutely unique.

terms of properties which they have in common with other entities.

For if they have anything in corjaon with something

else, they can be conceptualized. It should be added that

we are speaking here of the intrinsic properties of space

and time.

In so far as we cannot fora concepts of the intir^e, vje

trinsic natures of space and

must admit that they

are absolutely unique entities.

26
iiegarding the distinction between space and tiue,

we have, then, four main possibilities, which may be ordered

as rollov;s:
1. Jpace and time are adequately represented by dis-

cursive concepts which have nothing in

corainon.

2. Jpace and time are adequately ropresexited by dis-

cursive concepts which have something in

coiiimon,

3. Space and time are not re presentable by concepts,

but are unique particulars i^ith v/hich we can be acquainted.

Ui

Either space or time (but not both) is re pre-

sentable by a concept.
There is, ol course, a fifth possibility, which could
be expressed as follovjs: "Iveither space nor time is repre-

sentable by a concept and neither of them is an entity with

which we can be acquainted".

This would be the view of those

(e.g. Vaihinger) who hold that space and time are ''fictions".

But since this position is remote froiu the

vievv's

of both

Kant and Bergson,

vje

need not discuss it further.

From the point of view of maintaining a fundamental

distinction betvjeen space and time, the first of the above

possibilities offers obvious difficulties. These are analogous to the difficulties concerning the relation between
two substances, which
v^e

have mentioned.

If space and time

are adequately represented by concepts and these concepts

have nothing ivhatever in

corrjaon,

we should never be able to

27

conceive ox any relation between the t\w,

/e

could never

have a rational knowledge of how space and time are related,

for example, in the phenomenon of motion.

If this v;ere,

indeed, the case, it would be a highly undesirable state of

affairs from the point of view of human knowledge.

For al-

though

v;e

could lAnderstand space and time by themselves,

tvjo

our knowledge of the world v;ould display these

ted aspects, and so create a dualism which
v;e

unrela-

could never

resolve.

It should be added that if space and time are not

representable by concepts, we should not expect to have a

rational knowledge of their relations.
But if they are re-

presentable by concepts, we should, in metaphysics, expect

to have rational knowledge of their uiutual relations.
But

if these concepts have nothing in comraon, this expectation

would be frustrated.

This would seem all the more intoler-

able just because we were able to conceive space and time,

by themselves, adequately.

Ivnowledge v/oiild be bifurcated;

motion and change would be inex licable.

The second possibility represents a denial of any

fundamental or intrinsic distinction between space and time.

For it is clear that if the intrinsic or essential natures
of space and time have something in common,
v/e

can form a

concept of the essential nature which v;ill represent the

essential nature of both space and

v/ill be

ti;ae,

A single concept

adequate to repre.sent v:hat is essential in both

2g

apace and time,

Sxtensiveness, perhaps, would fulfill the

This second possibility is,

requirements of such a concept.

in fact, the one taken by those who uphold the theory of the

manifold.

Since this position x*ill be examined in detail

belov;, we leave it aside here, except to remark that the

question of

vjhat is

essential to space and time, and whether

space and time nay both be represented by concepts is a

highly debatable one.

It is immediately apparent that if
xve

choose the

third possibility, we rule out the possibility that space

and time resemble one another.

For if they resembled one

another, they v;ould not be unique entities, and the second

of the first
tvjo

possibilities v;ould be true.

v;e

In other words,

in choosing the third possibility,

make a fundamental
In this case we cannot

distinction between space and time.

expect to be able to understand the relation between space and time, although
v;e

may well be able to become acquainted

The extent of our under-

with it through some other means.

standing of the relation between space and time will be to

point out that
the;J^

do not resemble one another and cannot The metaphysical im^^'ay

be conceived in terms of one another.

plications of this position are as startling in their

as are those of the first possibility.
viev/,

According to this

space and time are simply two unique entities which

we may meet with in the course of experience, having no

29
conceivable relation to each other or to anything else.
In

view of

tlie

indubitable fact that space and time are

oui'

amont:;

the most pei'vasive aspects of

experience, this view

would be strange indeed.

It seems small comfort to note

that the relation between space and time would seem, on this
view, no more inexplicable than space and time themselves.

The task of metaphysics, here, v/ould simply be to point out

this state of affairs, and to attempt, perhaps, throup;h the

non-cornitive fujictions of language to communicate something

of the feel of liltiraate irrationalities.

Metaphysics could

proceed only in

r.

manner quite unlike the metaphysics v/hich

attempts to give an accouuit of the rational structure of

the vjorld.
The fourth possibility offers more scope to the in-

tellect.

Here either space or time will be considered to be

adequately represented by means of concepts, while the re-

maining one

v;ill be

considered an intrinsically imique enThat Bergson's philo-

tity, V7hich escapes concexotualiaation.

sophy falls under this possibility can easily be seen.

he considers space to be conceptualizable but time not.

For
The

other alternative of taking space as a unique entity and

leaving tine to the intellect,
X'^fould

have the rather unplau-

sible result that geometry would have nothing to do with

space, however accurately it might be held to represent tem-

poral determinations.

V/ithin this fourth possibility.

30

Bergsons choice seems to be the obvious one.

the point is not so obvious as to be

nevertheless,

massed without argu-

ment/

Me

s.'iall,

therefore, in subsequent chapters, present

Bergson's arguments on this point, and the relation of his

views to those of Kant,
It will be apparent from the foregoing tliat a tlieory

which deals

ivith the

problem of the fundamental distinction

between space and time must take one of the four possibilities
listed above.
Since neither Kant nor Bergson thinks of both

space and tine as adequately representable in concepts, poss-

ibility one is not useful for our purposes.

So we shall dis-

regard it.

Possibility

t\vo

represents a

viexv

opposed to any
And since Kant

intrinsic distinction between space and time.

and Bergson both v/ish to maintain such a distinction, this

possibility can also be eliminated from our discussion.

should be added that neither of possibilities one and
tv;o

It

can

represent Kant's views, since he repeatedly emphasizes that

space and time are not concepts.

There remain, then, possibi-

lities three and four v;hich provide the general means whereby
Kant and Bergson might explain the distinction between space

and time.

It will be observed that, within these limitations,

at least one of space and time must be recognized as a unique

entity not adequately representable by means of concepts.

It follovv's, moreover, that for a philosopher who holds that

geometry is applicable to, and indeed, requires an intuition

31
of, space, it will not be possible to maintain that space

is that unique entity which absolutely eludes conceptuali-

zation,
;vithin possibilities three and four, there is l?oom,

however, for a certain amount of eclecticism.

Space, for

example, might be considered to be a unique entity, but ana-

lysable in terms of intrinsic, derivative properties, v/hich

are represented by concepts pertaining to spatiality, the

discursive concept of space.

This would involve the doctrine

that the plurality of spaces, from which the concept of spa-

tiality is derived, are parts of the one imique entity, space.

As a whole it would be impossible to form a concept of space,

for there would be nothing else like it, which it could resemble.

There would be no properties common to space as a

whole and to something else, which would allov; us to form a

concept.

There

xi?ould

be nothing to prevent us, hovjever,

from forLiing a concept of spatiality from the common properties of a plurality of the parts of space.
And in so far as

what is true of the parts of a homogeneous medium is true

of the whole, our knovvledge of spatiality would be true also
of space, even though we could not directly have a concept

of it.

But such eclecticism will have sharp limits.

It

would not be possible to carry out the same procedure for

both space and time.
That is, it would not be possible to

say of both space and time that they vjere unique entities.

32

but both capable of oeing adequately represented in terras

of concepts which have any property in common, for example,

that of extensiveness or divisibility.

An eclecticism of

possibilities one and three might be possible if it could

be shown that the intrinsic properties of time had nothing in
coiranon

with those of space.

But if it i^ere supposed that

possibilities two and three could be combined, this v;ould

involve a clear contradiction,
i^'or

such a view x;ould be

committed to saying that space and time xvere unique entities

which had nothing in

corrjnon,

but could be adequately repre-

sented by concepts which had something essential to space

and time in common.

For this reason, we have not listed

this supposition as a legitimate possibility.

It might be objected to the above scheme that it

narrows the available metaphysical possibilities to the point

of absurdity. In one sense only, this is true, and the sense
xve

in which it is true, that

have narrowed
T|ie

dotvn

the possivie

bilities, is, we maintain, not absurd,

principle

have accepted is this: that eitner tine is essentially si-

milar to space or it is not.

And we confess to having enough

faith in ohe law of excluded middle to hold this proposition

true, simply as a matter of logic.

To sum up, then, a metaphysical accoimt of the dis-

tinction between space and time must take one of the three

possibilities we have indicated, or some penrdssible eclectic

33

combination of them.
tlie

It must, in aluioion, take account of

tiiiie

way in which space and

are distinguished in

x,he

special scionces.

uith reference to the latter requirement,

it will be necos.iar^'' to state in what the distinction con-

sists.
To return to our
ori;::;inal

point of depart ux-e, the

caii

question

ox'

the unity of experience, it

be seen that the

above possible ways of distinguishing beUveen space and time

impose conditions on the possible

ansv.'ers

to tais questioii.

If, for example, it is held that space and time are not re-

presentable by concepts, and yet it is also held, as Kant

seems to do, that experience exhibits a spatio-temporal unity,
it v/ill clearly be impossible to ^ive a rational account of

this unity.

The question of x^hat relations between space

and time can be conceived, i/ill, in terras of this position,

be meaningless.
The question of hov/ space and time come to

be united in one natur-e mist remain uixanswered.

The fiirther

question of

hov;

two unique entities, which do not admit of

a common conceptual representation, can be represented in

tenras of one

another will likewise be unansv/erable.

Also,

how it is possible that elements of experience which are

in time alone, can be spoken of and represented in tenas

which are appropriate to spatial elements,

which ivill be impossible to resolve,

v;ill be a

problem

we shall find that

34
these are quoGtious of a kind which hant, in accordance with
the dictates of his position, gives no answers.

35
2. 31milaritie3 and Differences bet;veeu Space aiid Tirae

The analogouG properties of space

aiid tirae

were notv/o

ticed by Locke,

"i;ho

distinguished space and time as

dis~

tinct, simple idcac. To conclude: expansion and duration do mutually embrace and conprehend each otlier; every part of space being in every part of duration, and every part of duration in every part of expansion. Such a conbination of two distinct ideas, is, I suppose, scarce to- be found in all that great variety i;e do or can conceive, and may afford matter to farther speculation. IB

Locke woiild have been aaazed, perliaps, to knov/

hov/

far speculation in this regard has been cax^ried out since

his day.

The ciathenatical treatment of space and time pro-

vides a conspicuous exaciple of the emphasis on the similarities betv;oen space and time.

Without coniniitting oneself to

the doctrine that space and time as they are dealt with ma-

thematically in physical science constitute, v/ithout qualification, what space and time are, it is possible, nevertheless, to maintain the philosopiiic importance of their simi-

larity in this context,

whatever theory is adopted on the

question of v/hether they are ultimately distinct or not, the

mathematical treatment of space and time cannot be ignored.

One of the strongest arguments in favour of tliis view is the

indubitable success of the mathematical treatment in allowing predictions to be made concerning spatio-temporal events in
the world.
Yet it should be noted that this point does not
viev;

necessarily involve the

that space and time are ultimately

36
or intrinsically similar,
v/hitehead's
fch-3ory hov/

of space and

time provides an interestin.^ example of

it may be pos-

sible to uphold a fundamental distinction between space and

time, while tailing advantage of the mathematical trer.tment

of thsni through their analogous properties.
In his criticism

of classical physics, Whitehead at the same time grants the

success of its uethods, even though, for him, they create

philosophic problems which demand reconsideration of the entire concept of nature.

\ie cannot T/onder that science rested content with this assumption as to the fimdamontal elements of natixre. The 2^3at forces of nature, such as gravitation, were entirely determined b^' the confi.'^urations of masses. Thus the circle of scientific thought v;as complete.ly closed. This is the famous raechanistic theory of nature, v/hich has reigned supreme ever since the seventeenth century. It is the orthodox creed of physical science. Furthermore the creed justified itself by the pragmatic test. It worked. 19

V/hitehead sav; clearly that part of the success of

this theory was due to Its taking advantage of a common fea-

ture of space and time.

This allov/ed a conception of matter

as having the property of simple location, v;hich './hitehead

e:cplains as "one major characteristic which refers equally

both to space and to tine, and other minor characteristics 20 vjhich are diverse as between space and tine". He explains
this common feature as follows:

37
The characteristic comuon both to space and time is that material can be said to be here in space and here in time, or h ere in space-time, in a perfectly definite sense which does not require for its explanation any reference to other recions of space-time. Curiously enough this charc.cteristic of simple location holds "whether we look on a recion of space-time as determined absolutely

or relatively,

21

Whitehead considers the assumption of the simple location of matter to be a fallacy if it is taken to express a

philosophical trutli about the v;orld.

At be.t the assumption

is an abstraction which mathematics is eminently capable of

expressing.
The seventeenth century had finally produced a scheme of scientific thoucht framed by raathenaticians, for the use of mathematicians* The great characteristic of the mathematical mind is its capacity for aealing with abstractions J and for eliciting from, them cloarcut demonstrative chains of reasoning, entirely satisfactory so long as it is abstractions v;hich you v/ant to think about. The enormous success of the scientific abstractions, yielding on the one hand natter with its simple location ih space and time, on the other hand m^a7~perceivi.ng, suffering, reasoning, but not interf ering , h^s foisted onto philosophy the task of accepting then as the most concrete rendering of fact, 22

What mathematics is concerned with, however, is "the

investigation of patterns of connectedness, in absoractiou

from the particular relata and the particular modes of con23

nection".
of space

It is not surprising, therefore, to find aspects

find

time, in v;hiteheads philosophy coalescing in a

corjinon

space-time continuum under che guise of their

charac-

ter as extensive relations ,

Yet Ivhitehead maintains an ultiIn their

cor.iuion

mate distinction betvieen space and time.

3^

character of extensity, space end

as "the extensive continuum".

tirie

are better npoken of

But, "we must not proceed to

conceive time as another

fom

24
of extensiveness"
IHiitehead

realizes quite clearly that the concept of space-time, conceived as a network of relations, obliterates the distinction
betvjeen space and time.

But this exhibition of the actual universe as extensive and divisible has left out the distinction betvieen space and tiiae. 25

There is a vjay out of ohis difficulty,

Ilathenatics,

which deals

v;ith the

extensive continuuw, involves only exThe relata in

ternal relations, abstracted from the relata.

concrete nature are internally related.

It is in the concrete,

internally related events of nature that actual time is laanifested.

'Whitehead's treatment of space and tine illustrates

the emphasis on their analogous properties from the standpoint of pure relation, and in a''dition shov;s how these ana-

logies may be dealt vjith in metaphysics without tliereby explainin,;^ av^ay a

real difference betv;een space and time.

It

is not to the present purpose to discuss the particular man-

ner in which Whitehead has accomj-jlished this feat, but merely

to point out that he
sa'iv

the necessity of doin^ 30.

This is

all the

laore

interesting in that V/hitehead's accoiuit of the

analoi^oua properties of space and time, particularly in his

early v;orks, 3 strongly influenced by his conception of

39
mathematics as a formal science of pure relations.
logiss for
not to be
hi3i

The ana-

become formal analogies, and as such turn out

at all of f;mdamentally different entities,

a nalo.f^ies

but are replaced by one relational complex, the extensive con-

tinuum, which constitutes all the spatio-temporal relations

betivsen concrete occasions. This is a rather different 'nethod

of attack from that of the pure matliiematician who, if he cares

to, may study the aspects of those relations usually acknow-

ledged to ba characteristic of time, and those usually acknowladged to be characteristic of space, separately,
tirithout

necessarily coiaraitting himself as to whether these relations

as QX9 aplif led in phenomena make up one complex whole.
It is,

however, an understandable step to take, and may, perhaps,

be the only possible one if the mathematical treatment of

space and tine is to be taken seriously.

It is interesting-^ to note in this regard, hov/ever,

that from the standpoint of pure geometry the fusion of space

and time may taka place in either of two ways.

Geometries

based on relations taken to be characteristic of physical

space
.':aay

incorporate tiwe as a fourth dimension; or rela-

tions characteristic of time may be considered to be prior,

and relations characteristic of space may be derived from
then.
At least, in one case, an attempt has been laade to

build up a geometry of four dimensions of the kind required

by theoretical physics, from the relations of before and

40
after, \.hich are taken to be the fimdanental ones ch^irac-

tcrisoic of tine.
Thus instead of re^arc'ln^ ourselves us, so to speak, swimming along in an ocean of space (as we usually do), we s.re to think of ourselves rather as scnehow' pursuing a course in an ocean of time; x^hile spacial relations are to be. rer;ardeQ as the rianife station of t h e fact thaT th e eTements of t ine f oriu a svstem in conical order ^ couc:vi,::L.lcn );!dc]i u^^y be ctualyned ij_;_ ten.::; oi" " the rela tions of before ancTafterV 2'6
:

The author of this geometry builds it up by a fonaal

procedure of deducing theorems fron postulates: "The geometry

as I have pointed out is a loyical structure built up from

certain postulates which

shall formulate".

Thus, if this

author has been successful, it would seem that relations taken

to be characteristic of space need not, from the mathematical

point of viev;, be considered to be fundamental.

in the v/ords of G.
D. Broad,
",
,

Vie

need not,

.talk, or listen to, non,

sense about Tiine being a foiorth dimension of Space*"

in the context of mathematics.

even

What, then are the analogous properties of space and

time, and v;here does the analogy v/hich they su-gest break off?
In addition to one characteristic to which V/hitehead
iias

dravm

attention, namely, that we can speak meaninj; fully of a "here"

in time as ivell as in space, there are a group of properties

which are thou^iht to be

cominon to

both space and time, and

which have often been referred to as their "extensiveness".

G, D.

Broad sums up these characteristics in an illustration

41

designed to drav; attention to the extensive aspect of time.

How the te:v;'Oral relations v;'r',ich vje perceive anong events are similar to the relations of partial or complete overlappin^j vmich we can perceive in the case of tv/o extended objects, like a pair of sticks. The possible tinerelations betv/een tivo events can bo completely represented by taking a single straight line, letting "left -to-right" on this stand for "earlier and later*', and taking tv/o stretches on this line to represent a pair of finite events, 29
Time, like space, soems to adnit of quantitative de-

terminations.

This was also

brou'::ht

out hy Locke, \vho, al-

though he regarded space and time as two distinct ideas,

thought that they might be nore clearly and distinctly'- knotvn
if they were compared with one another.
In both of these (vi?;. expansion and duration) the mind has this comriion idea of continued lengths, capable of greater or less quantities; for a man has as clear an idea of the difference of the length of an hour and a day as of an inch and a foot. 30

Locke also touches upon other aspects of the exten-

siveness of space and time, which have their counterparts in

Bread's representation.

For exarrple, Locke says that "All

duration are duration";

the parts of extension are extension, and all the parts of 31 in other words, space and tine in

their character of extensiveness are homogeneous.

ther
ccETT.on

Two fur-

proper-ties of the extensiveness of space and time

are mentioned by Locke.

Being influenced by I'lewton's theory

of absolute space and time, he emphasizes that both space and

time can be conceived of as infinite.

Expansion is not bounded

hj natter, nor duration by motion, but each may be thought of

42

as uriending once the idea o a particular length of ohe or

the other has been seized upon.

As a corollary to the infinite extension of spcce and time, Locke aads that they admit of finite determination, that is, they are divisible into particular lengths.

Time in general is to duration as place to expansion. They are so much of these boundless oceans of eternity and iiamensity, as is set out a.id distinguished froni t^e rest as it vere by landmarks; and so are laade use of" to denote the position of finite real beings, in respect to one anoth<3r, in those uiiiform infln^' -^ -'^ ^=in3 of duration and space, 32

All of these geaeral aspects of extensivenoss, viz,

homogeneity, divisibili&y, the possibility of infinite extension,
a:id

simple location, may be regarded as analogous

tiiae#

properties of space and

Whether s'trict adherence to

tlie

such a vievj is defensible in

final analysis, however, is

another question,
points out,

iinphasis on these properties, as Broad

is apt to obscure the very real difference be-

tween space and time.

At first sight the problems of Time look very much like those of Space, except that the single di-mension of Time, as compared with the three of Space, seems to rromise greater simplicity, V/e shall point out these' analogies at the beginning; but we shall find that they are somewhat superficial, and that Time and Change are extremely difficult subjects, in which spatial analogies help us but little. 33

If space and time are looked upon as certain kinds

of relations exhibited in experience, the differences between
the''"
hf^rio. n

fvir:cnt.

The gt.icce.isiveness and irreversibility

43 of the temporal order do not eeem to have an obvious counter-

part in the spatial order.

The relations themselves are

narhedly different.

Broad records that the triauic relation

of "between" occui's both in the linear spatic^l series and in

the temporal series, but emphasises that there is a profound

difference in the "betweenness" in each case.

Temporal re-

lations are analysable, Broad holds, In terms of a single,

asy^nmetrical dyadic relation.
is an intrinsic dii-ection, or
'

In the temporal series there

sense". In the spatial series

there is no intrinsic direction; direction can only be intro-

duced extrinsically, v/ith reference to some point outside the

series.
In addition, as Broad points out, space seems to pre-

suppose title:

Spatial extension and the occurrence of spatial relations presuppose temporal duration and a certain deterip-inate forsi of temporal relation. Shape and size are commonly ascribed to particulars which persist through periods of time and have histories of longer or shortar duration. 34
Robb expresses the difference between spatial and tem-

poral relations in texnns similar to those of Broad, except

that where Broad considers the spatial relation of "between"

to be partly symmetrical and partly asymraetrical, Robb omits

this point,

Kow in considering the subject of time as it presents itself to our experience there is one very important respect in which it appears to differ from our spacial experience. L*"J Of an3r two instants which one experiences in one s own mind one is after the other.
'

44
i'iiis relation of after is what is called an asym metrical relation by which is meant a relation ?. such that if B bears the relation it to A then A does not bear the relation R to B. ihus, in the parliicular case considered, if B is after A, then A is not after B, ihere are however relations which are syrjnetrical; such, for e>:ar.iple, as the relat:.on of e<^iialit. ^. where if 3 is equal to a then A is equal to B, i'ovj the relation of two points or two particles in space is a syuiuetrical relation and, if A and B be taken as tv;o distinct points, t^iere is no reason v/hy we should say that B is after A rather than that A is after B, 35
;

Pointing out that

an irreversible one
,.

tlie

thought proces s is essentially

iiobb

goes on to offer an explanation of

the peculiarity of time relations which gives them their asym-

metrical character.

The fundamental fact, for llobb, about

these relations is that an effect is always after a cause.

A present action of mine may produce some effect to-norrow, but nothing v.'bich I nay do nov; can have any effect on what occurred yesterday, 36

Thus Roub seems to take the causal relation as the

fundaiaentul one, which enables us to distinguish between

spatial and temporal relations.

It may be doubted in passing,

however, whether this is entirely defensible since it does

not

seei':

to be

tlie

case that all temporal series involve

causation.

The pain in my toe may be after my perception of

the flash of light, but does not seem to have been caused

by it.

The important points brought in ftobb's treatment of

temporal relations, however, seem to be the same as those

mentioned by Broad, namely, the intrinsic successiveness
and irreversibility of the
tii.ie

series.

It is these which

45

mark

tiiae

as different 10^.
cOiiiplete

i:,jj.ctif

uo

>viiao

u .e

attempt to

establish a
breaks doim*

or intrinsic analogy

befcv;een the:Ti

We must now consider, however, the arguiaents of

those who hold that the analogies between space and time do not break down, and that to a far greater extent than
v/as

formerly raalized, what

caii

be said of space can like-wise

be said of iiiue, and vice versa

This, we take it, is a

correct statement of the position of those who deny any intrinsic difference betvjeen space and time. We shall examine

first svme of the arguaients preseiited by Taylor, who a'ives

a clear statement of this position,
I want to remove sone of this mysteriousness

(of tine) by showing lations, contrary to radically aliks; or, ordinarily used in a

that tennoral and spatial remuch traditional thought, are more precisely, that (1) terms peculiarly temporal sense have spatial coijinterparts and vice versa, and that accordingly (2) many propositions involving temporal concepts which seen obviously and necesgaril]'- true, are just as necessarily but not so obviously true v/hen reformulated in terns of spatial relations; or, if false in terms of spatial concepts, then false in terms of temporal ones too, 37

Taylor makes use of the analogies already mentioned,

namely, simple location, and extensiveness involving divisibility into parts,
3ut he also introduces a supriosed

analogy in terms of direction , which denies Broad* s point

concGmtnj; the intrinsic direction or sense of the tineseries.

A^ain, the notion of direction has a use v.'ith respect to both spatial and tenporal" relations; one can, for irutance, speak of t'.;e diroction Troiij past to future, from future to past, frora north to south, and so on, none of v/hich di:.'ect Ions is any nore or less genuine or intrinsic than the others. 3^

Taylor makes

tliis

statement without attempting to

support it in any way.

This is unfortumate for his case,

since most of his proofs of analogies between space and

time

hinf-;9

on the truth of it

It is obvious that if we

deny the essential characteristic which distinguishes space

froin tirae, a

number of supposed analogies follow

analj-iiihov/

cally.

This raises the interesting question of

shov.'

one

could
It

that there are analogies between space and time.

seeiiis

clear that the final appeal laust be, broadly speaking,

V;e

to experience.

need not concern ourselves here as to

l/hether we mean "sense-

just what "experience" comprises,

experience'-'

(sooing, hearing, feeling, tasting, etc.), "cogni-

tive experience" (perceptual judgments, intellectual pre-

hensions, etc.), or experience in sone other meaning, is not

so important as that
vie

must have

sorae

non-arbitrary court

of appeal v/hich

iirill

enable us to distinguish betvjeen genuine

analogy and

laere

assumption. It is not unlikely, of course,

that there will be disagreement concerning the legitimacy or

possibility of experience in its various meanings. But it

seems clear that the responsibility of those who assert analo-

gies between space and time, or analogies in any context, v;ill

be to indicate vjhat experience of what kind illustrates the

47
the analogy which they assert.
In the absence of an attempt

to do this, we may feel justified in returning the Scottish

verdict: "Not proven".

No matter what analogous stateuents

\:e

can nake con-

cerning space and time, no analogy between space and time

themselves will be demonstrated until it has been shown that

such statements are synonymously true, that is, true in the
same sense.

They need not be entirely sjmonymous, perhaps,

but they must be synonymous in just those respects v/hich

make them analogous.

That is,

it.

mupt be shown that the

term "length" or the term "part", for example, has the same

meaning when applied to space and time respectively, and

that true statements can be made involving this meaning in

both contexts.

And this involves, of course, showing that

such terms, if meaningfully applied to space, do in fact

have a meaning when applied to time.

If they have no mean-

ing v;hen applied to time, there can be no synonymy, and

hence no analogy.

But S3monymy can only be finally shown

by appealing to experience, in some acceptable sense of the word.

Thus, for example, we say with perfect candoiir that

some two-dimensional spaces, i.e. patches, are red, but when

we say that some times are red, the statement seems to be

nonsense,
u'e

have no notion, in terms of our experience,

of what a "red time" as opposed to a "green firae" or a

"yellow time", might mean, although we have no trouble in

46

experience distinguishing between a red patch and a green or

yellow patch,
V/e

are not entitled to assume synonymy of ana-

logous statements concerning space and time, v;ithout reference

to our experience of space and time.

Thus, if
xje

say that time has no intrinsic direction,

we must look to our experience to find out whether this has

a meaning.
V/hen

we do this, it seems that the statement has

For we knov/ what we mean when
vje

no clear meaning.

say that

an observable spatial series has no intrinsic direction.

Part, at least, of what we mean is that when observing a

series of objects in space,

v;e

can, at will, with a flick

of our eyes, run through the series from either end, indif-

ferently, or perform similar operations tactilely.

But if

we are observing a series of events in time,

v;e

find that we

cannot, at v;ill, run through the series from either end.

This is not merely a practical difficulty, but is a logical

impossibility, for running through a temporal series from

either end involves holding that a past event can be present

to an observer.

That is, in order for an observer to run

through a temporal series from either end, it is necessary

that events
\irhich

are past to the observer should be present

to him so that he can run through them in different directions.

But it is a necessary although not sufficient condition for

an event to be past to the observer that it should not be

49

present to him.

He must then observe an event v;hich is both

present and not present to him.

iJhether ue consider his ob-

servation of this event to be instantaneous or to take an

enormous lapse of time is irrelevant.

For his act of ob-

servation must be considered to be a single act, by vihich

he observes, i.e. has present to him, a single event that is

not present to him.

w'e

add tiiat a necessary condition of

.at

being present to an observer is t

the observer should

have a sensation, (not necessarily atomic), which is cor-

related with the fiinctions of his receptors in such a manner

that it would not be called a menory.
If it be insisted that the observer's act of obser-

vation does take some tine during which the same event will
be both present and not present (past) to him, and so in-

volves no contradiction, it should be noted that if the event

becomes past to him, he is therefore not observing it.
Hence

his act of observation of the event comes to an end, just

when the event becomes past, i.e. not present to him. It will
therefore not be correct to say that he observes an event

which is both present and not present (past) to him.

But

the observer must make just such a logically impossible ob-

servation in order to run through a temporal series of events

from either end.

It is tempting to suppose that the same condition

might apply to the observation of a spatial series. This,

50

however, is not the case, for it is just the peculiarity of

spatial configurations that they can and often do remain present to an observer during the time he runs through them in

either direction.

In a single act of observation an observer

may run back and forth over a spatial series that is present
to hin durin.^ the entire time.
As in the case of a temporal

series, it is irrelevant how long a single act of observation

may take.

It may be considered to be instantaneous, or to

take an enormous lapse of time.

As a matter of fact the spe-

cious present is fairly short, but that has no bearing on the

theoretical aspects of the case.

as short or as long as
v;e

The specious present may be

please in terns of units of measured

time; the point is that the xvhole series of spatial configu-

rations is present to us during the entire time.

tv/o

If we hold

fingers in front of our eyes,

v/e

can run back and forth

ive

over them in a single act of observation, and while

doing so, they are both present to us.

are

They do not disappear

as a past event disappears. They are definitely not past to

us.

They are present.

It may be objected, hovjever, that
v/e

are using the

tenn "present" in a non-temporal sense, that we are using

it only in the sense of "immediately apprehended", witnessed,

or sensed.

We agree, of course, that when

vjas

xi?e

said that an

object or event

present to an observer, we meant that he

But this does not prove that

was aware of certain sensations.

51

the usage is

rxot

tsnporal.

It does, however, reveal the dif-

ference between time as conceived in terms of units of measurement, divisible into parts, and time as it is felt .
The ob-

ject or event which the observer saw was not present to hin
in the sense it was an instant or length a certain number of

units removed from some fixed date, it was present to

the sense that it was not future, and not past .

hira

in

In other

words, the difference between the ojserver's apprehension of

the temporal series and his apprehension of the spatial series
is that the fornier involves a reference to the future and the

past, while the latter does not necessarily do so.

The pro-

gress of the event from being future to the observer to being

present to him, and to being past to him,
is ;vhat raade hira

unable to run through the temporal series from either end.

We therefore maintain that the primary, irreducible,

and intrinsic temporal meaning of the term "present" is that of the experience of the present, which involves the distinction between the present which we do experience, and the future and past xvhich we can never experience.
speak of time or temporal relations at all,
vie

If we are to

must be pre-

pared to accept the temporal distinctions which are revealed

in experience, and these, ox course, are the distinctions bet\T/een

past, present and futui'e.

Me have an unmistakable

sense of the difference between the present and the past, and
betv;een the past and the future, and between the present and

52

both the past and the future.

First of all, we are never

in doubt as to v;hether or not we are in the present.

We can

always distinguish the present from the future and the past.
The future and the past are, quite simply, not present,
V/e

have an unmistakable sense of the difference between the fu-

ture and the past, however, and the clearest proof of this
is that we can remember the past, but not the future.

Russell

says that "it is a mere accident that vie have no memory of 39 the future". But an accident due to what? If we refuse to

accept the fact that we do not have a memory of the future

as prima facie evidence of the temporal distinctions betv^een

the past and the futuj*e, we shall never be able to decide the

question of whether the time series has an intrinsic direction

or not. For our awareness of the difference between past,

present and future, as Broad points out, is intimately associated with the intrinsic direction or sense of time.

How the intrinsic sense of a series of events in Time is essentially boimd up with the distinction A precedes 3 bebetv;een past, present, and future. cause A is past when B is present, 40
Furthermore, if the theory of the four-dimensional

manifold is true, it is not clear how anyone could possibly

know that we do not, in fact, have a memory of the future,
v;e

take for granted that memory gives us an awareness of

events which are not present to us.

Whether this av/areness

is "direct" or not, does not matter for the present case.

53

as Ions as it is granted that memory gives us avjareness, in some sense, of events which are not present.
Let us say,

then, that the occurrence of a memory A has a^memory relation"

to the occurrence of another event B.

Let us suppose that A

is present to some subject who is sitxiated at some specifiable

place in the space-time continuum,

Mov;, hov/

this subject

could know that A is related to B in one direction of the

time -dimension rather than to another event
C in

the opposite

direction of the tirae-dinension, is an interesting question.

What possible criteria could he use to decide?

Certainly,

he could never decide the question empirically, for the posi-

tions of A, B, and

C C

are fixed in the space-time cortinuujn,

are present to the subject.
He could

and neither 3 nor

never compare A with either B or C, as to likeness.

It is

tempting to suppose that at a later specifiable place in the

subject's history, say, at the point where 3 or C occurs, if
his history happens to include either of these points, he

might carry out a co-:iparison.

But, of course, A is, again,

not present at B or C, and no empirical check is possible.

It might be objected that memory requires repetition,

and that we could easily define a memory as that event vihich

followed a suitable number of repetitions of a certain experience.

This would not be

valid objection.

First of all,

memory does not alvjays require repetition, Secondly, the

objection does not hold anyway, for obviously
v^e

need the

54

memory relation to relate the memory to the last of the repetitions, v/hich by definition is not present to us,
i.'e

are

still forced to accept a memory which is present to us at

some point of the time-dimension, and v/hich
knovir
\je

still do not

to relate to an event in one direction of the tineoth:::r.

dimension or the

Incidentally, no sort of psycholo-

gical explanation will answer, for it will presuppose that

memory relates to the past, that is, one temporal direction rather than another, and further, is likely to involve the
use of memory itself, either in its explanation or in the

empirical investigation on which it is based.

It seems evident that if we do not know the difference

between the future and the past, we can never know to v;hich
of these ovt memories relate.
It is to be noted that we can-

not put this the other

v^ay

around.

Although the distinctions

between past, present and future must be revealed in experience,

(othendse we could never know of them at all),

xre

cannot define these distinctions in terms of our cognitive

acts, such as perception and memory.
As regards memory,
vje

cannot define the past as that to vjhich our memories relate,

because

v;e

will still not

knov/ xfhether the past is in one

direction of the time-dimension or the other.

For the same

reason we cannot define the future as that v/hich we do not

or cannot rea^iber, and which is not perceived by us.
In

addition, we shall always have difficulty in distinguishing

55

that which we cannot remember because it is future from that

vrhich we do not remoaiber because our memories are fallible.

Moreover, as 3road indicates, every event that an observer

knows of must be capable of being both remembered and perceived.

Hence these cognitive characteristics do not suffice to distinguish a past from a present event, sinc3 every event that knoi^s of has both these relations to him. If you add that an event always has the perceptual relation to C '...efore it has the memory relation, you only mean that the event of remembering something is prssent vjhen the event of perceiving it is past, and you have simply defined present and past for 0*s ob.iects in terms of preG?nt and past for his cognitive acts . If you then try to define the latter in tarns of different relations to 0s acts of introspection, you simply start on an infinite regress in which past and present remain obstinately uiidefined at any place v/here you choose to stop, 41

These difficulties concerning memory and the attempt

to define the past, present, and future, are connected with

the view that the analogies between space and time do not break down.
The supposition that it is a mere accident that

we have no memory of the future, presupposes that past, present and future do not represent unique temporal disi^inctions,

And the vievj that we can define past, present, and future

likewise assxmes that they are not unique distinctions, but

can be understood in terms of something else.

We must now

turn to some ox the specific arguments xvhich Taylor advances

to show that space and time are analogous entities.

We shall

not examine all of his arguments, since most of them rest on

the same assxamptions. If
vje

can question these assumptions

56

in one case, we can do so in the others.

We have already noted that Taylor does not prove

that
tii:.e

has no intrinsic direction, hut assumes it.

We

also note that he likewise does not prove synonymy of the

terms "place" and "part" when applied to space and time respectively.
Vn'ith

these things in raind, we proceed to ex-

amine his first arg;ument.

Taylor's procedure is to state

objections which are coniiuonly thought to express a radical

difference betvjeen space and time, and then to
shov/ that

the

objections do not express such a difference.

jection is as follows:

The first ob-

iUi object cannot be in tv;o places at once, though it can occupy two or more times at only one place. A.2

Taylor ar-^ues that since we say that an object is

in one place at two times only if it occupies all the time

in betTveen, either at that place or another,

v/ell say that an

v/e

can equally

object can also be in two places at one

time by occupying the intervening space.

A ball, for instance, occupies two places at once, if the places be chosen as those of opposite sides; but in so doing, it occupies all the places between. It is tempting to say that only part of the ball ic in either place; but then, it is a different temporal part of an object vjhich, at the same placf, is in either of two times. 43
We draw attention, first of all, to the point that

if

v;e

do make such statements as that "An object occupies

tim^e

all the

in between",

v;e

are presupposing an analogy

vje

betvjeen space and time.

If

can make such a statement

57

with respect to time in the same sense, (or in any sense),

that we can make it xvith respecb to space, we r^rant the whole

analogy, including the curious term "texi-poral part",

For ob-

viously if an objecfoccupies time" in the same sense that

it occupies space, we must admit that an identical object

can be extend&d through one time while its spatial parts

are in different t^paces.

v/hatever that
v;e

This would apply to any object

^ae point at issue, hov;ever,

can perceive,

is just whether we can use such language synonymously with

respect both to space and

has any
iiieaning

tirr.e,

v/hether, in particular, it
v/e

to nake such stateaonts about time,

must

reiterate here that when we speak of time in this thesis

we are not referring to measured time or physical time, the concepts of v/hich conform to the requirement of extensiveness, but about time itself,
that

For this reason

v/e

have stressed

a theory of time must tiake account of the v^ay in v/hich

l.e

time is dealt with in various sciences,

cannot assume

uncritically that time is identical Jith conceptual tiae.

If we do.
Vie

shall bag the question ox whether or not it

Time may be a unique entity, inca-

is analogous to space.

pable of being treated of in terms of coi^cepts such as ex-

tensiveness.
The premise of Taylor's ar^Uiiient assxnies without

discussion that tine and space may be conceived in the same

terms.
It assumes that if we can say that an object occupies

5?i

space, wo can also meaningfully say that it "occupies

ti:.:e",

and that x/hen we do say this, our statement is synonymous

with the analogous statement concerning space.

This, oi

course, is just the point at issue as to whether there are

analogiGG betv;een space and

ticie.

Hence Taylor's arguments

tiiiie.

do not prove analogies between space and

v/hole question,

They beg the

Taylor's arguments are based on another assumption,

which is related to the one we have mentioned. This concerns

the identity of individual tilings.

For just as vje can an^ordinarily do say that moving about in space i,e, acquiring and losing spatial relations v.-ith other things over a lapse of time does not destroy the identity of a thing, we have equal reason to say that movin^., about in tine i,e, acquiring and losing teiiiporal relations v;ith other things over a does not destroy it either, lapse Ox space

l^J^.

This assumption again presupposes rather than aen^on-

strates an analogy between space and time, and stands or

falls on the question of whether there is such an analogy.

For wenotiathat Taylor conceives of the teiiiporal relations

of a tiling in terms of extensiveness.
It is an easy step

to suppose that a thing can change its temporal "distance"

just as it changes its spatial distance with reppect to

other things, if we assume in the first place that the terra

"distance" is appropriately used of time.
This further step

involves a repu-iiation of the point that there is an in-

trinsic direction or sense of the time-series,

.Por

if there

59
is no intrinsic sense of the tine-series which fijces the

temporal relations of a thing with respect to other things,

we may, pei'haps, suppose that there are individuals v/hich

change their tei^poral relations.

This necessarily involves

changing the conception of what constitutes the identity of

a thing.

We cannot hold that the intrinsic sense of the

time-scries fixes the teinporal relations of a thing, without

also holding that there are identical things to he so fixed
in their temporal relations. But if we deny this doctrine,
U'e

we erase both aspects of it.

cannot deny that there is

an intrinsic sense of the time-series without also denying

that the individuals fixed by this intrinsic sense are in-

dividuals,

for the intrinsic sense of the ti2:e-series is

expressed by the fact that there are certain identical individuals, (otherwise identifiable than by their temporal

relations), which are fixed as to their tenporal relations

to other things, of a like nature.

It is obviously neces-

sary to hold that these individuals are othervrise identifiable than by their relations in the time-series, for if

their individuality depended only on their temporal relations, there v/ould be no means of fixing either their

individuality or their temporal relations.

This follows

as a direct resv.lt of conceiving time after the analogy

of a straight line,

Kow can we identify individuals on a

But if the individuality

straight line, except by convention?

60

of certain things, related either in a straight line, or in

time, (conceived in this manner), is g:iven or fixed by some

other means, then the ppatial and temporal relations in

these two cases, respectively, are also fixed.

Thus the

theory which holds that time is a series which has an in-

trinsic sense,
events in
tions.
tlie

laust
ijex'ies

also hold that the individuality of the

is constitutive of the temporal rela-

Individual events, so to apeak, shoulder their way

into their positions in the tine-series.

It must he admitted that this theory
;.;iiuii

c;jncyves

of tiwe as a series with an intrinsic direction., "the sort

arrow-head on it",

of thing Me usually represent by a straight line with an 45 It seems to is rather unsatisfactory.

be a sort ox half-way house on the road to the theory of

the spacs-iinie manifold.

This is not to say, however, that

there is not an important element of truth in the viev; that

time has an intrinsic direction or sense,
is adequately or falsely expressed in
a direction, however, is a inatter
on.
v/e

s.'hether

this point

tearias

of a series with

shall discuss further

For the present, it remains to compare the merits of

this theory which does maintain a fundamental distinction

betv;een space and time, with the theory which denies this
distii-^ction,
\ie

have already drawn attention to the point that

txieri is a close connection between the view that the

61

time-series has an intrinsic sense, and the identity of individuals.

The denial of this position involves holding

that not only is extensiveness and its related properties,

attributable both to space and time, but also that time

in no way has an intrinsic direction which space has not.
V/e

have noted that Taylor has not proved an analogy in this

respect, but has assumed it.

We enquire

nov; vjhether

anyone

could possibly prove it.

If we find reason to believe that

it is iapossible to prove this analogy, and we find good

reasons for not accepting it, we shall, by this token, have

good reasons for rejecting the theory of the space-time manifold.

It is to be observed that denying that the time-series

has an intrinsic diroction involves holding that individuals are not in any sense given.
A corollary of this is the view
v/e

that it is a matter of choice or convention as to v/hat

consider to be an individual.

For if we hold that indivi-

duals are in any sense siven to us in experience, or fixed

othen^fise than by convention, and if
v/e

hold likewise that

v;e

tine is adequately conceived in terras of extensiveness,

must adr.it that the occurrence of these individuals unalter-

ably fixes their temporal relations in time.

If we want to

hold that temporal relations of things can change, we must

abandon the notion that individuals are in any sense givtn, or fixed otherwise than by convention.
It is to be noted

62
tliab

if ve do not conceive of tine in terns of extensiveThe point is that extensiveness

ness, this does not hold.

itself,

Qvta

extensiveness, is not capable of cleterr.dning

The reason for this is that

or accoionting for individuals.

it is conceived as being honogeneous, everytjhere literally

the same.

If individuals are to appear in it, their indiuv-s'c

viduality

come from somewhere else.

Extensiveness

beins hoa'o^eneous, individuals arc identifiable in it only

by convention.

Extensiveness may relate individuals, but

it cannot account for, or be constitutive of different in-

dividuals,

3ein^ every;irhere the same, there are no diffe-

rentiations in it, hence no different individuals, except

in so far as these are chosen or identified by something

outside.

As we mentioned before,

hov/

can we identify in-

dividuals on a straight line, except by convention?

those
v/ho

Hence,

hold the theory of a space-time manifold, or an

extensive continuum, are taking a logical step if they hold

that individuals are identifiable only by convention.
It

is a logical step because the individuality of individuals

must proceed from something outside the extensive continuum

itself,
iliether it is a defensible step, hov,jever, is ano-

ther question, and in this regard one might well ask how
ariyone could

know that something was a matter of convention.

It might be asked as well whether conventionality itself is a matter of convention, but a discussion of this would

63

take us too far afield,

une of the primary reasons we find for rejecting the

theory of the four-dimensional manifold, in which time is

coaceived as a fourth dimension, is that it does not seem

to be the case that all events are capable of being located in such an extensive continuum.

Some events, e.g. feelings

let, if

and thoughts, seem to take place in time alone.

the four-diaiensional manifold contains all events, all

events

..mat

ipso facto be capable of being located by three

spatial coordinates and one temporal one.

In addition, if itAheld chat there is no distinction
betv;een space and time, and that time has no intrinsic sense ,

this will mean, as

v;e

have already mentioned, that indiviIf this is explicitly

duals v;ill be a matter of convention.

maintained, as Taylor and V/illiams do, it must also be ad-

mitted that it would be fruitless to refer to experience in

order to determine Vihether time has an intrinsic direction
or noi.

Hence

v;e

could never decide this question one v;ay

or the other, v/ith res;iect to our experience of space and

time, for it would depend on our choice of individuals. We

in time, or both, as Taylor, in fact, does.

could choose individuals which move either back or forth 46 This is a di-

rect result of supposing that an individual may chanse its

temporal relaUions v^ithout losing its identity.

From the

the standpoint of our experience of space and time, this

64
is imsatisfactory,

Goodman,

v:ho

rtoarks that we must recog47'

nize "that individuals do not change their temporal relationships to each other as they do their spatial relationships"
:^ives a co^^ent
,

reason

whj'-

we nust reject the view that in-

dividuals may change their temporal relations.

This may seen to depend on a mere verbal accident. Why not simply generalize the use of "change" a little so that a thing chanr^es in a ^^iven respect if different parts of the thing have different qualities of the kind in question Because, it may be fairly answered, this ignores the distinction* betv/een a 'linutc nobile thing that travels over a given region, and a spatially large thing that occupies a comparable region at n single instant. Sach of the tv;o thin.c^s has parts that differ from one another in location; but according to ordinary usage, only the foraer undergoes change. By applying the term "change" in the one case, out not the other, ordinary usage marks an important distinction, 4^
"

'

For reasons such as these, we reject the theory of

the four-dimensional manifold in which all events are located, and
vse

retain the view that there is a fundamental

To the examination of

distinction betxveen space and time.

the nature of this distinction,
v;e

must now proceed.

P.CSLIMIN/.iiY

DISCUSSION OF

THH;

DISTINCTION

BiiTWEEN SPACE AND TIME

3,

The Direction of Time and the Problem of Individuation

First of all,

-we

v;ill

enquire how it is that time

or the time-series, conceived as a unidimensional extensive

continuum,

.["lets

its intrinsic direction,

vie

have already-

pointed out that it involves a logical contradiction to suposo that end,

V/e
v.'e

can run through a temporal series from either

should have to observe at least one event that was

This consequence

both present and not present (past).

follows from the primary temporal meaning of the term "present^'

namely, the "present" of a conscious observer.
If
v/e

re-

gard the term "present" simply as a synonym for "being at"

some point in the time continuum, of course, such a present will be both present, and past
v.-ith

reference to some

other point on
"present"
v.'ill

t?i3

continuum.
;jivo

But such a use of the term

not

us any intrinsic direction of

the time-series, for we could equally well refer to the

past in either direction.

We should not know whether our

memories referred to events in one direction of the timeseries or another, and this diificulty would present
65

66

itselx for every separate memoiy that we had.

These

are some of the points relating to the intrinsic di-

rection of the time-series, which we

course of our arj^uraent.

liave laade

in the

Finally, we maintained that human

beings have an unmistakable awareness of the distinction

between past, px-esent, and future*
It
j.-eraains nov;

to

try to show whence those distinctions arise, and hence

ho\u

it comes to be held that there is a time-series xdiich

has an intrinsic sense or direction.

It is to be emphasized that this conception of

time, conceives it as a series .

How, then, we ask, does

any seiles get a direction'^

It is apparent that no

ser*ies can get a direction if it is

given whole.

It will,

as Broad says, have an order but not a direction. The peculiarity of a series of events in Time is that it has not only an intrinsic order but also Thi'ee points on a straight line an intrinsic sense , have an intrinsic order, i,e, B is between A and C, or C is between 3 and A, or A is between C and B, This order is independent of any tacit reference to something traversin:^ the line in a certain direction, 49
It is clear that if time is conceived as only
a line,

or series , in a one -dimensional continuum, siven

whole, as a series, it simply will not have any intrinsic

direction at all.

One can run through any series qua

series, in either direction provided it is suitably

finite.

Something more needs to be added to it, in order

67

to

,-iv9 it

the intrinsic direction.

The first thou;;ht

that occurs, as broad hints above, is that of something

traversin.3 the series in a fixed direction,
I7e

have our-

selves made use of this device in speaking of "running

throurjh" a series from either end.

It is a clear enough

means of expressing the idea of direction, but in it

defensible to use it with respect to the time-series?

Can
v;e

think of time as a series of events

alon,;:;

v;hich

something moves?

Broad has expressed this possibility

in a picturesque manner.

We are naturally tempted to regard the history of the v;orld as existing eternally in a certain oi^der of events, Alon; this, and in a fixed direction, vie imagine the characteristic of presentness as r.oving, somewhat like the spot of li;:ht from a policemen* s bulls-eye traversing the fronts of the houses in a street, 50 The difficulties with this, however,
ax^e

all

too evident, as Broad, himself, emphasizes,

V/illiams

briars them out sharply vjith one pointed remark:

bubble is pricked at once by the question,
the bull's-eye move?"^

"The

'How fast does

The answer is, of course, that

it can have no velocity at all, since average velocity

is expressed as the ratio of distance travelled ov-r

length of time.
spatially.

But the bull*s-eye does not move

How can

It moves supposedly throu:;h time.

something move without velocity?

If the bull*s-eye is

conceived to be the specious present of a conscious being,

as the analogy suggests, the situation is no uirferent.
In the process of pointing out xvhy we cling to an analogy

between space and time c^fter it breaks down, Goodman finds

the following mistake:
In the third place, I think our error is nciirished by a nebulous underlying notion of the self as something that flits through tine carrying its specious present along vjith it, I h.ve no idea of discussing the nature of the self here; but whether it is or is like a thing, event, or quality or v/hatever else it may be or may be like*- and however many are the times it lights upon, the statement that it lights upon or occupies or is at different times at different times will still

^ absurd.

32

It is apparent, then, that the notion of a moving

self or anything; moving along the time-series itself,

cannot account for the intrinsic direction of the timeseries.

What remains to account for it?

The ansxver can

only be, we believe, that when time is conceived as a

series of events in one dimension stretching into the
past, it is also conceived of as having one end open in

the specious present of a self.

At this open end of the

series, events occurring in the specious present are

added to the series.

The continuous adding on of events

at one end of the series gives it an intrinsic sense. The series goes in a fixed direction because it is contin-

uously projecting at the open end.

This coincides with

69
Kin^, 's viev;,

Speakin-; o tho present

ar,

al".;ay3

chan^in^,

Kin-

oes on to say that:

have no ri~ht to assume that the Present, ..'e as the locus ox motion and becomin;, proceeds throu,e;h a series of instants like that found in the slices of that v;hich has become, and upon ivhich ive have based our theory of chan.^e. Such an assumption results, as vje shall attempt to show, in the postulation of an absolute space and time xvithin which nature v;orks. And the failure to ::uard a,-ainst this coirjnor^Sense assumption ends inevitably in the spatial analogy fallacy, 53

This position entails the view that the future

does not exist*

Both Broad and King accept this result.

But as vjilliaras objects, arguments to prove the non-

entity of the futiire could equally

ii;ell

be used to prove

the non-existence of the past, the existence of v;hich

both King and Broad uphold, since the past is in the

category of that v;hich has become .

If this were a valid

objection, it would be fatal to Broad's and King's con-

ception of time as a series of events which have become, stretching into the past, and open at one end.

Clearly, time cannot be a series of events, if these

events do not exist.

There is the further difficulty,

analogous to tlmt of the moving bull*s-eye of the present, namely, the question: "How fast does the becoming

or change proceed?"

King answers this objection by

tliat

declaring that "it is meaningless to say

the Present,

70
the locus of the act of becoming, is either durative

or instantaneous, terms which have reforence only to

physical time and its divisions".

54

Although we disa

;ree v;ith

the theory

v^e

are now

discussing on essential points, we believe that Kinos

reply to the above question is essentially the correct

one, and that it represents an insii:;ht v/hich will lead to
the xmravelling of some of the mysteries associated vjith

time, space and change.

V/e

turn, therefore, to the

analysis of the relationships between these three.

To be^in,
i\'e

reaffirm a point which

xve

have

already emphasized, namely, thot if either space or time

or both are conceived as homo eneous, extensive con-

tinua, the individuals which appear in such homo:j;eneous

media must

,':et

their individuality from some source or

For a homo--eneous conIt is to be

principle outside the continua.

tinuum is literally, ev rywhere the same.

noted that when space and time are considered to be

extensive, they are also conceived to be homo :eneous,

infinitely divisible and continuous.

If we do not hold

that individuals are a matter to be decided by con-

vention, and we conceive of space and time in this manner,

we must admit that once the individuals are given to these

media respectively, the spatial and temporal i-^elutions of

71

them are forever fixed.

that the individuals

This does net, of course, mean

not "move", or that from the
v/e

coulc'.

point of viev/ ox our conceptions of notion,

should be

presented v;ith a static, "block universe".

Individuals

might move, i.e. chan^ie their extensive relationships,

either spatially or temporally, depending on the kind of

individuals that were
.n:iven.

If we conceive of space and

vjay in vjhich

time as extensive, the only rational

to regard

motion is to look upon it simply as the fact that an

individual
is_

at different places at different times.

As

Russell says:
It was only recently that it became possible to explain notion in detail in accordance TJith Zeno's platitude, and in opposition to the philosopher's paradox. V/e may nov; at last indulge the comfortable belief that a body in motion is just as truly where it is as a body at rest. Motion consists merely in the fact that bodies are sometimes in one place and sometimes in another, and that they are at intermediate places at intermediate times. 55

There is, however, a difficulty in this doctrine

of motion.

The difficulty is not merely the seeming un-

believablenoss of the doctrine, from the point of view of

our psychological experience of motion, but rather
the doctrine is epistemologically inadequate.
tte^t

For motion,

we suppose, always has some direction.

If we rer^ard
beinj^ at

motion as solely a matter of an individual

different places at different times,
xve

shall be hard

72

pressed to say In what direction the individual novos.

Direction has simply been loft out, except

v.rhere

it is

tacitly retained in the notion of an intrincic direction

of the time series, so that it is assumed that when an

individual is at different places at different times, the

times are in a linear series from earlier to later.
involves, as
vje

This

have seen, the view tliat there are

individual events the position of which is fixed in the

time-series v;hich has one open end constantly being projected.

We see, then, that theie is a mutual relation betvjeen the

existence of individuals and their extensive

Hov?

relations in space and time.

are these individuals

to be distinguished in the homogeneous extensive continua

of space and time?
The ansv/er must be that changes or

differentiations of something extended in space or time, must be given.

These individual changes or dirferences

of the extended in space or tine must be the distinguishable

individuals in these continua.

But if these were the only

kind of individuals, we would not be able to say that

anything moved , either in space or in time.

Movement con-

sisting, as we have seen, according to this way of re-

garding space and

tir.ie,

of the fact that something

is,

at

73

various places, all of these individual differences being

just fixed differences in the continuuin, all of them will
be where they are and i;ill not move,
vve

add here that we

are not speaking of the phenomenon of motion, but of

motion of an individual considered abstractly in the continuum of either space or time.

Such "motion^' will be

equivalent to the locus of a point as treated in mathematics.

The point does not move in the phenomenal

sense, but is at various points capable of bein^^ located

by means of distances from fixed coordinates.

Clearly, the differences in the extensive con-

tinua of space and tine could not be considered, by themselves, to be individuals in motion, althou<;:h they are

individual differences .

The re must be something in

addition to these, or unuei-lyin^ them, the conceptual

function of v/hich is to constitute the identity of an

individual in these di-'feronces, or, through changes.

The something which ansv;ers to these requirements is,
of course, the concept of substance.
\Je

hasten to add

that we wish to be undez-stood to mean no more here than

the principle which accounts for unity or continued

identity of an individual in various differentiations

of the continua.
For our purposes,
vje

wish to ignore

other possible connotations of the term "substance".

74

anu coruoicLor only

dividuation.

iot;

Tunction aa a principle of in-

Traditionally, the substance which has been thought

to have an intimate relation to time, rr^ther than to space
and the material world, is variously desir-nated as mind .
s:3irit
.

intellect , consciousness , self , the subject,

These terms, xvhatever else they may convey,

the soul .

all carry the connotation of something v;hich retains its

identity in different states in time, or, in

ouj?

sense,

moves through temporal differentiations v?ithout losing

its identity.
On the oth3r hand, the substance which is

thought to retain its identity in different parts of space,

or move through space, is matter , body or material sub-

stance.

Hence spiritual substance is the principle of

individuation of something which moves in time, and

material substance is the principle of individuation of

something v;hich moves in space.

And in both cases, it is

simply a matter of beinj;, in some sense, in various

differentiations in the extensive continiia of space and

time. This dualism of space and time, corresponding to

the dualism of matter and spirit, hov;ever, gives rise to

certain problems.

There is apt to be confusion in dis-

tinguishing the roles of the two substances, and stating

75

the relation, if any, betv/een them*

V/e

have here de-

liberately drawn the distinction as sharply as possible

in order to clarify the issues involved,
3ut it is clear

that the distinction is unsatisfactory from the point of

view of explaining human experience, v/hich, as we mentioned earlier seems to involve a spatio-temporal unity,
We may v;onder, for exaraple, v^hether two substances are

necessary, or v/hether one would be adequate to account

for all aspects of humaji experience. So far, we have

spoken as if the spatial differences and the temporal

differences yere entirely distinct.

This obviously needs

If we

to be modified if the dualism is to be removed.

suppose that
cessarj'-

vje

need only one substance, it will be ne-

not only to reject the redundant substance, but

also to brin^: those differentiations native to the rejected substance into relation
v>;ith

the accepted sub-

stance and its differentiations in its particular continuum.

This \;ill involve

tvjo

points:

(a)

we must lit-

erally identify the differentiations native to the rejected substance with either all or some of the differen-

tiations in the continuimi through xAich the accepted substance moves, and (b) we must try to show that laws or

principles

v.-hich

^^ovum the movenients or activities of

".overn

the accepted substance in its ovm continuiim also

76
tho differentiations in the othor continuum.
alon:: theoe lines,
xve

Thinking

will be led to choose one of two

possibilities
(A)

v.'hich

are as follovi's.

There is a principle of individuation

(substance) for space, and this accounts for the events

in the tine continuum.,
(B)
Ther-e is a

principle of individuation for time,

and this accounts for the objects in space.

Insofar as metaphysics takes space and time to be

extensive continua, these are, we take it, two basic

metaphysical positions.
v;ill do

In both, the accepted substance

txjo

double duty in the

continua, or

r.

A.hjr will

have one function for those differentiations v;hich are

common to both space and time.
But it is possible, of

course, that either of these positiions may be maintained

vjith respect to space and time

without an attempt being

made to carry through the requirements of a complete

metap>hysical system.

Possibility (A), therefore, may be

i-)art

of a

full-blown materialistic system like that of Hobbes, or

it may be presented simply as a theory concerning space

and

'cirne.

Ic is in latter guise that we meet with it

in classical physics, and in the views of 3road and others.

77
It allows one to mako a fundaraential diatinction betv;een

space and time, because the chanijes in the spatial re-

lations Ox objects

ax-e thou^jiht

to be events in time.

Provided it is thought that time is a series with one end

open and with events constantly being added, time v/ill be

different from space in that it will have an intrinsic

sense or direction.
It is to be noted that this view

involves a necessary reference to consciousness or the

self, in order to account for the direction in v;hich motion

proceeds in the physical world, as we have explained

above.

This reference can be expressed in two ways, either

by

sayin;-;

that time is a series with a projecting open

end, or by referrin^: to the observation of the direction

of notion by an observer.

Since the latter, (e:ccept for

special cases in the "specious present") alv;ays involves

the distinction of earlier and later, or past and present,
in the recognition of the dix^ection of motion, the two

amount to the same

thini;:.

Although the physical world

accounts for events in the time-series, the temporal

distinctions made by the self account for the direction

of the time-series, and hence for the direction of all
motion.

Considering objects in the physical world to

be simply individuals or substances, which are at

different places, neither the direction of the tine-series.

n
nor the direction oi any motion would be possible -without
this irreducible reference to the observer or the self.
In empirical science, of course, this reference is taken

for granted as generalizations concerning events in the

physical world are only possible ivhere empirical observations are possible.
It is clear that according to possibility A,

when reference is

j:iade

to the teiaporal distinctions of

the self, that spatial relations of objects may chanije

but temporal relations of events may not.

56

Thus, spatial

relations of individual objects may change, and these

changes are considered to be events which take their place
on the end of the time-series.

This involves sayin

that

the object or substance endures through changes in its

states or spatial relations.

Since the object retains

its identity through changes in its spatial relations,

and since these changes mark "events" v;hich are fixed

in the time-series, we say that an objects endures .

or is permanent, through time.

Here

v^'e

note the first

requirement of the theory being presented, that is, the

literal identification of changes or differentiations

in the spatial continuum with differentiations in the

time -continuum.

Thus differentiations in the time-

continuum are considered to be individuals, and these

79

get their identity from the changes in the spatial r3lations of objects.
But, as we pointed out earlier, vjhere

differentiations in an extensive continuum are considered themselves to be individuals, they cannot move ,

They are irrevocably fixed.

The kind of individual

which is thought to move through time, namely, the self,

has been rejected and the differential individuals must

remain in the position in which they occur in tha timecontinuum.

There is, however , the other requirement of

a complete metaphysic, viz., that all events in the tine-

continuum must be subject to the

lav;s

which goveni the

activities of the substance which moves through space.

This involves accounting for all those phenomena for

which the rejected substance might have been thought

to account.

Since the spiritual substance which moves

through time has been rejected, it is necessary to show

that all those different states through which the spiri-

tual substance is thought to retain its identity at

different times, are, in fact, subject to the laws which

goveni the activities of material substance.

This iS a

corollary of maintaining that all events in the time-

continuum are quite literally changes in the spatial

continuum. This necessitates an attempt to formulate a
57

materialistic psychology.

go

Metaphysical systems stemmin,: from possibilities

A and 3, both involve holdin,^ that thare is one nature
.

the differentiations of v;hich constitute dirforentiations

in the extensive continuum of both space and time.

There

is this di. forence between them, hov/evor, that ivhile the

materialistic system must deny that there are differentiations

which appear in the time-continuum alone, for example,

states of mind which are in tine, but are non-spatial,
the system which accepts the self as its sole substance

may maintain that there are differentiations which are

in time alone, without abandoning the idea that there
is one nature.

It v/ill only be necessary to maintain

that all spatial differentiations are in time, while time

may contain differentiations which are not in space.

This, as
\ie

shall see, is the position adopted by Kant.

The corollary of this latter position is that it will

be necessary to shovj that those differentiations which

are in the spatial continuum are subject to those

principles vjhich ^'ovem the activities of the self.

V/e

add that there is, of course, a wide latitude of

possible opinion as to what these activities peculiar

to the self are.
The self may be conceived as a set

of purely logical functions or it may be conceived as a

centre of feeling, and prehensions.

In addition, the

SI

self or subject v;hich fimctions as a principle of in-

dividuation may be considered monistically or pluralir^tic^.lly.

Nature may be the result of activities ox one

subject or a plurality of them.

For either theory, the same chan^ces of the ex-

tended are considered to be in both space and time#

If

this step were not taken, there woiild be no explaining

a unity of nature.

Accordinf^ to possibility (A) chanriies

of objects are considered to be events which are added to the end of the time-series.
exist.

Future time does not

According to possibility (B), there are events

which occur on the end of the time-series, and these

events
brin,,~

objects in space into existence.

Time is
In

the becoming or passing av/ay of objects in space.

both cases, the same, often empirically observable,

chant-es are dealt with.

Thus, the main difference be-

tween the two theories is a difference of emphasis, and

a difference of opinion as to what the substance or

principle of individuation relates to.

Accordin,- to

(a), substance relates to bodies, objects in space, and

the changes through which substance retains its identity

are thought to be events in time.

Accordingly, time is

merely a device for recording the changes of objects.

It is derivative , and since it cannot be perceived, is

82

sometimes thought to be unreal.

Although it is neceasary,

on this theory, to refer to tho temporal distinctions

of the observer in order to account for the direction of

motions, it need not be considered that anyohin.;; arises

or passes away so far as the representation of motion is

concerned.

Motion need be considered as nothin:^ more

than the order of the e:cistence of parts of an object.

As to how it comes about that an object is at different

places at different times, this theory need not be concerned,

Jut accordini:: to (B), the principle of in-

dividuation relates directly to change itself, to the

successive advance of time.
The self remains the same,
Ob acts are derivative.
.1

Accordin:- to (A), space

exists as a

vfhole,

and the principles which i^ovem the

movements of bodies may be thought to determine what

changes shall occur in the spatial continuum, and hence
also -what events shall appear in time.

According to

(B), space does not exist in itself, as a void aivaiting

changes in it, but space as extensive is only an abstract aspect of the primordial becoming of things. The

changes which bring objects into existence also bring ex-

tended space into being.

There is just so much space as

(A)

the changes of time determine that there shall be,

S3

considers that space and its objects are Ic.ically

prior to time and change.
(B)

considers change to be

lojically prior to space and its objects.

There is apt to be confusion as to the terra
"tine", v/hich has different meanings according to each

theory.

For (A), tinie is measured tirae, ultimately de-

rived iron the regularity of events, which are, in turn,

dependent on the movement of bodies in space.

time is
a

For (B),

successive series of acts of becoming itself.

Logically conceived, time is for both theories a continuum made up of points v/hich are events or changes.
Dut (A) disregards the fact that the continuum is made

up of changes and considers only its logical aspect as

a

continuum.

The result of this is that time is con-

ceived as a kind of extensiveness.

In conceiving of time

in this manner, theory (A) points in the direction of

the theory of the four-dimensional manifold in vchich

there is no necessary distinction

betv.'een

space and time.

It is thus as we have said a half-way house position,

which only manages to make a distinction between space

and time by supposing that the time-series is projecting

itself at one end in the forra ci

nev.

events.

Strictly

speaking, it is impossible to conceive how time, on this

view, could project itself.

For the instants of time are

S4

analogous to iJOints on a lino, and no natter

v;g

v.'hat

point

take, there is no next point, just as there is no

v.'hich

fraction

is next after 1.

But it is necessary that

the time-series should project itself, or it could have

no intrinsic direction.

If this intrinsic direction is

^iven up, there v/ill be no distinction betv;een space and

time, and theory (A) will liave developed into the theory
of the manifold, the difficulties of
v.'hich v;e

have already

examined

Theory (3), on the other hand, does not lose

slight Ox

the fact that time is conceived as a series of

It is possibly for this reason that time has
V'e

changes.

often been said to flovf.

date changes, not objects,

and time is conceived as an extensive continuxan of changes.

The thought may arise that the whole continuum is changing

from point to point.

Time floivs.

But this is a nistake,

chan.^^es,

because even though the continuum is a series of

it does not itself change.

The mistake is encoura-^ed,

^erhaps, by the fact that in terns of measured time

phenomenal changes take some tine to occur.

Nevertheless, the emphasis which theory (B)
,>laces on the fact of change is not

lightly to be cast

aside.

1/e

have seen that the temporal distinctions of

chan,r:;e.

the self, which are related to the apprehension of

65

are necessary in order to accoiuio for the direction of

notion.
Ox

Theory (A) points towards a complete rejection

temporal distinctions knovjn to the self, (with dis-

astrous results).

Theory

(D)

takes these temporal dis-

tinctions seriously and attempts to combine the fact

of
chani,:e v/ith

an extensive theory of tine.

For this

reason it considers chanj;e to be logically prior to objects and space, and its principles relating to the \mity

of nature are principles which apply directly to time

itself.
The principles are taken to be associated with

the self or subject, which retains its identity throiigh

chan,,i;e

of time, or through the process of becoming and

passing avjay ox objects.

Theory (3) attempts to take

advanta^^e of the clotie connection betv^een the self-

identity of the subject and chancre, so

tliat

activities

of the self are thought to determine the occurrence of

events and hence the coming into beins of objects.

But

it leads logically to a rejection of any concept of the

extensiveness of

tit/ie.

It leids to the fu.llest acceptance

of the subjective apprehension of becoming,

'.Je

have here intended to present in ^eneral the

outlines of these two possibilities before turning to a

specific examination of Kant*3 vievjs.
l/e

hope to show

that Kant* 3 theory of space and time approximates to

86

possibility

(D)

rathor than (A).

But before returning

to these two possibiliti'-js, we must try to clarify Kant's

Method of distinijuishinj between space and

distinction between inner and outer sense.
we
nfxi;

tiiae,

and his

To this task

proceed.

CHAPTEIL THREE

KANTS THiCORY OF 3PACS AND

TII-dE

1. Kant^s Distinction bobween Space and Time

The doctrine that concepts, (intellectual functions) can only yield knowledge when related to in-

tuitions

rr^iven

to sensibility may be

reri;arded as a

view

typical of Kants niature critical philosophy.

It is a

view v;hich Kant had not clearly arrived at in the

Inaugural Dissertation , indeed, since he does not seem

then to have separated so clearly and sharply the intuitive elements in knowledge from the intellectual.
Yet the trend towards the critical philosophy is qiiite
evident.
V/hile it is not the

purpose of this thesis to

trace the development of Kants views which culminated

in the position adopted in the Critique of Pure iieason .
a short excursus into the difference between his dis-

tinction betv;een space and time in that v;ork, and the

distinction put forth in the Inau':ural Dissertation

may serve to throvj the doctrines of the Critique into

sharper relief.
This thesis takes these doctrines as a

point of departure in showing the systematic relation

between Kant's and 3ergsons theories of space and time.

One aspect in particular of the difference between Kants
viev;s in these two vjorks is important.

This concerns

S7

the relation of the intellect to tine.

It was said above that Kant did not seoarate

purely intellectual functions from intuitive functions

in the Inaur^ural Dissertation as sharply as he did in the

Critique ,

Similarly, the position adopted in the former

work on the question of the intellect's relation to time

was expressly repudiated in the latter v;ork.

In the Critique the hi:;hest principle of all

analytic judgments is held to be the principle of contradiction, a purely negative criterion of truth, ne-

cessary but not sufficient to guarantee the correctness of all judgments.

The principle of contradiction must therefore be recognized as being; the imiversal and completely sufficient principle of all analytic knov.'led,:;e; -but beyond the sphere of analytic knox^^led.^e, it has, as ^ sufficient criterion of truth, no authority and no field of application. 5^

The principle of contradiction, as characteristic

of the logical functions of the intellect is not, indeed,

ignored or omitted in the Inau^:ural Dissertation .

There,

instead of making the critical distinction betv;een analytic and synthetic

.judp^roent
.

the use of the intellect

is likewise two-fold, but is on the one hand real, in

xvhich "the very concepts of objects or relations are

raven "

'
.

59

and on the othar hand, lo :ical . in ivhich

b^9

"concepts, whencesoever jiven, are only subordinated to

one another, the
lov.'er

to the hi:^her (the comraon marks),

and compared with one another according to the principle

of contradiction".

In both works, then, the principle

of contradiction is held to be characteristic of one use of the intellect, and in each case, this is its purely
lov;ical use.

But the relation of the principle of con-

tradiction to time is si-:^niiicantly different in each

case.

In the Inau ;ural Dissertation time is held to be

intimately associated with the operations of the intellect in the use of the principle of contradiction.

Further, though time does not indeed prescribe laws to reason, it yet establishes the chief conditions by the help of which the mind am order ibs notions according; to the laws of reason. Thus I cannot decide whether a thin^ is impossible, except by predicating A and not-A of the same subject at the same time . 61
That Kant meant
x-zhat

he said can be seen from

vj'here

his repetition of the same point in the very place

he is at pains to expose the errors due to the con-

tamination of intellectual knov/ledge by the sensitive in

the metaphysical fallacy of subreption.

Here sensitive

or phenomenal characteristics are attached to purely

intellectual concepts in such a manner that phenomena

are intellectualized, that is, as Kant was later accusto-

90

med to put it, treated as things in themselves.

That is to say, our intellect x^ecognises impossibility only where it can remark the simultaneous enunciation or o;.';>03ites about the same subject, i.e. only v-'here a contradiction occurs. The human intellect, tho^-efox-e, Cin make no jud^ent of impossibility in cases in which this contradiction is not 62 is not found.
In the Critique , on the other hand, the doctrine

that time "establishes the chief conditions by the help

of which the mind can order its notions according to the

-63
lav/s
lav;

of reason",

is expressly repudiated.

Thus of the

of contradiction Kant says that:

Althoui^h this faiaouB principle is thus without content and merely formal, it has sometimes been carelessly formulated in a manner which involves the quite unnecessary admixture of a synthetic element. The foraula runs: It is impossible that something should a_t one and the same time both be and not be... -the proposition is modified' hj the condition of time... The principle of contradiction, however, as a merely logical principle, must not in any way iii/dt its assertions to time-relations. The above fonnula is therefore completely contrary to the intention of the principle. o4

The loc;ical principle of contradiction needs no

help from time, and this being the case, it is apparent

that time bears a different relation to the intellect in
the Critique than is the case in the Inau":ural Dissertation In the former work the intellectual functions of the self
,

or subject, from which the principles xvhich bring about

the iinity of nature must ultimately derive, are freed fix>m

any confusion with either of the two extensive continua

91

of space and ;;ime.

Kant has become clearly aivare that

the principles which produce this unity are separate from

the msdia to v/hich they apply.

Nevertheless, in both

the Critique and the Inau":ural Dissertation there is a

clear indication of the importance of

tiiue

in contra-

distinction to space as regards this application.

Kant

expresses this in the Inaa":ural Dissertation by saying

that "time approaches more nearly (than space) to a

universal, x-atioiial conce-it, in that it embraces absol65

utely everythin:; within its survey".

And in the

Critique Kant explains why the categories may be sche-

matised by means of time-determinations, as follov;s:

Now a transcendental determination of time is so far hono;;eneous with the cateivory, which con stitutes its unity, in that it is- universal and rests upon an a priori rule. But, on the oth3r hand, it is 30 far homo, ;eneous with a;:pearance, in that time is contained in evsry empirical representation of the manifold. 66
Both in the Inaur.ural Dissertation and the
Critique of Pure Reason Kant distin.Tuishes between space

and time as forms of sensibility, and as pure intuitions.

The distinction betv.'een them, then, is a distinction be-

tween two different a priori formal conditions of sen-

sibility or two different

..ure

intuitions.

This is stated

in the following quotations.

92

Time is not somethinr- ob.iectivo mia real , It is neither substance nor accident nor relation, but is a subjective condition, necessary ov/in;; to the nature of the huraan mind, of the co-ordinating of all sensibles according to a fixed law; and it is a pure intuition, 67

Kant describes space in almost identical language.

After

statin,'- tliat

"the conceot of space is thus a pure

68

intuition",
69

and also "the fundamental form of all outer

sensations",

Kant

f-oes on

to repeat what he

lias

said

about time in almost identical terms.

Space is not somet'iin objective and real , neither substance, nor accident, nor relation, out subjective and ideal and, as it v.'ere, a schema, issuin by a constant lav; from the nature of the mind, for the connecting of all outer sensa whatsoever, 70
:

AlthoUi':h there is a ceriiain

chan,:;:e

in the dis-

tinction between space and time in the Critique of Pure

Reason , from that of the Inauf;ural Dissertation , they
f.re

still distinguished in the same manner, namely, as forms

of sensibility.

The

chan,':;e

is due to Kant^s sharper

realization of the difference between intuition and

thought.
Hence in the Dissertation Kant could say that;

So far is it from bein^ possible that anyone should ever deduce and explain the concept of tine by the help of reason, that the very principle of contradiction presui'poses it, involvin^j it as a condition. For A and not-A are not incompatible tinless they are jud,~ed of the same thin^^ tor:;ethQr (i.e., in the same time) 71
On the oth
.r

hand, in the Critique Kant

lias

completely

93

severed any necessary and intimate connection between the

peculiar functions of conceptual thinkin

and intuition;

althou:^h, of course, he stresses their union in empir-

ical knowledge.

Indeed, it is only because Kant has been

at rreat pains to isolate and examine the peculiar nature

of each, and make clear what the functions of each consist

in, that he can present a clear account of that union.

Thus, in the introductory remarks to the Transcendental

Aesthetic, Kant announces that "In the course of this

investigation it will be found that there are

tv/o

pure

forms of sensible intuition, servin;- as princijles of

72 and continues a priori knowledj^e, namely, space and time",

on to present his arguments for this contention, vjithout me nt ion in:; any necessary relation of one of these pure

intuitions, namely, time, to the operations of the intellect,

^'Je

emphasize this point because we wish to

stress thc.t the singular role of time in the philoso hy of Kant does not consist in time enterin.- into the

operations of the intellect as a condition of them. The

principle of individuation, and its related principles,

must in the nature of things be considered as distinct

from the medium to
v.'hich

it applies.

The differences between space and time in Kant's

philosophy, however, are not limited to their

beinj.:

dif-

94

forent forms of oensioility.

ferences for
Ki.nt

The sirail^i^ities and difv;e

correspond, in fact, to those v;hich

pointed out above in coimectlon with the views of C,D.

3road,
r>..ucj

unv.

^ii.ie

uro cooa cuiiuinua,

ox*

73
v;ords, "Space and time are quanta continua ".

ivriiu's

and there-

fore extensive,

Space and

tiaie

are "the two original but there is an intrinsic

"Time

74

quanta of all

oior-

intuition",

difference between them in their extensiveness.

of all series",

is in itself a series, and indeed the formal condition 75 but space, on the other hand, "is an

76
not a series".
By this Kant seems to mean

that space or a spatial series has no intrinsic direction, for he says that", ...in space, taken in and by
itself, there is no distinction between progress and re77 gress". There is, hov;ever, a pro-ress in the time-series

from past to present.

The present moment can be regarded only asconditioned by past time, never as conditioning it, because this moment cones into existence only throush past time, or rather through the passing of the preceding; time, 7^

Kant clearly regards time as a series which has an

intrinsic sense or direction from past to future.

Further-

more, he re^iiards this series as deriving its intrinsic

direction from tho fact that it has one end open

to which

95

new events are conotantly

bein;.:

added,

Althou:;jh he

puts

this point in temio of conditioning and the conditioned,

his meanin^-, is clear.

Thus v/e necessarily think time as having completely elapsed up to the .^iven moment, and as being itself siven in this completed form. This holds true, even though such completely elapsed time is not determinable by us. But since the future is not the condition of our attaining- to the present, it is a mattjx' of entire indiff^i-ence, in our comprehension of the latter, how we may think of future time, whether as comin,_, to an end or as flowing on -to infinity, Vie have J as it were, the series m, n, o, in which n is ^^iven ug concliticnoU by m, ana" at" the same tTme as bein:-; the condition of oT The series ascends from the from the conditioned n to si (1, k, _i, etc.), and also descends from th^ oonH'ition n t'o the con"" ditioned o (, , r, etc). 79
Time, for Kant, flows, as it v/ere, "down hill",

from tho past to the present; and, by means ox

on in GO the future.

nev;

events,

Space and time, however, in and by themselves are

not concepts,

Kant is emphatic on this point.

They must

therefore be unique, having nothing in common with each

other or with anything else.

For if they had anything in

common v;ith anything else, the common elements would

provide a basis for a discursive concept.

But it is

extremely difficult to try to reconcile Kant's doctrine

that space and time are intuitions and hence singular,

with his numorous statements about their analogous

properties.
To a certain extent the difficulty can be

96
resolved.

For example, the irnity of the horaor^eneous

manifold in either space or time is the result of

synthesis, but, as Paton points out, the explanation
of this cannot be siven in the Transcendental Aesthetic,

where Kant is primarily concerned with sensibility in

isol tion.

urthennore, until Kant has explained his aocurine of synthesis, he has to speak as if the uJTiity of space and time were . iven in intuition. It can, hovJever, be ^iven only because of a synthesis which does not belon-:; to sense. The necessary synthetic unity of space (and of timej depends upon, and presu 'ooses, the pure catec'ories of the understanding. All this is onitted from the Aesthetic, but it seenis to me tliat such an omission is defensible, SO A nan cannot explain his whole philosophy at once.
,

Insofar as the analo^^ies stem from this unity,

there is no necessary inconsistency with Kant's view that
space and time are pure intuitions.
But insofar as space

and time are of such a nature that an analo/^ous unity,

and common concepts, must be formed of them, they cannot

be considered imique.

As intuitions they would then have

It is clear that if the categories

marked similarities.

of quantity are applicable to both space and time, they

ipso facto have that much in common.
And it is ab\m-

dantly clear that Kant did think that space and time
were capable of bein:^ represented in the same terras.
Besides those definite statements to which we have already

97
draivn

attention Kant

'lays that "A

concept of space and tine,

as quanta, can be exhibited a priori in intuition, that is,

constructed, either in respect of the quality (figure)

of the quanta, or throui];h number in their quantity only gl (the mere synthesis of the homogeneous manifold)".
No'.v

it is quite possible that there rir:ht be

tvjo

pure intuitions, which unique in themselves, yet have

somethin.3 in common.

This is apparently Kant's view.

Nevertheless, such a doctrine, if it is to avoid the con-

tradiction in the kind of eclecticism which

above in Choptei*Tx-^o,

^^'e

mentioned

must maintain

tl-iat

the essential of

either space or time is omitted when they are represented

in similar conceits.

Otherwise the essential of both

space and time will be adequately represented in the same

concepts and a distinction between them as intuitions

will be unnecessai'y.

One intuition, at most, v/ould be

enough, for they would represent a common type of intui-

tion oven if there were a plurality of them

v.'hich

dif-

fered in unessential ways.

There would be no more point

in distinguishing them than there would be in distinguishing a multiplicity of separate intuitions of space, and supposin-'-

them to ba different kinds of intuition,

Nevertheless, Kant does distinguish between space

and tine, not only in that tine is all-pervasive in the

sphere

oi'

sense, as we have seen, but also in

oliat

every-

thinji that is in space is -*lso in time, vjhile some states

Oi the self are in time alone,

Ve shall explore the

implications of this latter point in subsequent sections.

Meanwhile it can be noted that these two points involve

one another.

For ix

tii.ie

alone is allfcervasive, in con-

tradistinction to space, it follows that something must

be in time aloiie.
sta.;.es

..hen

it is held, as Kant does, that

peculiar to the self such as thoughts and feelin<:s

tiiae

are in

alone, this is a clear indication tiiat Kant

has rejected tliat theory vdilch we labelled above as

possibility a,

i'hia

theory is coiiraitted to attemptin^^ a

materialistic psycholo^^y, v;hile Kant, as we shall see,

rejects the possibility of any sort of science of

psychology entirely,
Kant's position is not, in ibself, radically
different from that of Broad, for both of them, v/hile

conscious of the fundamental distinction betvjoen space

and time, attempt to express this distinction by means

of spatial analo^ries.

Time is distinguished from space

time is a series, which has an in-

conceptually in

tliat

trinsic direction.

V/e

have already expressed doubts as

to v/hether the difference between space and time is ad-

equately expressed in these terms,

Kc-mt's denial that

99
jipace and time are concepts lends force to this kind of

criticism.

This is particularly true in viev; of the fact

that this conceptual distinction still involves a tacit

but essential refej-ence of the self.
If the self did not
coming;';

have a sense of the becomin: of things, the

into

existence of new events, it v/ould, as

v^e

hive seen, be

impossible to find any difference betv^een a temporal

series Lnd a spatial series.
The points which
\:e

hc.ve

already mentioned concomins Kants distinction

betv.'een

space and tine preclude his acceptance of any such view

as the one
vje

have designated as possibility A.

The

bare recognition of non-spatial mental states is enough to ensure this.

But further considerations associated

v;ith his Copernican revolution led Kant to adopt the theory

we have called possibility B,

Before discussing this,

hovjever, it is necessary to attempt a clarification of

v.'hat

Kant meant by inner and outer sense.

chapt;i;r

three
tifie

kant's theory of space and

2 Inner and Outer Sense

Kant*s doctrine of innor and owtar sense is difficult, and while we shall present what
vje

hope is a

consistent and intelli-^ible interpretation thereof, it

cannot be do;^atically asserted that this presentation
is what Kant had in mind. It is to be feared that

Juntas terminology in regard to this doctrine is not

entirely consistent, and although this does not entitle

us to conclude ipso facto that the doctrine is thus ne-

cessarily inconsistent v^ith itself, it does make it hard

to know whether one is understanding Kant rightly.

Nevertheless, the interpretation to be here expounded is

based on documentary evidence, and makes it possible to
view in a
nev; li:;ht

some puzzling features of the

critical philosophy which seem inexplicable according

to alternative treatments, for example, the account of

the schematism of the categories,

Kant's distinction between inner and outer sense

is based, at bottom, on his belief that not all the matter

of experience is ^iven in spatial relations, and this

insi-'^ht

he expressed as early as the Inaur:ural Dissertation ,

100

101
sayirii:

of time, th;rt it "embraces absolutely everything

\vithin its survey, namely,

space itself, and in addition

the accidents which are not comprehended in space-

relations, such as the thou/^hts of the soul".

This in-

troduces another aspect of the distinction which Kant also anticipates in the Inaugural Dissertation , namely, that inner sense is bound up v;ith the staces or re-

presentations of the soul, and that outer sense is associated with spatially extended objects.
or these concepts, the one (i.e. space) properly concerns the intuition of an. object, the other {i.e. time) a state , namely, that of representation. S3

Kant is not very explicit concerning what the

term "state" comprises in this context.

hovt'ever,

In one place,

he seems to indicate that he does not mean

no\i

what would

be termed affective states, i.e. feelin^-^s.

it is especially relevant to observe that everything in our knowledge v;hich belon;:s to intuition feelin^ of pleasure and pain, and the will, not beins knowledge, are excluded contains nothing but mere relations, &k

...

Kant is here referring to knovjled

,e .

hox^evar,

and, as we shall see, the appearances of the inner states

of the soul are not objects of knowledge.

In other

places Kant definitely includes thourhts and feelings

as

inner states.

But the cataloguing of all theit inner

102

sense co/aprises
basis
lOi-

raay

not be necessary as long as the

the distinction between inner and outer sense

can be made clef;r.

Unfortunately, Kant presents the

distinction in a

VGi?y

brief form.

Inner and outer sense

are described as means whereby we become conscious of

inner states and outer objects, but Kant does not elaborate on the nature of these means.
3y means of outer sense, a property of oiir mind, we represent to ourselves objects as outside us, and ail-without exception in apace. In space their shape, magnitude, and relation to one another ar^e determined or determinable. Inner sense, by means of ivhich the mind intuits itself or its inner state, yields indeed no intuition of the soul itself as an object; but thei-e is nevertheless a determinate form (namely, time) in which alone the intuition of inner states is possible, and everything which belongs to inner detenninations is therefore represented in relations of time. S6

This

passa._;e,

hoivever, yields us one point of

importance to which we can hold fast, namely, that vjhatever else Kant may mean by inner and outer sense, the

former contains time-relations and the latter spacerelations.

3ut it should be added that Kant makes no
txvo.

mention of a mutual exclusion of the

Although this
raay be,

would seem to be a natiiral inference, it many of Kant's

tc^rms,

as with

dangerous to jump to conclusions.

As it turns out, thei-e is ample evidence that they are not

mutually exclusive.

To be more explicit, it is not the

case that what is intuited in inner sense in time-

103

relations excludes whyt is intuited in outar sense in

space-relations.
But since all i*epresentations, whether they have for their objects outer things or not, belon,']; in themselves, as detornilnations oT the mind, to our inner staoe; and since this inner state stands under the formal condition of innar intuition, cand so belongs to time, time is an a priori condition of all a-ipearance whatsoever.

Kant makes the same point in terms of empirical

knowledge
All increase in e.^pirical knowledge, and every advance of perception, no matter v.'hat the objects may be whether appearances or pure intuitions, is nothing' but an extension of the determination of inner sense, that is, an advance in time. This advance in time determines everything and is not it&S self determined through anything further,
Thus, as Kant puts it, "There is only one

whole in which all our representations are contained,

namely, inner sense and its a priori form, time".

Inner

sense with its form, time, seems, as did time in the

Inaug^ural Dissertation , to be "absolutely primary in the

domain of sense".
There is, then, some reason to believe that v;hat

Kant means by inner sense is consciousness, in .-eneral,

of anything that may be an object for us. This empirical

consciousness, of course, must be distinguished from the

transcendental unity of a;>perception,

throu,;5;h

which alone

104

any empirical consciousness of the 3elf is rosrible.

Apperception and its synthetic irnity is, indeed, vary far from being identical vi/ith inner sense. 90
Inner sense, in this meaning of the panorama of

changing conscious states, may be entitled, says Kant,

en;

irical a^joerce'^tion .

Consciousness of self according to the determinations of our state in inn r perception is merely empirical, and alvays chan<;;ing, Ko fixed and abiding self can present itself in this flux of innor appearances; Such consciousness is usuallynamed inner sense, or empirical apperception , 91
It is to be noted that Kant says it is no fixed

and abidin

self that can present itself, but not ne-

cessarily no fixed and abiding appeax^ance .

This in-

terpretation of inner sense requires, hoviever, thst represent rtions of outer sense be included
x-jithin

inner

sense, and this seems to be flatly contradicted by some

of Kants statements, for example, "Time cannot be out-

wardly intuited, any more

tlian

space can be intuited as

92

somethini' in us",

and:

Time is nothing but the form of inner sense, that is, of the intuition of ourselves and our irmer state. It cannot be a determination of outer appearances; it Irias to do neither X'\?ith shape nor position, but vdth the relation of representations in our inner state. 93
It should be noted, however, that Kant does not

here specifically say that outer sense cannot be in-

105

eluded xvithin inner sense, but only that time cannot be

outwardly intuited, and cannot be a determination of outer

appeax'ances.

In short, he is not speaking of outer and

inner sense at all per se, but of the nanner in which we

can intuit time.
It is only by means of an inference

that

ive

can construe these statements to mean that outer

sense, v;ith its spatial relations, cannot be inclu-

ded in innor sense v;ith its temporal relations.

It is,

of course, perfectly obvious that Kant needs temporal re-

lations to obtain in the sphere of spo^ieULy extended

objects,

ether v/ise, he could not possibly ansv/er the 94 queotion "How is pure science of nature possible?",

which is one of his avowed purposes.

If we construe inner sense to mean empirical

apperception,

v;e

may still maintain that time cannot be

in the sense that time cannot be in-

outv;ar-dly intuited,

tuited in that which is purely spatial.

Likewise, time

woxild not then be a determination of outer appearances.

Outer sense, by this token, would be a term standing

for our ability to a prehend objects in spatial re-

lations, and only in spatial relations.

U'liat

is ap-

pi'ehended in outer sense would be an abstraction from

inner sense.
distin^'-.uishin.-

This is, perhaps, only a

10i,;ic.l

result of

between space and time as two different

106
pure intuitions.

If space and tine are separate and

follov.'s

distinct ways of intuitin:^, it

that determinations

peculiar to the one will not be peculiar to the other,

Kant*s distinction between inner and outer sense
thus seems to amount to this.

When we are perceiving

anything, we are also empirically conscious that we

are perceivin
;

it,

'.'hatever is ,i;iven is alv.ays -jiven

to us with our peciilif.r nodes of sensibility.

In the

totality of all that we experience, however, some things

are ^iven,
v.-hich,

considered in abstraction by them-

selves, are purely in spatial relations; other things are jjiven vjhich are purely in temporal relations, namely,

certain

ir,3ntal

states, states which

w^e

clearly and
But nothing is given

intuitively recO':^nize as non-spatial.

to us which is neither in space or tirue.

Kor is any-

thin

ever .iven which is not in time, but only in space.

Hence "inner sense" includes
tTA'o

meanings, or

to put it otherwise, covers

tv;o az-eas

of representations,
On the

which are not necessarily mutually exclusive.

one hand, inner sense may mean vjhatevor pertains to my

non-spatial mental states.

innor sense.

This is a narrow meanins of

On the other hand, inner sense may mean all

my 3:cp^rience of whabev5r kind.

This is a

v.;ide

meaning.

It is not certain th^at Kant does not alternate these

meaninjs.

It is fairly certain tliat he uses "inner sense"

lu7
in the \vide meaning; often enough to mal:e that meaning

of it clear, as can be seen from the rore:^oinf, quotations,

as well as the follov;in.3:
ViiTiiatever the origin of our representations, whether they are due to the influence of outer things, or are r^rocluced through innar causes, whether they arice a priori or bein^;; appearances have an empirical ori;-in7 they nunt all as modifications of the mind, belon^; to inner sense. All our knowledge is thus finally subject to time, the formal condition of inn3r scnoe. 95

appearances are not thin^js in them . selves, but are the mere play of our representations, and in the end reduce to determinations of inner sense. 96
Yet Kant also often uses the term "inner sense"
to mean whatever pertains to non-spatial "states", and

distinguishes these from the contents of outer sense:

In outej;- sense we find no other outer effects save chanfres of place, and no forces except mere tendencies Vv'hich issue in spatial relations as their effects. I/ithin us, on the other hand, the effects ax-e thou.jhts, amonj vjhich is not to be found any relation of ^lace, motion, sha-e, or other s-oatial determination 97

...

The distinction between inner sense is

v;nat

has

been here called the narrovj meaning, and outer sense,

thus seems to rest solely on the distinction between

space and time themselves.

Inner sense contains that

which is only in time-relations, and outer sense that

vjhich which
ars

well as being in time is also in space.

log
The trojiscendental object is equally \mknown in respect to inner and to outer intuition. But it is not of this that we are here speaking, but of the efiipirical object, which is on lied an external object if it is represented in sr ce , and an inner object if it is represented only in its time - ref; tion st Neither space nor time, Kowever, is to oe found save in us, 9^

In tryin

to explain more fully the difference

betv;Gen representations

which are in inner sense and

those which are in outer sense, a distinction which seems

to require an intuitive grasp of the difference betv;een
space and tiiao themselves, Kant allov;s himself some rather

picturesque language:
I'lattar, therefore, does not mean a kind of substance quite distinct and hetero.^eneous from the object of inner sense (the soul), but only the distinctive nature of those appearances of objects in themselves unknown to usthe representations of which W8 call outer as compared with those we count as belongin,; to inner sense, although, like all other thoughts these outar representations belong- only to the thinking subject,- They have, indeed, this deceptive property that, representing objects in space, they detach themselves as it v/ere from the soul, and appear to hover outside it, 99
'

If we begin with what might be called "total experience", inner sense in the narrow meaning of nonspatial thoui^hts flittin3 through the soul, is an ab-

straction from total experience.

Similarly, outer

sense, conceived as appearances v;hich are only spatially

extended is also an abstraction from

tot-^;l

experience.

In this meanin:, of "inner" and "outer" it would be cjuite

109
true bo say that "Time cannot be outv/ardly intuited any 100 more than ap-.ice can be intuited as soniethin;; in us",
but these abstractj.ons \;ill have relevance only for the

pure intuition of splice and time, that is, they will be

useful in explainin;: the two basic forms of sensibility. They

v.'ill

not be useful in explaining the universal

sensible characteristics of all experience whatsoever,

- possible expirienco,

?or that, a wide laeanino of

In that

"inner" and "outer" sense will be required.

context, inn^r sense v/ill turn out to be coextensive

with all experience whatsoever; and outer sense

v;ill

en-

compass representations v/hich, as part of inner sense,

are ipso Tacto in time-relations, and distinctively

characterised by the fact that they are also in space.

It is apparent that Kant is thinking in his doc-

trine 01 innor sense, of the paramount importance of the

train of non-spatial representations in our apprehension

of time, a fact which has been reco/jnized by some '..riters
on psycholo:jy.

For exaiiiple, in Stout's Iianual of

as the author points out, from the
v;e

Fsycholony .
"^;^enetic

x.'ritten,

point of view",

find the follov;in2 state-

ment:

The apprehension of temporal relations, as they exist for hroman consciousness, is an extremely

110
The part coraplex pro-Iuct of montal devslormGnt, played in it by trains of free ideas is of pre101 dominant importance.

Similarly, Kant seems to be drawing attention to

trains of ideas when he says that "everything that is in 102 Kant holds these innei- sense, is in constant flux".

changes in inner states to be so continuous that "No

fixed and
c.bidin.
.

self can present itself in this flux

103

of inner appearances".

But the train of ideas through

v/hich we !\ve conscious of both our ohenonenal selves and

of temporal series cannot, be separated except ab-

stractly, from those appearances which we entitle outer,

a point which Kant himself makes in
sayin;;',

that "the

reoreGant:.uions of tba outer senses constitute the 104 proper material viith which we occupy our mind".

According to Stout, the trains of ideas, the sort of

thing which Kant would say comprises inner sense in the

narrow meaninjr of the

terra,

is absolutely indispensable

to the apprehension of time.

Without such trains there could be no such thing as the definite apprehension of a time-series, havinr: a distinguishable beginning and end, connected by a train of intermediate events, each havinp its own position determined by its relation to other events which have come before and after it. 105

Perhaps the difficulty we have in understanding

Kants theory of inner sense is due to our not taking

Ill

soriously enough this fundamental psycholo :ical T ct,

which Kant ooems to have hit upon.

If it is indeed a

fact that v;ithout the train of ideas in the constant flux

of inner sense (in the narrow

nieanin;^^^;)

vje

could have no

a^^prehonsion of time, it is not surpi'isins that Kant

should havo extended the meaning of inner sense to include all time-relations v/hatever
.

If, as Stout points

X'jould

out, i.dbhout tr:.in3 of ideas, there

be no ap-

prehension of time-series then wliatever time-series we do

in fact apprehend, for exanple, a tiiae-series in spatially

expended objects, v/ould ipso facto in some way involve

inner sense.

Thus, it would be only because

v;e

did not did not,

take this psycholo^^ical fact seriously

enou,^:h,

that is, see the generality of it, that we xfould suppose

that time-series of outer appearances did not involve

inner sense, and

drax-;

the incorrect conclusion that a

theory of time based on inner sense could take no account

of such outer "oiae-series.
V/hether or not Stout *s

contention is the whole truth about our apprehension

of time is not to the present purpose to discuss.

Nevertheless, sequential experience whether of ideas or

of other psycholor:ical states seems to be a sine qua

non of any theory of the apprehension of time,

Bartlett,

who suiigests that "V3ry likely it is

tz-ue

that the makins

112

of tirae-dintinctions is, rsycholo iccilly, fundamentally

lub
a socic^lly cletormined resronse",

also says that one of

the essentials of a psycholo ical theory of the ap-

prehension oi tino is "the formation of or';:anized chains

of successive activities throLigh the operation of some 107 Successiveness biolo'iiical or psycholo:ical interest".
of pnycholo :ical stages, then, is intimately associated
v;ith the

apprehension of

tirse,
1-vas

and this is, at bottom,

It is here

the point to vjhich Kant

drax-m attention.

that Kant and the psychologist part ways,

For Kant the

fact of the succession of psycholo deal states is only

explicable in tsrrns of a transcendental analysis of himan

kno'vledge, part of which necessarily deals
v.'ith

sensi-

bility.

For the psycholor^ist interest terminates in

the det irnination of the fact and its place in the theory loa of empirical psychology.

Kants particular interest in drav/in

tinction between
third
meanin;-:;,
inn^^i-

a dis-

and outer sense necessitates a

not incompatible with, but differant from

the others, vjhich he jives to these tanas.

raeanin,

This third

is apparently obtained by abstracting pure re-

lations from inner and outer sense in the \vider meaning.

Here he is not speaking of inner and outer sonse as the

whole of appearances and the spatial appearances within

113

that whole, v.hich contains appearances exejiplifyin,.: both

matter (of sonoation) and form; he is

speakin,^:

of inner

and outer sense as complexes of pure relations.

Inner sense in the meaning of the totality of

iOwSible experience contains (a) non-spatial representations in time-series, and {b) spatial representations

sIgo in time-series*

Here, outer sense is simply that

But in this

area of possible experience comprising [o).

concrete signification inner and outer sense involve

the matter of sensation as well as the form.

If ab-

straction is made from the rrattar, however, inner sense

v.'ill

be the net-work of time-relations throughout posv.'ill

sible experience, and outer sense

be the net-work

of spatial relations throughout possible experience,

iiecauce inn or sense in the more concrete meaning of the

"sum of all representations" contains an area of representations which ate both in spatial and temporal relations, by the same token, inner sense in this
riore

abstract meaning v/ill contain temporal relations which

relate representations
\>jhich

are also in s::atial relations,

Thus, determination of time throu:;:h the transcendental

unity of apperception will affect outer appearances,

Kant speaks of the "ideality of both outer and
109

inner sense",

just as he asserts the empirical reality

114
of 3pace and time and thoir transcendental ideality.
conrin.i;.tion of the fonnor assertion, and hence of the

In

lattor, he takes oains to make clear tliat inner and otiter

sense contain nothin;j but relations.
. outer sense .jives us nothin,_; but nore . relations. . This also holds true of inner sense because uhe tine ?.a .'lich ue set theoe re. presentations . , itoelf contains only relations ... Ill
,

Here, finally, we arrive at the specific doctirino

of inner and outer sense which Kant needs to acPux'e

complish the purposes of the Critique of

"reason .

Inner sense as transcendent ally ideal is nothin^ but the

netv;orl: of

time -relations throughout possible experience.

Outar sense is the netv;ork of spatial relations throughovA possible experionce, and the objects thus related in

space are also related in time in innei* sensei (as tran-

scendentally ideal), because, "Just as I can say a priori

that all o-utcr appearances are in space

...

ci,ui

also

say, ron the i.rinciples of inner sense, that all

appearances

x;hat soever ,

that is, all objects of the

lations".

senses, are in tine, and necessarily stand in tine-re112 Inner and outer sense, then, like time and
space, are om-irically real, but transcendent ally ideal,

or as Kant puta it, they

ai-e

nothing in thenselves.

The

distinction between

inn-Ji-

and outer sense rests on

the

115
Tundaiaental distinction
b'^f.-.'sen

space and tine.

As em-

,drical realitisG inn

t.ar sense

becdno the msans

of distincuishin^ betvisen external^ spatially extended

objects, and int.-imal^ temporally related thou;;ihts.

Thus external things exist as well as I myself, and both, indeed, upon the immediate witness of my S8lf-con3ciounn8;os, The only difference is that the representation of myself, as the thinking subject, jolon:G to i;in-.- sen.^e only, ivhilo tha representations which raark extended bein^^s belong also to outer sense. 113
As the above quotation indicates, inner sense, on the empirical level, does not exclude outer sense and
its appearances, but must necessarily include these

"upon the immediate witness of my self-consciousness".

In referrinj to outer and inner sense, then, the emphasis

should

b'3

on j^^nse^, as for Kant, these terms seem to be

names for the capacity to sense spatial relations and

the capacitj'- to sense temporal relations^ respectively,
VJe

m.u3t vi

r.;

all our perceptions, "whether

v;e

call them

inner or outar, as a consciousness onlv of v;hat is 114 Outer objects are redependent on our sensibility".
present;: tions, "-vvhich are entitled outjr because they

depend on vjhat we call

115
is space ";
'

outer sense*,

^-ihose

intuition
116

a representation is an "inner object if it

is represented only in its time -relations".

Thus, in

the final analysis, inner and outer sense must be

115
fandaiaental distinction b'jtv/ean apace and time,

A3 em-

^drical realitias inn^.r and outar sense becoao the means

of di3tin::ui3hin^ boUr'jon external^ spatially extended

objects, and intomal^ tenporally related thoughts.

Thus external things exist as well as I nyself, and both, indeed, upon the immediate witness 0." my 3elf-con3ciou:-,n9oS, The only dirxGi^?nce is that the representation of myself, as the thinking subject, jalon;G to inri-^ 3en.-e oaly, '.'hlL^ t'la representations which roark extended beinrs belon:^ als o to out or sanse. 113
-

.-

As the above quotation indicates, inner sense,

on the empirical level, does not exclude outer sense and

its app3arances, but must necessarily include these

"upon the immediate ivitness of my 3eir-con3ciou3ness".

In refarrln^ to
c

inner sense, then, the emphasis

should bo on Gcnse . as for Kant, these terms seem to be

names for the capacity to sense spatial rel^.tions and

the capacitj'- to sense temporal relations, respectively,
VJe

must vi

n;

all our perceptions, "whether

v;e

call them

inner or outsr, as a consciousness onlv of v.;hat is 114 dependent on our sensibility". Outer objects are re-

presentations,
115
is snace":

"-vvhich

are entitled outir because they

^-/hose

depend on v/hat we call Pouter sense*,

intuition
116

a representation is an "inner object if it

is represented only in its time -relations".

Thus, in

the final analysis, inner and outer sense must be

116
re;3arded as aspects of possible experience, "the objective

form of experience in general, (which) contains all

117

synthesis that is required for knovjledge of objects".

One fiirther point concerning inner sense remains

to be dealt with.

As erapirically real, inner sense refers

to the network of time-relations throughout possible experience, but the inner sense to v?hich Kant refers in
the explanations of the conditions of the possibility of

this experience cannot be a netvjork of relations, but

must be the pure intuition of time itself.

Apperception and its S3mthetic unity is, indeed, very far from being identical xi/ith inner sense. The former, as the source of all combination , applies to the manxfold of intuitions in general , and in the guise of the cateroFies , prior to all sensible intuition, to objects in general . Inner sense, on the other hand, contains the raere form of intuition, but without combination of the manifold in it, and therefore so far contains no determinate intuition, which is possible only through the consciousness of the determination of the manifold by the transcendental act of imagination, , ,11S
'

Just as the reproductive imagination presup'119 so our emposes a productive imagination, a priori .

pirical inner sense presupposes a pure inner sense, which

contains a manifold, the content of the pure intuition of

time, capable of determination in accordance v;ith the

categories.
The understanding does not, therefore, find in inner sense such a combination of the manifold, but 120 produces it, in that it affects that sense.

117

After unfolding so many different possible meanings of the distinction between inner and outer sense, it

may be advisable to

sura

them up.

Starting at the tran-

scendental level and working down, inner sense may be

regarded as that v/hich contains or comprises the pure

intuition ox
tirae;

outer sense is that which constitutes

As such, inner and outer

the pure intuition of space.

sense are separable presuppositions of experience in

general.

But although they are both a priori presup-

positions of experience in general, they are not related

to experience in the same way.
The form of inner sensi-

bility, namely, time, is that in vfhich all appearances

whatsoever are ordered; the form of outer sensibility is

that in which only those appearances called spatial are ordered.

Inner sense relates to all experience; outer

And the recognition of the

sense to only a part of it.

difference between inner and outer sense, respectively,

depends on the intuitive grasp of time and space, themselves.

Considering the whole of possible experience,

'121 of vdiich Kant never doubts the exthe "one nature",

istence, inner sense must be regarded as identical with

the temporal aspect of the v;hole of it.

The advance of

perception is a revelation of what is in inner sense,

lis
"Tor the original appcrcoTntion stands in relation to 122 inn^r sense , ( the sun of all representc-tions )". In

this

v.'idor raeanin^j,

inner sense may be considered to be

the totality of all eripirically real appearances, "which

-123 in the end reduce to determinations of inner sense",

and time is "the form of inner intuition wherein all por124 ce^'tioriL; h l^':.^ a position". In thid wide meaning of inner
sense, outer sense must be regarded as a part of inner

sense, for inner sense is our mode of inv/ardly intuiting

ourselves, but as vjell it is "the mode of intuition in

of which we like^;ise take up into our facilty of 125 representation all outor intuitions ". Empirically extei-ms

ternal objects are "thin,",s v;hich ar3 to be found in 126 space", and "space is a form of that intuition vjhich 127 we call outer". Thus, for Kant, inner sense in its wide,
-

concrete meaning comprises the whole of phenomena, all

of which is in time, and within this whole, cuiier sense
co.-.iprises

those appearances which, as well as bein-^ in

time, are also in space. As we have already noted, if abst--action is

made from all the matter of sensation, inner anC outer

sense
'..'ill

comprise networks of temporal and spatial rev/oi^ld.

l-tions, respectively, pervadin<5 the phenomenal

119

Finally,

the narrow meaning of inner

sense, as that intuition by v/hich we are conscious of our

tr-^in of non-s;-.atial

thoughts,

"

aelf -l:no^^led-;e by more

inner concciousnass, that is, by dGtormination of our

12d

nature 'vithout the aid of outer emi-irical intuitions".

In this Tiieaning, inner sense contains, thou(::hts, con-

Gciousness, desires, which cannot be outwardly intuited, 129 "All these belon;,T to inner aense". Inner sense in this
rceanin^ has the soul or self as an object, -"Now I am

an object of inner sense, and all tine is merely the form

130
of inner sense".

For Kant, inner sense in this narrow

ide

mean in:.; is distin,j;ui3h3d from inner sense in tha

raeanin^j;

not intrinsically as

ti-jo

absolutely separate

spheres, but only so far as inner sense in the two dif-

ferent meanings has a different object of consciousness,

that is different kinds of representations in the phe-

nomenal

.orli,

,hile inn";r sense in the

v.'ide

meaning com-

prises all appearance, inner sense in the narrow raecming

comprises the peculiar non-spatial thoughts, desires,

etc., of the so-ul.

But both kinds of appearance, the

spatial and the non-spatial are appearance, and nothing

more, so that if we take Kant'J repeated assertions tliat

all appearance is contained inner sense, (in the

meanin,;:),

^..'ide

we must include inner sense in the narraj raeaning,

120
x.'itliiii

ol.l'i

larger v;hola,

ough the I as represented throu^jh inner sense in time, and objects in space outside me, are Si.:.ciric.--lly quito uiGtiiict ai rea. ances, they are not for that reason thou' ht s^'bein- different things. 131

Opposed to

xni.cx

..onae

m
quf.

.:;..

i-a:.-j.

a;

meinlng

is outer senae in Ihe abstract cleaning, outer objects

inouited in sj^atial relations,

spatial x-elations only .

aS such, certainly, time is not a determination of them,

and as Kant hastens to jjoint out, "it has to do nsither

with

shax-;e

nor position, but with the relation of re132

pr^esentations in our inner state".

Space and time,

then, are ultimate and irreducible and absolutely

distinct from ona another, just as the sense of space

and the sense
oi'

tine,

(outer and inner sense in the

narrovver, abstract meaning), are distinct from one another.

Inner and outer sense in these narrower meanings are both

abstractions from concrete experience;

reach the
piire

throxi^^h

them

v;e

intuitions of time and space.

They are

products of a slightly different mode of abstraction.

Inner sense abstracts from all appearances only those

v/hich are non-spatial;
i"rom v.'hat is

outer sense abstracts the spatial

also in time-relations in concrete exporBut inner

ionce, naiaely, outer appearances in general.

121 and outor sense, a.

tality
sense,
ox
v"
ouov.i-

.-J3';r
_

Cwionc,
_

' r:l.ted to tho to.'1:73.

o-r-v"

Inner

Lme, is related to the ^liolo of

appearance, wuilc

sense is related only to a part

within this v/hole.

Inner sense in the narrower meaning

thus not only comprises the empirical subject natter of psychology, i.e. thouchts, non-spatial appearances, it
also in 3ome fashion reveals inner sense in the trans-

cendental meaning, that inner sense, which, containing

the nere form of intuition v/ithout combination, is a
pre sup ,^0 sit ion of all possible experience, and thus is

related to all c:cn3rience as "the one v/hole in vjhich all

133
our-'
;

resentations are contained",

i^^or- inner experience in general, isia its possibility, or perception- in general and its rel^ri-iou to othor perception, in .-.'hich no spscial distinction or empirical determination is given, is not to JO rejja-'ded an eiipirical'laiouled:-;.3 but as knov.'ledge of the empirical in general, and has to be reckoned^ u'ith the inve3ti.;j;:-.tion of the posoibiliti'' of any rnd every experience, which is certainly a transcendental Gn-piii-y. 134

122
Tine and the Unity of Ilature

3.

We have already dravm attention to the point that

in so far as a fvindaiaental distinction betv/een space and time

as extensive continua is made, there are

t;7o

possibilities
V/e

as to

hov;

individuals may appear in these media,

have

discussed these possibilities only with reference to the

category'-

of substance, but we must now consider them in con-

nection with those wider principles vAich bring about unity

in nature.
is
V/e

must determine, moreover, how far this analysis

applicable to Kants philosophy, and to what extent it

nay be said that Kant^s views represent that position which

we have called possibility B.

This necessarily involves a

decision as to whether time is more important than space in

the unity of nature which Kant proposes to explain*
It is commonly thought that Kants philosophy re-

presents an ansv/er to Humes scepticism, and in particular

to Ilume^s criticism of causality.
In the course of our in-

vestigation we attempt to show that Kant does, indeed,

formulate an ansvrer to Hume, and that this answer depends
on the role of time in the unity of natijire.

Possibility B is stated as follovjs:

There is a principle of individuation for time, and

this accounts for objects in space.

123

More explicitly, possibility B requires that principles which govern the activities peculiar to the self,

which are manifested by means of time-determinations, account

for objects in space.
Just as laws which govern the move-

ments of objects (substances) in space are thought, according

to possibility A, to determine in general v;hat kind of events

will occur in the time -continuum, so according to possibility

B, the rules which govern the activities of the self in re-

lation to time nust determine in general what kind of objects shall appear in space.
In both cases there must be

both constitutive and regulative principles.

only must it be shovm
hoxv

That is, not

events and objects, respectiveljr,

come to be, there must also be principles which indicate

their natures and their relationships.

Now either of the positions can be given with imaginative connotations,- such as the conception of a self-

existing substratum beloxv or beyond the two media with their

given differentiations,- or each can be reduced to the bare

logical essentials which are necessary as principles or laws

of the unification of nature, according to either possibility,
It is with this latter contingency that we wish to deal.

Since philosophers are ultimately all attempting to explain

the same Nature, it is not surprising that the logical

essentials of the two possibilities should show a marked

124

similarity.

There is, of course, room for much disagree-

ment concer/:J.ng those basic principles v/hich are presupposed

by empirical
lav/s

of nature, and in this respect there is

But in

room for dissirailarity betv/een the two positions.

so far as nature permits of quantitative deteiiainations, the

category of quantity will be essential to both; and in so

far as nature exhibits enduring entities, some doctrine of
substance will be necessary.
Similarly, a recognition of
seera

causation and quality would

to be imavoidable .

These

are, perhaps, the most notable points on which there may be

agreement between the two positions, although even concerning

these, agreement may not be complete.

Considering, for

illustrative purposes, such principles as these to be examples

of common principles, then, the main difference between the

two positions will be that the one will

consider these

principles to be ?ssociated with matter and space, and the

other will consider them to be associated with the subjeet

and time.
3ut v;here material substance and the subject

are thought of simply as complexes of functions embodying

the necessaiy principles, the

tv;o

positions will represent

a clear antithesis such as we have drawn betvjeen the two

possibilities.

Sither one must explain events in time

through principles pertaining to space, or one must explain

objects in space tlirough principles pertaining to time.

125
On epistenological questions, the two positions may

agree or disagree.

Either matter or

tjie

subject may be con-

sidered rationalistically so that the principles pertaining

to each nay be conceived as necessary and apodeictic.

Al-

ternatively, matter may be thought to exhibit only probable

laws, or the principles pertaining to the self may be con-

sidered to be merely probable, as for example, Hume's

of association.

lavjs

The intrusion of epistemological questions,

however, reveals the close connection between the self and

time.

Hume's criticism of causation woiJ.d have been impos-

sible without his distinction between the future and the

past. That is, it v/ould not have been possible for Hume to

maintain that the primary impression of the idea of necessary

connection derives from customary association if he had not

believed that huiaan experience was ordered in time in such

a way that there were past repetitions which may or nay not

be exemplified in future events.

Similarly, as we hope to

show, Kant's Copemican revolution would have been impossible

had he not shown that the principles pertaining to the knov/ing subject, v;hich are constitutive of objects, have appli-

cation to experience through

tir.ie

alone.

For Kant, the whole phenomenal world has reality

only in relation to the cognitive faculties of the self, or
the human mind.
But those cognitive faculties which produce

the order and unity of this phenomenal world are not

126

subjective in the psychological sense.

us a genetic psychology.

Kant is not giving

In what sense, then, does the

phenomenal world have its character only in relation to the

knov;er?

The whole transcendental apparatus, including the

Transcendental Aesthetic, the categories, the Transcendental

Unity of Apperception, the Transcendental Iraagination, is

not psychologically subjective, and does not issue from the

activities of a numerical plurality of nxman minds.

Ue

must be entitled to say in this regard that if Kant's philosophy is trua, it must have
life ever arose on earth.
heeii

true long before human

Granted that the phenomenal world

has raality only in relation to these transcendental conditions, in what sense are we entitled to call these conditions

subjective?

As Patoii points out, the self in relation to

which the phenomenal world is real, (as appearance) is not

the phenomenal self.

If the self which is knovm in inner sense is only phenomenal, what are we to say of the self which Icnows? Is the kno\ving self a thing-in-itself , although the knoivn self is only an appearance? To this question Kant's answer is obsciore: but perhaps v;e may say, in the light of his moral philosophy, that the -self does belong to the realm of things-in -themselves, although 135 as a thing-in-itself it can never be known by us.

If it be conceded that the aelf which knov/s must be

a taing-in-itself , on what basis do we distinguish one thing-

in-itself from another as subject and object?

127

Above all, can v;e believe if this is Kant* doctrine --that the v;orld as vje experience it is due' to the interplay of t^;o unknovm thin^js-in-thenselves, one of which is a self. v;hile tiie other is perhaps not It is difficuJ.t to accept one \jholly xmknovjn a self? factor. It is alr.iost impossible to accept two. If they are v;holly unknown, how can they be distinguished from one another? I36

Even if we delate the thing-in-itself from Kants

philosophy, and consider that his main point lies in his

discovery of the logical presuppositions of empirical knowledge, the difficulty remains. If

v;e

strip Kant^s termino-

logy of its psychological connotations, the nature of the

difficiil.ty becomes clearer.

A basic element in Kants posi-

tion lies in his view that universal and necessary preposi-

tional forms lie at the base of the unity of nature.

This

order of one nature is, we take it, quite independent of any

particular judgments made by human beings.

As Kant himself

often points out, the activities v/hich make up the psychology

of any particular human being are just another part of the

phenomenal world, in so far as they are sensational.

Thus,

the subject-object distinction of psychology is quite irrele-

vant to the subject-object distinction which figures in the

Copeinican revolution.

The knov/er in relation to which the

phenomenal vjorld

;:ets

its order and unity is not the knower

How, then, can this knower, as

qua psychological subject.

a transcendental presupposition of empirical knov/ledge, be

distinguished from the ordering principles of an objective

I2e5

world?

Kants viow seems to be that the intelligible, the

l-.nov/ledge

logical presuppositions of empirical

pertain to

the activities o an ego or self, and yet his v/hole criticism

of rational psychology rests on the point

tliat e::cept

for

its purely lo:::ical functions this ego is eiapty.

That the ^I* of apperception, and therefore the in every act of thought, is one ; and cannot be resolved into a plurality of subjects, -and consequently signifies a logically siinple subject, is something already contained in the very concept of thought, and is
'I

therefore an analytic proposition. But this does not mean that the thinking I* is a sinple substance . 137
It is evident that one reason at least for Kant's

association of the ordering principles of nature with an

ego or subject lies in his view that logic pertains to the

thinking activities of a subject, and in his doctrine of

judr-yient .

But if
h\ar.ian

v;e

approach the question from the point

vje

of viexv of

knov;ledge, as Kant did,

discover a se-

cond set of reasons for making the subject-object distinction

at the transcendental level.
A priori knowledge is know-

ledge which can be had prior to experience, i.e. knov/ledge

which is independent of experience and v;hich does not issue

from the senses.
Thus a priori Icnowledge is independent of

the passage of tine.

Because the nind contributes the form

v/e

of possible experience,

can according to Kant, Icnow some-

thing about future experience prior to our having it. Kant,

indeed, does not linit his statement of the scope of a priori
knowled;;e to future experience, but extends it to all possible

129 experience.
that
v.'e

Nevertheless, his doctrine necessarily implies

can know something about future experience prior

A posteriori knowledge, on the other

to our experiencing it,

hand, is knowledge v/hich must await experience, and which

therefore depends on the passage of time.

If our knowledge

conforms to objects, we can know nothing of objects prior

to what is revealed in experience vjith the passage of time.

We can know something about objects prior to experience only if objects conform to our peculiar modes of cognition. If intuition must conform to the constitution of the objects, I do not see how we could knavi anything of the latter a priori but if the object (as object of the senses )''mu3t conform to the constitution of our faculty of intuition, I have no difficulty in conceiving such a possibility. Since I cannot rest in these intuitions if they are to becone knotm, but must relate them as representations to something as their object, and determine this latter through them, either I must assume that the concepts , by means of v/hich I obtain this deterr.iination, conform to the object, or else I assume that the objects, or what is the same thing, that the experience in v;hich alone, as given objects, they can be "kn ovm , "c onf orm to the concepts. In the former case, I am again in the same perplexity as to how I can know anything a priori in regard to the objects. In the latter case the outlook is more hopeful, 13^
;

This is Kant^s statement of what is involved in the

Copernican revolution. for its success: concepts v;hich
ma^?-

There are three points necessary

Kant must shov; that there are
piur-e

(a)

be constitutive and re.e;ulative of our

(b) he

knowledge of objects;

must

shoii?

that these cate-

gories do necessarily apply to objects of knowledge, or to

put it othenvise, he must show that if
vje

are to have any

130
knowled^je of oojcc^s oi experience at all,
v;e

must think

them by neans of the categories;

(c)

he must show that ex-

perience is, in fact, constituted in such a fashion that

the categories are constitutive and re^^'ulative of it.
In

order to make clear v/hat

vje

mean by ()

perhaps it would

be better to speak of the phenorienal world instead of ex-

perience.

For this third requirement does not concern a

genetic psychological explanation, a thing quite foreign to Ilant's purposes, but concerns how the categories are, in
fact, constitutive and regulative of objects in the phenome-

nal world.

Without the fulfilment of this third requirement,

Kant has himself

Kant*s Copemican revolution is incomplete.

clearly acknowledged the necessity of these three require-

In the metaphysical deduction the a priori origin of the categories has been proved through their complete general logical functions of thought; in the transcendental deduction x^e have shovm their possibility as a priori nodes of knowled;,e of objects of an intuit ion^in general. We have nov; to explain tlie possibility of knowing a priori by necjis of cdte[:,ories whatever objects may present thenselves to our senses , not indeed in respect of the forn of tiieir intuition, but in respect of the lav;s of their conbination, and so, as it vjere, of prescribing laws to natujre, and even of making nature possible. For unless the categories discharged this function, there could be no explaining why everything that can be presented to our senses must a be subject to la\\?s which have their origin "" priori in the understanding alone. 139
agreer.ient witl.. the

Cur chief interest in this third requirement, the

second part of the Transcendental Deduction as stated in the

131

second Edition of the Critique , lies in the point that

tlie

"combination" of which Kant speaks must take place in accordance with categories schematised by means of time-detemina-

tions alone.

For we could never knov; anything prior to ex-

perience if the passage of time itself did not embody those

very principles v/hich are said to contribute the form of

possible experience.
If the passage of time
t;ent on

inde-

pendently of these principles, we should never be entitled

to suppose that the future v/ill resemble the past, or that

future experience v;ill have any of the characteristics of

past experience.

Although the principles themselves are

independent of time, they must be constitutive and regulative of objects of experience through time, and only through tine : otherwise there v;ould be no necessary reason for

supposing that these logical principles, rules of s^mthesis,

were constitutive of future experience.

It is just this

requirement which must be met if Huine*s statement that "the

supposition that the future resembles the past , is not

founded on arguments of any kind, but is deriv'd entirely 140 from habit" is to be refuted.
Thus, if r.ant is to be considered as giving a con-

clusive answer to Hume, he must

shov;

that contrary to Hume's

contentions, we have reason to believe that the future will

resemble the past.

This can be shovm not merely by exhibiting

the principles constitutive of objects of knov/ledge for

132

intuition in ^^eneral, but by showing that the passage of

time, v;hich allov/s a continuous increase of sensible intui-

tions, combines these intuitions in accordance with the

principles constitutive of objects.

In other words time

itself must bring about the existence of objects of future

experience.
The first
pairt

of Kant^s Transcendental Deduc-

tion does not suffice to prove that this vjill be the case,
as Paton observes.

To put the matter xu ^acu>'^er vjay it night be a were accident whether appearances conformed to the

categories or not. No doubt if they did not conform, we should cease to have experience; but in that case so much the v;orse for us. oo far as the argument has gone at present, the conformity of appearances to the cate.~ories must indeed have occurred; for there has been such a thing as human experience. That conformity might, hov;ever, have been due to a pre-established harmony arranged by tjie beneficence of God; and any view of this type is emphatically rejected by Kant. 141
Ue might show conclusively that
v/e

can know of ob-

jects only those

v.'hich

conform to our peculiar modes of cog-

nition, but this does not by itself in the least show that

we shall ever in the future meet with such objects.

The

statement that either knowledge conforms to objects, or

objects must conform to our ways of knowing objects, and
the defense of the latter alternative, does not refute
Hxime*s position.

Dven though objects must conform to our

knovjledge, we are not entitled to assuiiie th.at future ex-

perience will exhibit objects v;hich conform to our knowledge,

unless time itself and its irreversible passage is shown to

133
brin.-:;

such ob.jects into existence .

v;e

mi^ht cease to have

i.e. as empirical

experience in Kant's sense of the

knov/ledge.

terra,

But is that not just Kunies point, namely, that

human experience does not necessarily represent empirical

knov/led^e, since there is no proof that the future v/ill

resemble the past?

Thus, if Hume is to be effectively an-

swered, the a priori principles constitutive of objects

must manifest themselves by means of time-determinations,

and not tlirough spatial determinations.

This is,

;-;e

feel,

a reasonable explanation of why Kant has schematised the

categories by means of time-determinations alone.

There are, however, other reasons for this conclusion.

Some weight must perhaps be given to the fact that

Kant had ample opportunity to repair his oversight in omit-

ting space from the Transcendental schematism, if, indeed,

it represents an oversight on his part.

He could, if he

had desired, have included spatial schemata in the second

edition of the Critique .

But he did not avail himself of

this opportunity.

V/e

must conclude, then, that either his

acumen failed, or that his philosophy did not necessitate

spatial schemata.

Even if Kant was not prepared to draft

a lengthy explanation of these schemata, he could easily

have included a foot-note indicating that later v/orkers in

the new science might appropriately fill in the lacuna.

134 For Kant, tine cannot be said to be an ultimate

reality.
It is not a thing-in-itaelf .

Yet time, as a de-

termining condition throughout the phenomenal world is allpervasive.

It is through time alone that the catetiories

become schematised, and hence are able to determine the

character of appearances.
We thus find that the schema of each category contains and makes capable of representation only a 142 deteriiination of time.

The principles which represent the rules under which the existence of all appearances is ordered are all deter-

ninations of time, and

tirae

alone.

The principles can therefore have no other purpose save that of being the conditions of the unity of empirical knowledge in the synthesis of appearances. But such unity can be thought only in the scheEia of the 143 pure concepx; of the understanding,
The unity of empirical knowled^ce is a result of the

function of the original synthetic unity of apperception,

and it is this
tliat is

the condition of the regulative and

V/ithout this synthetic uiiity of

constitutive principles.

apperception there could be no unity of experience at all.

But this unity is manifested in principles v/hich express

only time-detenainations.
In this third [medium], the essential form of which consists in the synthetic unity of the apperception ox all appearances, ..'e have fovoid a priori conditions of complete and necessary determination of time for all existence in the (field of] appearance, without which even empirical determination of time 144 would be iiapossible.

135

Kant hinself gives the clearest indication that he

is maintain inij the essential requirement of possibility B,

namely, that principles or

I'loles

which bring about the unity

of nature apply directly to time ibself and not to space.

Time, indeed, is not given whole as the deter.ainable in which

every experience finds a position.

Every appearance is de-

termined in its position in time because the categories

apply directly to tiae.

irnat detormines for each appearance its position in tiue is the rule of the understanding through v/hich alone the existence of appearances can acquire synt'xetic unity as regards relations of time; and that rule consequently deteiTiiines the position [in a manner that is) a ])riori valid for each and every time. 145

Also, Kant gives the clearest possible expression

to the other main requirement of possibility B, namely,

that everylihing else in nature is determined by time, and
not the other
v/ay

around as would be the case according to

possibility A.

All increase in empirical knowledge, or ex-

perience, Kant tells us, is nothing but an advance in time.

But follov;ing this statement is a most important assertion
in which Kant says, vjithout qualification, that
" This

ad -

vance in time deterraines everythinp; . and is not itself

determined through anything further

der Zeit bestiramt Alles
ijiid

Dieser Fortgang in

ist an sich selbst durch nichts

weiter bestimmt)'-.

In addition, Lant hov;s himself to

have been fully conscious of the point we have mentioned

136
in connection with :;iving an effective ansv/er to Hume,

namely, that a priori principles constitutive of objects

must nanifost thsiaselves by means of time-determinations.

It is Just Kant's point tliat futt.ire experience must have
a priori features because the advance of time itself guaran-

tees it.
In the sane manner,, therefore, in vjhich time contains the sensible a priori condition of the possibility of a continuous^advance of the existing to v;hat follovvs, the understanding, by virtue of the unity of apperception, is the a priori condition of the possibility of a continuous determination of all positions for the appearances in this time. 147
It is true that here Kant is referring principally

to causation, but it is clear that all of the other categories

must

likexi'ise be

determinative of objects through time alone.

For if they were not, the certainty of our a priori knov;ledge

in respect of them would still be endangered by Hume's inex-

orable claim that we have no proof that the future will resemble the past.
ISven

though we might be certain that every

v/e

future experience will be causally determined,

would not

be entitled to assume that it would have any other a priori

features.

ICant

must therefore show that all of the catego-

ries have application to experience by means of ti:ue-deter-

minations.

This is, in feet, v;hat he does, and so

vje

can

say that his answer to Hume is complete.

Kant held that nature was a dynamical iMhole rather

than a purely Eiechanical one.

He was suspicious of the

137

concept of a void, or empty space and sought to show, in

the Critique of Pure Reason and in the Iietaphysical Foun-

dations of liatural Jcieace that the concept of empty space

was unnecessary for physical science.

For this reason alone

we can say that Kant definitely rejects possibility A, in

which the givenness of all space either as a relational

complex or an absolute container is an essential point.
Although the spatial aspect of appearances is not by any

means ignored in the adumbration of the ilnalytic of Principles,

Kant expresses these principles in terms of determinations
of time.

The concept of magnitude in general can never be explained except by saying that it is that detemination of a thing vjhereby we are enabled to think how many times a xmit is posited in it. But this how-manytimes is based on successive repetition, and therefore on time and the synthesis of the homogeneous in tiirxe. Reality, in contradistinction to negation, can be explained only if we think time (as containing all being) as either filled with being or as empty. If I leave out permanence (which is existence in all time , nothing remains in the concept of substance save only tne loI4S gical representation of a subject
)
-

Only the three Postulates of Empirical Thought are

elaborated v/ithout specific stress on time, although these

three postulates are so defined as to presuppose the other

principles v/hich are stated in terras of time-determinations.

In any case, as Kant remarks, "The principles of modality 149 are not, hov/ever, objectively synthetic". They add nothing

to the order in time -relations of the objective existence

136
of appearances, but rather depend on that ordered existence

for their usefulness as niles of guidance concerning the

possible, the actual and the necessary in nature.
It is different with the other principles.

They

are rules which enter into the dynamical process of nature

itself.

The Analogies of Experience, for example, are "Prin-

ciples of the determination of the existence of appearances

in time, according to all its three modes, viz., the relation

of tine to itself as a magnitude (the magnitude of existence,

that is, duration), the relation in time as a successive

series, and finally the relation in time as a sum of all 150 simultaneous existence". Similarly, even thoizgh the .^ioms

of Intuition and the Anticipations of Perception are mathe-

matical rather than dynamical principles, that is, they are

necessary formal conditions of experience pertaining to

'151 rather "the mere intuition of an appearance in general",

than with the dynamical, on-going progress of nat\ire as it exists , they are nonetheless all dependant on time .
This

requires fizrther comment.

One concept v/hich is involved in the /ixioms of In-

tuition and the Anticipations of Perception is quantity or

magnitude, the schema of v;hich is number, which is "simply
the unity of the synthesis of the manifold of a homogeneous

intuition in general, a imity due to my generating time

itself in the apprehension of the intuition".

sistentlj'-

139 152 Kant con-

tells us that magnitude comes about by means of

t/ithout this

a successive synthesis of part to part,

successive synthesis,

v/e

c,&t

no detenrninate intuition.

Inner sense, on the other hand, contains the mere form of intuition, but without combination of the manifold in itj and therefore so far contains no determinate, intuition, which is possible only throu^K the consciousness of the determination of the manifold by the transcendental act of imagination (synthetic influence of the uiiderstanding upon inner sense v;hich I have entitled fisurative" synthesis, 153
)

In support of this statement Eant draws attention

to the fact that in order to conceive of a magnitude, we

must construct it in thought, by a successive synthesis.
This vje can always perceive in ourselves.- V/e cannot think a line without drawing it in thought, or a circle without describing it. 154
Kant holds that this is true not only of geometry
but also of numbers, for, indeed, the schema of quantity
is niaraber.

Thus our coxmting, as is easily seen in the case of larger numbers, is a synthesis according to concepts, because it is executed according to a common ground of unity, as, for instance, the decade. In terms of this concept, the unity of the synthesis of the manifold is rendered necessary. 155
It is clear that space and time, or rather the spa-

tial and temporal aspects of the phenomenal world, have

magnitude only because of the successive synthesis due to

the transcendental imagination.

This is why, according to

In a passage

Kant, space and time are continuous magnitudes.

140

of unusual clarity, Kant explains how determinate magnitudes

of space and time are generated.

Such raagnitudes may also be called flowing , since the synthesis of productive imar-ination involved in their production is a progression in tiiae, and the continuity of time is ordinarily designated by the tern flowing or flowing axvay. I56

This reveals, as well, a point of comj>ari3on v;ith

Bergson, namely, that continuity does not stem fron the in-

tellect itself, but is a property of time.

this point later on.
V/hat is

We shall discuss

important here, however, is

that quantity of anjrthing in nature in general, is the result of a successive synthesis in time.

Thus the principle

of the Axioms of Intuition, namely, that all intuitions are

extensive magnitudes, does not depend on space itself, or

the application of the concept of qtiantity to space, but

depends on a successive synthesis in time, which is the means

whereby all quanta arise.

Kants doctrine seems to be that

the successive transcendental synthesis by means of time-

determinations, that is, by means of the schema of quantity,

brings determinate spaces into existence.

synthesis there are no objects.

Prior to this

Objects are not spatial

except in so far as spatial magnitude is generated by means

of the successive advance of time.

That is, as time itself

is extended by successive addition of part to part, space

is also generated in the sane synthesis
.

as a determinate

quantum.

For this reason, no doubt, Kant can say that as

141

forms of intuition neitlier space nor tine are objects.

The mere form of intuition, without substance,
is in itself no object, but the merely formal condition of an object (as appearance), as pure space and tine (ens iiaa^inarium) , These are indeed something, as forms of intuition, but are not themselves objects which are intuited, I57

As we have seen, outer sense is an abstract aspect

of inner sense,

l.'hat

is in space is iijso facto in time.

Thus, we can say that in that area of appearances which are

in both space and time, the transcendental synthesis pro-

ductive of these appearances is a successive synthesis of

both space and time, productive of determinate magnitudes

of space during a determinate length of time.
^rovrtih

This parallel

of spatial and temporal determination seems naturally

to sujsest that it is necessary for Kant to say that the de-

termining principles must apply to both space and time.

But

this is not only unnecessary; it would be of no use to Kant

in carrying out the essential tasks of his Gopernican. revo-

lution.

This is a corollary of what \m

liave

said about

Kants answer to Hume,

Let us suppose that there are spa-

tial schemata of the categories vjhich enter into the con-

stitution of objects of experience, guaranteeing that objects

shall bx that means have certain general a priori features.
By what right are
v;e

entitled to suppose that such spatially

extended objects will ever be present to us in experience?

142
:r the spatial features of objects ware insured by that

means,

vje v;o\ild

not necessarily

liave

synthetic a priori

knowledge about the spatial constitution of future experience (in the Humean rather than the Kantian sense), because

we would have no guarantee that future experience would reveal such objects.
Future experience might, for all we know,
It is undoubted-

reveal the utter emptiness of such knov/ledge.

ly hard to
ina,::ine

such a contin::ency, but our subjective

it, does not, in itself, constitute a

inability to

inia:::;ine

conclusive argument that the future v/ill reserable the past.

For this reason, spatial schemata would, so to speak, be a

fifth wheel in Kant's system.

Hume's very point is that

psychologically

v;e

have become accustomed to certain regular

vje

features of experience, and for that reason

expect a re-

petition of them in the future, and Hume shows conclusively,

if he shows anything, that this psychological conditioning

gives us no certairity ii/hatever about future experience.

And once this point is admitted, we
raay

say that as a matter

of fact objects do have spatial features, but that is far

from guaranteeing that they must have them.

Thus, just as

Hume's arguments rest An the fundamental distinction be-

tween the future and that past, so likewise, Kant's arguments that objects of experience must have certain a priori
spatial features must rest on the view that the passage of
time will bring such objects of experience into existence.

143
.Ind
\iQ

for this purpose spatial schemata are xiiisuitable, Tor

must reckon with the possibility that the spatial features

Kant,

of objects are merely accidents of a particular tine.

himself, reckons v;ith this possibility, for althoush he ar-

gues that we can never in principle represent to ourselves

the absence of space, "v/e can quite well tliinfc it as empty 15^ of objects". That there are objects in space, and hence
havini^ certain spatial features, must depend, then, on time-

determinations.
din^2

Thus, it is only because na^initude, accor-

to Kant, is alvjays the result of a catesorical deter-

mination of tiue, that space itself can have magnitude,

I cannot represent to myself a line, hov;ever small, without drawing it in thoui^ht, that is generating from a point all its parts one after another, 159

The mathematics of space (geometry) is based upon this successive synthesis of the productive imagination in the generation of figures. 160

The Axioms of Intuition do not, therefore, we main-

tain imply spatial schemata, but quite the reverse.

It is

just because the spatial features of objects do not depend

on the direct spatial schematism of the concept of quantity,

that we can say a Toriori of experience in general (including

future experience) that objects of experience must in their

intuition, be extensive magnitudes.

This implies, hov;ever, an intimate connection betv;een

space and time, the two fundamentally distinct forms of in-

tuition, such that if a certain quantum of time is generated^

144
a certain quantuu of space will by this process also be

iienercted.

If our analysis ox the meanini^s of inner and

outer sense is correct, hoxvever, this automatically follows.

For if outer sense is an abstract aspect of iiiner sense in
the wide meaning, then, if time is geuex^ated in that area

of sensibility which has a spatial aspect, a determinate

magnitude of space nust also be generated, just because

outer sense is only an abstraction rrom inner sense.
But

this implies soae kind of union of space and time, and from
the standpoint of the understanding this is a perplexing

thing.
Here we must recall what we have said in Chapter
Two, Section One, about the difficulties of
co::^ni2iing

the

relatedness of space and

ticulars.

tiine if
v:e

they are two unique parcannot understand this

Strictly speaking,

union, for if they are essentially unlike, there is no-

thing in

comrr.on

between then.

For this reason we have said

that in thinking about inner and outer sense, the emphasis

must be placed on sense .

concepts
hovj

We can never eiiplain by means of

it is that an advance of tirae can generate a

x^e

magnitude of space, any more than

in general.

can explain becoming

For this reason, we take it, Kant very properly

does not attempt to explain it, but refers to our acquain-

tance \;ith this union, in effect, by sai'ing of the transcen161

dental synthesis,

"this

v;e

can always perceive in ourselves*'.

145

That Kant is rererrin^i to a kind of becoming can be seen

from the fact that he refers to it in terras of inotion.

does not rotcrd

He

to be able to deduce a priori the notion

of any particular object.

Motion, ho\;'ever, consic'.ered c.s the descriDing of a space, is a pure act of the successive sjmthesis of the laanifold in outer intuition in general by neans of the productive iciar:ination. . ,l62
For Kant, only time is intrinsically successive, or
serial.

Space, in itself, is an aggregate not a series,

Kant tells us, furthermore, how we happen to have the concept of succession.

We derive it from time itself, as the

form of inner sense, and the 'motion* of synthesis in time,

Motion, as an act of the subject. ..and therefore the synthesis of the manifold in space, first if we abstract produces the concept of succession from this manifold and attend solely to the act through which we deterr^.ine the inner sense according to its form. 163

Unless we are prepared to reject these statements

out of hand,
i;e

must admit that Kant is here attempting to

explain not becoming itself, but the manner in which the

categories apply, as principles, to the passage of time

itself, and hence
hov;

time brings into existence determinate

quanta of space.

It must be added, of course, that Kant

later attempts to specify only in ;^eneral the a priori spa-

tial featiires of objects.

He does not try to tell us why

a particular tning is round or square, etc., a matter which,

like particular motions, is a thin^r about v/hich

\ie

can never

146
have a priori traowieajo of such a specific character.
But

it seems clear that this ^:eneral feature which we can know

a priori about objects of experience, depends basically on

the fact that the categories, by means of schemata expressed

as time -determinations, apply directly and solely to time

itself, the form of inner sensibility.

The understanding does not, therefore, find in inner sense such a conbination of the manifold, but produces it, in that it affects that sense. 164

We have also proved that the only manner in which objects can be given to us is by modification of our sensibility; and finally, that pure a priori concepts, in addition to the function of un'^er standing expressed in the catec;ory, must contain a priori certain formal conditions of sensibility, namely, those of inner sense. I65
On the other hand, the schema of a pure concept of understanding can never be brought into any image whatsoever. It is simply the pure synthesis, deterriiined by a rule of that unity, in accordance with concepts, to v;hich the category gives expression. It is a transcendental product of imagination, a product v;hich concerns the deterraination of inner sense in goneral according to conditions of its form (time), in respect of all representations, so far as these representations are to be connected a priori in one concept in con166 formity with the unity of appercsption,

The unity and consistency of Kants thought is truly

remarkable, for he does not in any essential respects depart from his view that the order of Nature is the result
of the categorical determination of time.

Even the a priori Time and the prin-

spatial features of objects depend on it.

ciples relating to it are logically prior to the spatial

147

aspect of the phenomenal world,

Gince it uas Hume who

v/e

awakened Kant from his "dogmatic sliombers",

may conclude

that Kant took particular pains to ensure that all synthetic

a priori knowledge,

(including that relating to the a priori

I-Iuraes

spatial features of objects), should escape

destruc-

tive criticism, his relentless dictum that we have no reason

to believe that the future will resemble the past.

Not only are transcendental schemata for space un-

necessary and redundant for Hant's purposes, the acceptance

of them would create grave problems relating to other parts

of Kant's system.

The inclusion of space in the schematism,

and according it an equal status 'with time in the formulation

of the r)rinciples which ensure the unity and order of Nature, 167 to regard the mind itself tempts one, as Paton points out, as spatial.
It raight be more accxarate in this regard to say

that if the categories were schematised in space as well as

time, this would tend to obliterate the distinction between

inner and outer sense on which so much of Kant's doctrine

rests.

We assume that it was Kant's purpose to explain the

If it is true

a priori principles governing on e Nature.

that the principles v/hich ensure that there shall be one

riature must do so by functioning through the media of both

space and time together, then the logical result would seem

to be that the sphere of application of space must be absolutelji^

coextensive with the sphere of application of time.

14fi
[loitf

could there be uiity in the functioning of these prin-

ciples, v/hich, according to this view, require both space

and time for their expression, if they apply to an area of

appearances

v.'hich

are in tine alone?

For if the categories

must be expressed in spatial deteniiinations as v;ell as

tiue-deterrriinations in order that there should be a unity

in Nature, it follov/s that they do not function separately

in their separate media, but function in unison in space

and time.

The v;hole phenomenal world, including pheno-

menal selves, must, if it is to be one unified nature, be

everyt/here spatio-temporal.

For if the categories are the

ground of all determination of appearances as phenomena,

then their schematism through both fonns of sensibility

will mean that all appearances will be deteimined both in

space and time. In this case, it will be necessary to deny that the mind is non-spatial.

Mental activities, of coiu-se, need not be considered to be in space, since they are not appearances.
But

mental activities, or psychological events, would have to

be explained as being, nothing more nor less than events
in the phenomenal world.
It vjould have to be held that

they were in principle amenable to a physical explanation

in the same raanner as any other events in the phenomenal
v;orld.

This v;ould be to take the approach of that theory

which WG have called possibility A, for in the one Nature

149
v/hich is completely spatio-tenporal the only possible events

in tine are those v;hich are marked by the 2iovements,

(changes

of place) of objects in space.

Sven thoijgh it can be said,

on this viev;, that riental activities are in time alone, the

point is that they are In time alone in the same sense, and
only because, all events are in time.
There is thus no dis-

tinguishing between mental events and any other kinds of

events.

They all occur in one spatio-temporal nature.

From

this follov7s the inevitable result that a materialistic

psychology must be formulated, one moreover, which v;ould

explain all aspects of mental activities, including logical thinking,
diates.
/aid

this is a thing which Kant explicitly repu-

For apperceptioa is soiaetlring real, and its simplicity is already given in the mere fact of its possibility. Now in space there is nothing real v/hich can be simple; points v;hich are the only simple things in space, are merely limits, not themselves anything that can as parts serve to constitute space. From this follov-s the inpossibility of .any explanation in materialist terms of the constitution of the self as a merely l6^ thinking subject.

Kants idealism is not to be gainsaid, even though

it figures in his oystem merely as the bare logical func-

tions of the mind.

This is why we have devoted some dis\vhy

cussion to pointing out that the reason

Kant regarded

the unifying principles of Nature to stem from the subject,

rather

ihan as being the ordering principles of an ob-

jective world, rests, at least in part in his doctrine of

150
.jud.-ment

The independent origin of the categories, them-

selves, in the nature of hxaman thotight would be quite in-

explicable in terms of

materialistic psycholof;:'.

We

cannot avoid this difficulty simply by falling back on

their purely logical nature, for locic itself must have

been discovered by human minds, and logical thinking must

always involve mental activities which take place in tiue.

But
hov;

can the forms of judgment ori^^inate in objects?

There is a further awlward result associated with

the viex^ that the categories must be schematised by means
of spatial determinations, and this stems from Kant^s

fundamental distinction between space and time.

V/e

assume

that Kant must have thought that space and time ware ul-

timately separate and distinct, or he would never have distinguished them as

tv/o

different forns of sensibility.

The synthetic a priori knowledge v;hich Kant believes is

possessed in physics concerns, above all, the relations

among phenomena.
The Analogies, taken together, declare

that all appearances must lie in one nature, and this is

because the transcendental ujiity of apperception imposes

this unity on all appearances through one kind of relation
only , namely , the time -relation .
If this tmity is imposed

on two different fon.is of sensibility, then on principle

there seems to be no reason why there would not be two

xinified natures of phenomena determined in space and time

151

respectively.

The simplest, if perhaps the roughest v;ay,

to put this point v/ould be to say that if the schemata give

the cate3ories meaning, and if there are two schemata, one

spatial and another temporal, for each category, end if,

further, there is a fundamental distinction between space

and time, then each category will have two separate and

distinct raeanin:js.

There will thus be

tv;o

natures.

Kant's

i'r.ctives

in schematising the categories by

means of time-determinations display several aspects, and

these are all connected
xv'ith

his fundamental distinction

between space and time:

the mind, or the subject;

(a)
(b)

logical functions stem from

the mental activities, inir^

cluding logical thinking, of human subjects are

aloxne;
(c_)

tirne

a materialistic psychology is impossible; (d) a

complete justification of s'/nthetic a priori knov;ledge,

even that relating to the spatial features of objects, can
be carried out by neans of categorical detenuinations of

time alone;

{e)

for this justification, spatial schemata

are not only not needed, but they could not possibly serve
the p\xrpose.
Thus, in general, there are two basic reasons

for liant's making a fundamental distinction between space and time, for considering tine to be logically prior to
space, and for giving time a position of oven-fheLming im-

portance in the unification of nature.

These are, first,

the indispensable and irreducible idealistic element in

152
i:ant*s system woxild othen/ise have been jeopardized.

The

very l>nich-pins of i:ants thought, namely, the transcendental unity of apperception, and the independent origin of
the categories in the nature of logical thinking, v/ould

have been difficult, if not impossible, to maintain.

Apper-

ception and

lo/:"ical

thinking pertain to the subject, and

in terms of human psychology, it is only the fact that these

mental activities are in time alone and not also in space

that distinguishes them from events in the spatio-temporal

phenomenal world.

Only

som.e

quasi-miraculous doctrine of

emerr;ence could locate them in an exclusively spatio-temporal

Nat\ire.

Kant, at least,

:;,ives

no indication that he

x-zas

tending in this direction.

For hLm, logic and its peculiar

norms are a possession of the mind independent of the world

of physical objects, in the sense that knowledge of them
is not be explained by means of the motions of bodies.

If

they were not independent, it is extremely difficult to

see how Kant could hold that forms of judgipent are prior

to :md constitutive of such objects.

And even as phenomenal

selves, we must be able to be conscious of the forms of

jud.^ment, psychologically, that is, we must have been able

to discover them, and be conscious of

ou-r

logical thinking

by means of mental activities.

The mental activities of

the phenomenal self must therefore be separated clearly

from events in the external v;orld,

Kant*s distinction

153

between inner and outer sense, v/hich rests, at bottom, on

his fundamental distinction betxveen space and time, is,
v/e

maintain, the chief means whereby he accomplishes this important separation,

Ue find, therefore, that it is very misleadins to

speak of Kant's view as a "metaphysics of experience". The

expression is correct enough as far as it goes, in that it

does imply what Kant's chief metaphysical interest
vjas.

But it also seems to sur^est an omission of the independent,

idealistic source of the principles constitutive of experience.

For Kant's philosophy is not primarily a metaphysics

The fun-

of experience, but rather a metaphysics of logic.

damental metaphysical fact, for Kant, is the primacy of

.judgment .

The second basic reason for the importance of time

in Kant's philosophy, and for the consequent derivative role

of space, relates to the effectiveness of Kant's answer to

Hume.

Since we have already discussed

tliis at

length, we

need not lin^^er further over it here,

we must take the op-

portunity at this point, hov/ever, to remark that had Kant

not made a fundamental distinction between space and time,-

had he obliterated all distinction between space and time,he would never have been ab3ie bo rebut Hixme's criticism at

all.

For KwaG's main argument can be stated equa.lly in

'

154

terms of diverse re^'ions of space-time, as it can be in terms of the difference between the future and the past, as long
as it is assumed, paradoxically enough, that an identical 169 self can be at different regions of space-tiiae. Kant's an-

swer depends on the view that the categories condition the

temporal successiveness of the transcendental synthesis of

imacination.
But if the universe is laid out completely

in an eternal four-dimensional manifold, there can be no

question of successiveness, and hence no synthesis at all,

at least in Kant's sense of the term.

In addition, accor-

ding to a manifold theory in v/hich there is no distinction

between space and time, individuals become a matter of convention.

It is hard to see how there could be any but an

arbitary

order in nature, if things can "nove back and

forth in time", depending on how we choose to regard them.

There ceri^ainly could be no necessary logical order,
A

development of Kant's view in this direction v;ould seem to

be extremely un-Kaiitian,
On the other hand, the physical

theory of space-time seems to retain an intrinsic difference

between space and tine, a distinction which is involved in
the reference to the observation of motion.
Time is dis-

tinguished from space by its successiveness to the conscious

mind.
And this, of course, is the very basis of Kant's
In this respect at least, the physical theory

distinction.

of space-time does not seem to be incompatible v/ith Kant's

155 views.
In so lar as the physical theory of space-tiffie pre-

iupposes a metaphysics similar to possibility A, however,

it \}ovld be opposed to Kantfi metaphysics of space and time,

out this opposition mi^ht possibly be removed by a proper

distinction between measured time and primordial time, i,e.

becoming.
Kant's metaphysics is, if nothing else, a unified
and amazingly consistent body of doctrine,
Uxs view of the

transcendental conditions of hiiman experience, his recognition of the

I'ole

of lo.^ic

and apperception, necessitate a

distinction between space and time such that the subject,

qua phenonenal self, is distinguished from the material world
.nd

objects in space, in that the self is in time alone,

'./e

can express this by saying that, for Kant, part of the dis-

tinction between subject and object consists in the fact

that the inner activities of the self are in time alone.
Kant makes use of this point and the consequent flux of

inner sense in his proof of the emptiness of rational

:'sycholOi3y.

Thus, in Kant's viev;, human mental activities

X'/hich

have a very close association with time, a topic

we

must now discuss.

CHAPTEH FOUR
i3::;RGS0K'S

THEORY OF SPACE

/iMD

TIKE MID ITS

RELATION TO THAT OF KANT

1.

Time and Inner Experience

Having clarified sufficiently for our purr>ose the

role of space and time in Kant's philosophy, we are now in
a position to introduce Bergaon's views, and to compare them

with those of Kant.

We have already

seei:

that Kant's vievvs

necessarily involve the point that the self is in time only.

Me must now explore the significance of this intimate rela-

tion between the self and tine,

V/e

shall present evidence

that Bergson's philosophy parallels that of Kant in several

important respects, namely, in holding that the inner

experience of the self allows a direct apprehension of the

realit]''

of time and change, that this inner experience is

not explicable in terms of science apr^ropriate to outer

sense, that the concept of the self is empty except in so

far as it is conceived as a logical

fiaactioqi,

and that in

several essential ways Bergson's distinction between inner

and outer sense resembles that of Kant,
Kant and Bergson a,^ree on a point of fimdamental

importance, namely, that the empirical self is characterized

by a perpetual

fliix.

In so far as we are conscious of

vje

ourselves and our inner state,

156

are conscious of a

157

continuous chan^je.

For Ilant, when

v;e

empirically attend
(in the

to that area of appearances called inner sense,

narrow meaning), our primary intuition is of this change.

Consciousness of self according- to determinations of our state in inner perception is merely empirical and alv;ays changing. No fixed and abiding self can present itself in this flux of inner appearances. I70

Bergson makes the same point in more concrete,

figurative language:
The existence of v;hich we are most assured and which v;e know best is unquestionably our own, for of every other object we have notions x^hichraay be considered external and superficial, whereas, of ourselves our perception is internal and profound. I'lHiat, then, do we find?. . . I find, first of all, that I pass from state to state. I an warm- or cold, I am merry or sad, I work or I do nothing, I look at what is around me or Sensations, feelings, I think of something else. such are the changes into which volitions, ideas my existence is divided and v;hich color it in turns, But this is not I chan,:e, then, v/ithout ceasing. saying enough. Change is far more radical than we are at first inclined to suppose. 171

Bergson, of course, reveals here the importance he

attaches to this intuition of change, in sioggesting that

it is more radical than one might suppose.

While Kant

does not attach the same kind of importance to this matter that Bergson does, it is nevertheless an indispensable

element in his philosophy, opening up as it were, the

avenue through which Kant approaches time .

For, as we have

already emphasized, time is the form of inner sensibility

in the narrow meaning as v/ell as in the wide meaning.

15S
In addition, we must draw attention
liere

to zhe similarity

of Ber^son^s point that our perception of ourselves is in-

ternal and profoimd, while our perception of objects is

external and superficial, to some of Kants statements.

Kant tells us that "All that we know in matter is merely

relations

(v/hat

we call the inner determinations of it are

172

inward only in a comparative sense)".

Since Kant also

remarks that "we know of no determinations v/hich are abso173

lutely inner except those (given) through our inner sense", we may conclude that, for him, the primary meaning of the
terra "inner" is

derived from inner perception of ourselves.

This point, in fact, constitutes one of his main v/eapons

in. his attack on Leibniz.
life

do not knovj the inner nature

But

of anything except that of our phenomenal selves.

Kant quite clearly distinguishes this inner nature from

the determinations of external objects.
But that which is inner in the state of a substance cannot consist in place, shape, contact or motion' (these determinations being all outer relations), and we can therefore assign to substance no inner state save that through which we ourselves inwardly determine our sense, namely, the state of representations . 174 One of the first questions v;hich occurs concerning

this self -consciousness or perception of the self, as re-

vealing a constant state of flux, is as to vjhat meaning the

term "self" can have in this context.

Kant distinguishes

between inner sense or empirical apperception, which is in

159

constant flux, and the facxiltv of apperception, and tells

us that "we intuit oujrselves only as v;e are inwardly 175 affected". The problem of how that v/hich is in constant

flux can be determined in any way, we shall consider later.

For the present, we note only that ones inability to find
a suitable terra, which at once indicates something; deter-

minate enough to be an object of knoii/ledge, and at the same

time does not destroy the primary fact of the flux, is

revealing,

iiThat

is uncovered in this inner intuition is,

so to speak, so slippery and transitory, th^t to apply

substantive terras to it seems to be a mistake.

At any

rate the ansv/ers which Kant and Bergson give to this ques-

tion are for-thright and strikingly similar.

There can be

little meaninc for them in the notion of a permanent ego

or self in the flux of inner experience.
But, as our attention has distinguished them and separated them artificially, it is obliged next to reionite them by an artifical bond. It imagines, therefore, a formless ego , indifferent and unchangeable, on v;hich it tlireads the psychic states v-zhich Instead -of a it has set up as independent entities. flux of fleeting shades nerging into each other, it perceives distinct, and so to speak, solid colors set side by side like the beads of a necklace; it must perforce then suppose a tliread, also itself solid to hold the beads together. But if this colorless substratum is perpetually colored by that v/hich colors it, it is for us, in its indeterrainateness, as if it did not exist. . , 176
Vife

see here, in Bergson 's words, something of the

Kantian doctrine of the indetenninateness of things-in-

160

themselves.
of the soul

Indeed, for Kant, the emptiness of the concept

is one of the charges he brines a^iainst rational

pS3rcholocy, and the effectiveness of this charge rests on

Kant's viev/ that inner experience reveals a constant flux.

For in what we entitle 'soul', everything is in continual flux and there is nothing abiding except (if v;e must so express ourselves) the I which is simple solely because its representation has no content. 177
It would be incorrect to call this inner intuition

itself time, for as Kant often remarks, "time itself cannot

176^

be perceived".

Nevertheless, Kant holds that the nat\ire

of time, the continuity of v/hich "is ordinarily desir^ated 179 by the tern flov/ing or f leaving away", is the determinate

form "in v;hich alone the intuition of inner states is posIgO sible". While ive cannot be said to perceive time, or to
have a knov;ledge of it by means of a concept, ("Time is

not a disciirsive, or

vjhat is

ISl called a general concept"),

we are nevertheless conscious of it as the peculiar form

in v;hich this inner intuition takes place.

As such, we

can never be conscious of time as an object, especially as

a thing-in-itself .

Ue have the sense of time, hov;ever,

and recognize it as absolutely different from the sense

of space, althoTOgh the way in v;hich
v;e

can be said to have

these must be carefully specified to avoid misunderstanding

of Kant's doctrine. We have a sense of time and a sense of

space, not as objects, or even as sensations; what we

161
actiially sense are appearances, so that
v/e

have the sense

of time and the sense of space, only in so far as we recog-

nize certain kinds of appearances as differently related,

i.e., as in time and as in space.

We have no sense of these

relations apart from the appearances which they relate, yet

we can in one way be said to have the sense of time and the
sense of space, in that we do recognise appearances as dif-

ferently related, sone in time alone, ("the empirical object, which is called.
.

.an inner object if it is represen) .

ted only in its time-relations "

others in both space and

time, "the conditions of sensible intuition, which carry

1(53

V7ith

them their oxm differences".

Granted that the self as it appears to inner sense,

in the narrow meaning, is not time itself, but is

jto

time,

and taking Kant's ivords at their face value, that this self
is in constant flux, it is navi necessary to enquire into

the character of the

fli;ix

itself, and to determine

hov;

Kant

thought of it.

It will be helpful in this regard to intro;and

duce his views on psycholofry

what kind of a science it

may be said to be.

There are, for Kant, two species of sense objects,

according to
divided.

x^?hich

the whole of nature as phenomena is

162

Now Natxire, in this sense of the word, has two main divisions, in -accordance with the main distinction of our sensibility, one of which coi.iprises the objects of the outer , the other the object of the inner -thus rendering possible a two-fold doctrine of 'Mature, the UOCTRIIJii OF JODI and the DOCTRIIiS C? 30UL, the first dealinj with extended ^ and the second v;ith thinking. Nature, 1^5
;

Here Kant is referring to inner sense in the nar-

row meaning, a point which he makes clear in his subsequent

proof that all matter as such is lifeless.

know no other internal principle of and no , other internal activity v/hatever but thought . witK that -which depenas upon it, feeling of' pleasure and pain, and inpulse or x;ill. But these groiinds of determination and action in no wise belong to the presentations of the external sense. 166
Kov; viQ

a substance to change its state but desire

It is, then,

^ propos to enquire

xvhether the objects

of inner sense can be objects for scientific study in the

same v;ay that external or spatially related objects can be.

Such a science would be a physiology of inner sense.

If our knowledge of thinking boincs in general, by means of pure reason, were based on more than the cogito , if v;e likewise made use of observations concerning the play of our thoughts and the natural lav;s of the thinking self to be derived from these thoi;^hts, there would arise an empirical psychology, which would be a kind of physiology of inner sense, capable perhaps of explaining the appearcinces of inner sense, but never of revealing such properties as do not in any way belong to possible experience. . 1^7
-

Kants hesitancy about affinning such a science is

apparent in this statement, for the sufficient reason that

he holds that there can be no synthetic a priori loiov/ledge

163

of the appearances of imier sense.

If we compare the doctrine of the soul as the physiology of inner sense, v/itli the'^octrine of the body as a physiology of the object of the outor senses, vje find that while in both riuch can be learnt empirically, there is yet this notable difference. In the latter science nuch that is a priori can be synthetically known from the ir.ere concept of an extended impenetrable being, but in the former nothing ivhatsoever that is a priori can be knovm synthetically from the concept of a tjiinking being. iSg
The implications of this statement are far-reaching.

Concepts employed in physics are not blind because an intuition i3 given for them, because in fact appearances are

determined a priori in accordance with the categories.

on the other hand, we have inner intuitions to which no

If

body of a priori synthetic knov/ledge corresponds, this can

only be because such appearances have not been determined
in accordance with the categories.

The difficulty seems to

be due to Kant's doctrine that inner sense is a perpetual

flux, with nothing permanent in it, and this in turn is

because the appearances of inner sense are in time, and

only in
tjxie
.

For space alone is determined as permanent,

v;hile time, and therefore everything that is in inner

sense is in constant flux.

1^9

Alongside this we must place Eant's statement that

if everything in the world is in a flux, and nothing is
-190 a doctrine of

permanent, substances are inadmissible,

which he takes advantage in proving the emptiness of the

concept of the soul in rational psychology, v/here he

164
re-afflrras the flux of inner sense.

For v;e are irnable from our o^.vn consciousness to decide whether, as souls, we are permanent or not. .i'or since the only permanent appearance which v*e encounter in the soul is the representation *I* that accomijanies and connects them all, we are unable to prove that this 'I', a mere thought, may not be IXj. the same state of flux as the other thoughts. . .191 Kant unhes'titingly draws the logical conclusions

from this doctrine, in other parts of the Critique .

Since
192

time itself does not change, but only appearances in time,

"Only through the permanent does existence in different

parts of the time-series acquire a magnitude which can be
193 In other words, that which is in time,

entitled duration".

and only in time, cannot be an object of scientific knovjledgej all alteration, if it is to be perceived as altera-

tion, presupposes something permanent in intuition, and

"in inner sense no permanent intuition is to be met with". But it is an even more notev;orthy fact, that in order to imderstand the possibility of things in conformity viith the categories, and' so to demonstrate the objective reality of the latter, we need, not merely intuitions, but intuitions that are in all cases outer intuitions. Ivheni ^or instance, v/e'take the pm^e concepts of relation , we find, firstly, that in order to obtain something permanent in -intuition corresponding to the concept of substance , we require 195 an intuition in space (of matterTI
It v^ould be quite easy at this point to take the

view that we are faced here with another of supposedly

many cases in which Kant contradicts himself, maintaining

on the one hand that we are somehow sufficiently aware of

165

the nature of time to distin-xulsh spatially related objects

from the flux of tine in inner sense, and on the other hand
that time itself does not chani^e, but only appearances in
it.

In other vjords, time flows, and it does not flov/.

It

would be quite impossible to convict Kant of contradiction

here, however, for as has been pointed out above, he does

not say that

v/e

ever have an intuition of time only, as an

object or as a sensation; time is only the vmy in which

some appearances are intuited.

Although we are acutely

aware of the passage of time, through the constant flux

of our inner sense,

vje

never perceive time itself flowing.

It is one thing to say that time itself flovv's, and another

to say, as Kant seems to be doing, that anything, (v/hether

we call it object or appearance,

v;e

shall be hitting wide

of the mark) that is in time alone will be in constant

flux.
To say that the essence of time is that it flov;s,
thin.gjs

and to say that the essence of time is such that

in it flou, or change, is to say two different things,

V/hen Kant says

that the continuity of time is a flowing

or a flowing away, he is simply referring to the fact that there are no gaps or bi-eaks in it.
Time itself

cannot flow, for it is not a reality in itself. It is

transcendentally ideal, but also an empirically real way

of intuiting anything that can be given to the senses.

Inner sense gives us a heightened av?areness of this

vray

166
'A'e

have of intuitin,'^, because in inner sense

v;e

intuit in

this way and only in this way .

Me do not at the same time

have intuitions of outer sense, which involve another, quite

different, way of intuiting.

Thus, throu;h inn.er sense we

become aware of time as a way we have of intuiting, different

from space, which is another

v;ay.

In this manner, the im-

mediate apprehension of time is of fxindamental importance

to the critical philosophy, revealing as it does one of the

transcendental elements of sensibility.

The existence of appearances which are solely in

time, and time alone, poses serious problems for Kant, however, ones vjhich he does no more than touch upon in a cur-

sory fashion.

For example, it is apparent that the kind

of change involved in such purely temporal succession as is exemplified in inner sense will be radically different

from the kind of change involved in appearances v;hich are

both spatially and temporally related.

Strictly speaking,

such change cannot be cognized. It is not alteration of

something which abides, namely, substance.

Kant, himself,

clearly recognizes this.

Although both are appearances, the appearance to ou.ter sense has something fixed or abiding which supplies a substratiim as the basis of its transitory determinations and therefore a synthetic concept, namely, that -of space and of an appearance in space; v;hereas time , which is the sole form of our inner intuition, has nothing abiding, and therefore yields knoviledge only of the chanp:e of determinations, not of any object that can be thereby determined. I96
'

167
Wg can only conclude that Kant is using the term

appearance very loosely, in speaking of appearances of irmer

sense, since appearances are nor often spoken of by him as

deteminate appearances, the results of synthesis and determination in accordance with the categories.
The self,

of course, as an entity, plays no part in inner sense,

except as a "universal correlate of apperception and itself 197 a mere thougnt. , .a thing of undefined signification".

Personal identity is accounted for by the duration of the

hiuaan

body in space, as an object of outer sense capable

of being perceived by others. Thus the permanence of the-soiil, regarded merely as an object of inner sense, remains undemonstrated, and indeed Indeaonstrable . Its penaansnce diiring life is, of course, evident per se, since the thinking being (as man) is itself likev/ise an object of the outer senses. 19^

Thus apparently nothing is capable of being knoim

about the appearances in the perpetual flux of inner sense,

neither the ego or self, nor the appearances, nor the kind
of change which characterizes them.

Since

v/e

are directly

aware of them, the status of these things is not quite the

same as that of things-in -themselves; nevertheless, they
are equally as indeterminate as things-in-themselves.
The

things of v;hich

i';e

feel that we are most intimately aware

It is characteristic of Kant's

are not objects of knowledge.

honesty as a thinker that he does not boggle at this result

I6i

but unwaveringly points out the reasons v/hy, according to

iiis

principles, thiu uust be so, in his analysis of the

oi"

claims

psycholoey to be u science.

His most important

point is that:
.as iii every natural doctrine only so . much science proper is to be liiet ;/ith therein as there is co'^ition a priori a doctrine of nature can only contain so''niuch science proper as there is in it of applied mathematics. 199
,

iiant ,;:oes on

to point out tiiat raatheraatics is

aiid

"inapplicable to the phenomena of the intexnal sense

200 its laws", for in it all that mathematics could lay hold
Ox would be the laxv of permanence in the flovi of its in-

ternal changes.

Me are able to get as much knowledge

j;;et,

about the soul froin this, as we could

for example, in

geometry merely from the properties of a straight line.

In short, Kant implies, the knov;ledge
v;e

can get is very

little.

This point is scarcely less important than the

one which Lant raises, nanely, that
"tlie

neict

manifold of internal

observation is only separated in thought, but cannot be kept 201 It is to be separate and be connected again at pleasure".

noted that both of these shortcomings are due to that very

perpetuity of flux in inner sense, v/hich Kant has so often

himself pointed out, and which lies at the

veiry

base of

the whole critical philosophy, in that it allows Kant to

maintain that time is the form of inner sensibility.

169
But it carmot be said that Lant has fully apprecia-

ted the implications of his ovm criticism of psychology.

If there is no a priori knov.led^e of the phenomena uf in-

ner sense, they can hardly be phenomena at all in the usiial

Kantian sense.
Our means of acl.lyvinc
'.'Jiov/ledge

of pheno-

mena can alvjays be traced back to the a priori determination of phenomena in accordance v^ith the categories.
vie

If

cannot get knowledge in a certain sphere of appearance,

it can only be because these appearances have somehow

arisen vjithout being subject to determination through the

categories under the transcendental unity of apperception.
In that case they are not objects of possible experience,
in the sense of empirical knowledge,
'202
/uid

since the cate-

gories provide the ground of all determination of the

existence of appearances,
the appearances of inner sense

cannot even exist as objects of laiov/ledge connected by

definite rules to other objects of knowledge.

But since

these limer appearances most manifestly do exist, a point

which Kant never denies, their existence must be radically

different from appearances in the more usual sense of

appearances subject to determination in accordance with

the categories,
liow

the existence of these appearances is intimately

bound up with tine, another point which Kant never fails to

stress.

If we did not have an awareness of the difference

170

between the way we intuit our inner states, and the way we
intuit spatially extended objc^cts, there
^^rould

be no poss-

ibility of distin.^ui shins between forms of sensibility at

all.

For if space and tine are

forr;is

of sensibility, and

we are not aware of the difference between one mode of

sensible intuition and the other,
vie

could not distinr,uish

There

between spatial and temporal determinations at all.

could, in short, be no transcendental aesthetic as applied

to human faculties.

Thus the immediate apprehension of

time as a fomi of sensibility is essential to Kant's

philosophy,
Kant has offered a solution to the problem of the

fundamental distinction

betv.'een

space and time in terms

of forms of sensibility, which involves at the very least

a recognition of the intimate relation between time and

the flux of inner experience, and hence of the relevance

of the psychology of inner states to the determination of the nature of time,

Kant's view that psychology can

never be any-thing more than an approximately systematic,

historical description of inner sense, rather than
a science, points to the peculiar role that time plays

in Kant's philosophy.

Not only are those appearances

which arc in time alone incapable of being cognized,

but time itself cannot be represented in thought.
aware of time as a form of sensibility, but
vje

We are

cannot

171

obtain a conception of it.

Hven time itself we cannot represent, save in so far as we attend, in the drai\?inp: of (v.'hich has to serve as the outer figurative representation of tir^e), merely to the act of the synthesis of the nianifold whereby v^e successively determine inner sense, and in so doing attend to the succession of this determination in inner sense. 203
a strai^^ht line

Kant apparently holds that this figurative repre-

sentation is a legitimate one, but he passes over in silence

why this, according to his principles, must be so, just as

he ignores for the most part the striking difference betVv'een

substance as it figures in outer sense, and substance as

a possible concept applicable to the flux of inner sense.

The essential characteristic of substance in outer sense

is that it is capable neither of increase nor diminution;

hence the application of the concept of quantity to it is

possible only through the aggregate pf parts outside one

another.
But only in space can we have parts outside one

another, and if we attempt to extend the concept of substance to a sphere which is not in space but only in time, the

concept of quantity cannot apply to substance in this

v;ay.

Kant attempts to preserve some meaning to the concept of

substance in inner sense, by applying his conception of intensive magnitude.

That, on the contrary which is considered as object of the internal sense may have a quantity as substance not consisting of parts outside one another, whose parts are therefore not substances, whose origination or annihilation therefore need not be the origination or annihilation of a substance. 2(U

172

This kind of substance, If indeed, it could be

called a substance at all, has nothing in comnon with that

of outer sense, a point which Kant makes clear when he

says of the ego, as object of consciousness, that it is a

"thing of undefined signification, namely, the subject of

all predicates without any condition distinguishing this presentation of the subject from a something generally,
in short, substance, of which no conception of xvhat it is 205 (is conveyed) through this expression". Altogether,

Kants attempt to grapple with this problem, namely, of

what meaning the categories can have for the appearances

of inner sense, is highly ambiguous.
On the one hand, Kant

does not want to admit that the soul is a substance, its

liability to destruction seeming to endanger the principle

of the permanence of substance.
Yet, at the same time, he

seems to want to be able to apply the concept of quantity

to inner states, in respect of intensive magnitude.

Thus consciousness, in other words, the clearness of the presentations of my soul. . .has a degree that may be greater or smaller, xvithout to this end any substance requiring to arise or be annihilated, 206

Those sensations of outer sense may legitimately admit of intensive magnitude, according to Kant, since

they are only states of substance that change, while the

substance remains the same.
Their arising and dying away
The appearances

cannot affect the pennanence of substance.

173

of inner sense, however, are in constant fl\ix; there is no

underlying permanence; they are not states of something

else that never changes.

Their arising and perishing

chanj:-e

through various degrees is not the mere state of

of

something that endures; it is a genuine arising and perishing, not explicable at all in terms of substance, in
short, the very passage of time itself, as revealed in

inner sense.
Kant has bought the means of distinguishing between

space and time at a rather expensive price, in terms of

the generality of his theory of knowledge.

Tine is, indeed,

clearly distinguished from space in that appearances in it

are beyond the pale of co^^jnition, but Kant has accepted a

point of fundamental importance in 3ergsons theory of time, namely, that we are aware of it through inner ex-

perience.

For Kant, we are aware of time merely as the

form in which our inner experience is solely conditioned,

but not as an all-encompassing reality. Yet in respect

of the all-pervasiveness of time, Kant's views are not

entirely dissimilar to those of Bergson, for, as

v/e

have

seen, time, for Kant, is the medium by means of which the

categories determine objects and bring about order throughout the phenoiaenal vjorld.
In addition, the appearances in

time alone bear a strong similarity to things-in-theraselves,

(the ultimately real which we know only as appearance).

174
in that both are indeterminate and unknowable by means of

concepts.

Before taking up these points, hov;ever, we must

examine Bergson's position on the immediate apprehension

of time.

Bergson also distinguishes between inner sense and

outer sense, although, for him, the distinction is less sharp than it is for Kant.
The spatially extended is, in

Bergson s view, something quite different, indeed, radically different from what is revealed by the deepest in-

trospection.

But the transition from what is absolutely

outer to what is absolutely inner in intuition is blurred,

for Bergson; the transition involves intermediate states,

unlike Kants sharp dichotomy.

When, xvith the inner. regard of my consciousness, examine my person in its passivity, like some superficial encrustraent , first I perceive all the perceptions vjhich come to it from the material world. These perceptions are clear-cut, distinct, juxtaposed or mutually juxtaposablej they seek to group themselves into objects. Kext I perceive nemories more or less adherent to these perceptions and which serve to interpret them; these memories are, so to speak, as if detached from the depth of my person and dravm to the periphery by perceptions reserabling them; they are fastened on me without being absolutely myself. And finally, I become aware of tendencies, raotol? habits, a crowd of virtual actions more or less solidly bound to those perceptions and these memories. . .But if I pull myself from the periphery tov;ards the centre ^ if I seek deep dovm within me v/hat is most unifonaly, the most constantly and durably myself, I find something altogether different, 'what I find beneath these clear-cut crystals and this superficial congelation is a continuity of flow comparable to no other flowing I have ever seen. 207
I

175

Ber^son announces the

ence as does Kant.

sarae

fact about inner experi-

His exposition differs only in that

while Kant was not particularly at pains to communicate

the exact character of this internal flux, speaking of it
in terms of phenomena and appearances,
v.'lach

strictly

speaking, as determinate entities cannot figure in the

flux at all, Ber^son exerts all his literary pov/ers to

try to make us see what sort of thing this inner flow
is.

His works are replete with metaphorical devices and

figurative language quite consciously used in an attempt

to express the quality of this inner fliox.

This technique

was necessitated by the kind of philosophy Bergson desired

to expound, rather than by a mere personal preference for

the niceties of literary style.

On the contrary, for Kant,

such a manner of exposition was out of the question, not

merely because it was foreign to his cast of Kind, but for

the reason that he wished to lay the groiond-work of philo-

sophy in an exact conceptual scheme vjhich wo;ild deserve the

name of science. Bergson 's view that "metatjhysics is,
'20a
.

the science which claims to dispense with symbols",

set

him on a path that Kant could not have followed without

abandoning his most cherished hopes in philosophy.

It is

not surprising, therefore, to find that Kant on the theo-

retical side of his philosophy says very little about the

character of the

l\xx.

of inner states.

Even though he made

176 the same point that Bergson does, namely, that inner ex-

perience yields an awareness of time, beyond the bare fact

of this, necessary for the exposition of the elements of

transcendental aesthetic, Kant could say nothing further.

The essential point in this is that Kant's theory of time

points in a direction in which Kant was imivilling to go.

This refusal to envisage further development is,
of coiorse, bound up with Kant's v;hole viei; of human know-

ledge as unable to extend beyond a possible experience

defined, in effect, by the scope of the applicability of

the categories, limited, in fact, by the kind of catego-

ries Kant deduced from the forms of judgment he expounded

as the basis of transcendental logic.

Yet Kant himself

raises certain problems which could only be answered by

taking this flux of inner states more seriously, and of

these one concerns the relation between inner sense in
the narrov; meaning and inner sense in the wide meaning,
or, briefly, the relation betv^een inner sense and outer

sense and the two forms of sensibility, time and space.

How is it possible that these two disparate kinds of sen-

sibility become combined in one unified experience, in

which the intimate relation between them as exhibited in

motion is evident?
One might put this in Kantian terras

and ask, "How is change in general possible?". It is to be

noted that this question concerns not motion alone, but

177

all change.

Kant has dealt with the concept of raotion as,

V.is

in his opinion, it is needed for natural science, in

Metaphysical Foundations of Natural Science , but his definitions presuppose that change is possible, (something,
of course, which no one will doubt, even though it may be

difficult to understand in terras of concepts how it is

possible],

ouch a question demands an answer showing not

merely how space and time are related to change but also
how psychologically . inner sense is related to outer sense.

This in turn involves a more exact explanation of how that

"train of ideas" which Stout maintained is absolutely in-

dispensable to the perception of time is related to the perception of space.

But the ansvjer to the psychological

question v;ill no doubt turn on the doctrine of what is re-

vealed in inner experience, in other

time itself.

\i;ords,

the nature of

At any rate, as to the relation between the

;Tive

two kinds of sensibility, Kant flatly refuses to

opinion.

an

The iuujh discussed question of the coiEinuiiion between the thinking and the extended, if v;e leave aside all that is merely fictitious, coraes then simply to this; hov; in a thinking sub.ject outer intuition , namely, that o? space, with its filling-in of sbape and motion, is, possiule . And this is a question which 209 no man can possibly answer.

Kants refusal is, of course, again dictated by

his general theory of knowledge, but that he has seen the

question in its generality is indicated by the following

statement;

17fi

Consequently, the question is no longer of the communion of the soul v;ith other known substances of a different kind outside us, but only of the connection of the representations of inner sense v/ith the modifications of our outer sensibility as to hov/ these can be so connected v;ith each other according to settled laws that they exhibit the unity of a coherent experience. 210

VvTiether

Kant's position, in refusing to deal with

problems raised by his theory of space and time, is a

justifiable one, need not concern us here.
Our present

purpose is merely to point out that Kant's theory seems

to point in the direction of a further development which

could deal with such questions,- one v;hich, however,

v.'ould

have been impossible for Kant, but which could have been

carried out by a thinker v/illing to follow the implications of it beyond the limiting scope of Kant's categorical
scheme.

Without undue anticipation, it can be said here

that Ber^-cson's theory deals with just those questions which Kant would not attempt to answer, deals with them, moreover,
in a way which stems from a basic position similar in many

ways to Kant's.

He agrees with Kant, in effect, that the

understanding does influence inner sense, producing definite

determinations in it of objects and spatial representations.
a

But for him this function of the understanding hasAmeaning

profoundly different from that involved in Kant's philosophy.

Bergson gives us a different interpretation of the spatial

representation of time, one v/hich allov;s him to go beyond

179 Kant in his theory of the relationship between space and

time.

For example, Bergson attempts to explain the rela-

tion between inner and outer sense; he attempts to tell

us how it is that we have spatial perceptions and
vjliat

is

the relation of these to time.

^Ind

this involves his

placing a much greater emphasis on the significance of intuition for metaphysics, a development which Kant would

probably have rejected as schwgrmerei . but one which,

nevertheless, is a development from Kants own assumptions

concerning space and time.

For Bergson, inner experience reveals the ongoing flow of duration itself, a terra v/hich Bergson uses to dis-

tinguish his intuition of time from the "spatialized" time,

conceived by the understanding.
Time, for Bergson, is not

a conceptualized entity, not a thing-in-itself , conceived

in general terms; it is the reality of concretely felt

change, and this change can best be felt v/hen the intellect does not interfere to separate the continuous flux

into separate states.

Piire duration is the form which the succession of our conscious states assumes v;hen our ego lets itself live , when it refrains from separating its present state from its former states. 211

In other words, the succession of conscious states

cannot be apprehended in terras of concepts, a point which

Kant also mentions in saying that "the manifold of inter212 nal observation is only separated in thought".

180
Just as, for Kant, there is no underlying permanence
in inner sense, so also for Bergson the reality uncovered
in inner experience is not a change of the state of some

underlying substratum.
There are changes, but there are imdemeath the change no things which change; change has no need of a support. 213
I/ith Kant,

the "parts" of that which changes in

consciousness are not parts outside of one another, such

as we find in spatially related objects of outer sense.

This is also 3ergsons view. Thus I said that several conscious states are organized into a whole, pemieate one another, gradually gain a richer content, and might thus give anyone ignorant of space the feeling of pure duration; but the very use of the word "several" shows that I had already isolated these states, externalized them in relation to one another, and, in a v/ord, set them side by side; thus, by the very language which I was compelled to use, I betrayed the deeply ingrained 214 habit of setting out time in space.
The above quotation also indicates the similarity

to Kant, v/hich has been mentioned above, namely, that in

order to obtain a representation of time, of the form of

our inner experience,

in spatial terms.

xie

must resort to representing it

Another similarity is that both Kant

and Bergson have recourse to the notion of intensive magnitudes in referring to the succession of psychic states,

although Bergson doubts whether this concept is legitimate.

Pure duration, that which consci-msness perceives, must thus be 'reckoned araong the so-called intensive magnitudes, if intensities can be called magnitudes: strictly speaking, hoi;ever, it is -not a quantity, and as soon as we try to measure it, we unwittingly replace it by space. 215
'

It is clear that Kant and Bergson h^ve the same

thing in mind in incorporating into their philosophies the

doctrine of an intimate association between time and inner

experience,- they have the same thin^ in mind in the sense

that they have the same phenomenc.i, (in the non-Kantian
sense) in mind.
It should oe added, hov/ever, that Bergson

develops the implications of this phenomenon further than

does Kant.
Thus, for Bergson, duration as revealed in

inner experience is an all-encompassing, basic reality. And the reasons for this further development, the sources,

that is, from which it springs as an answer to perplexing

problems, and the next step in their solution, given the

basic premises, is to be found in the relation of the

understanding to space, and the doctrine of the cognitive

representation of time, topics to be pursued in subsequent

sections.

1S2

2,

The Cognitive Representation of Tlrae

In one way, Ber^son and Hegel represent an interest-

ing contrast of development from Kant, v;hich

vje

may mention

in order to bring out the character of Bergsons development.

They both take a point of deimrture from Kants doctrine of

judgment, and both find in this doctrine an inadequacy

amounting to artificiality.

The traditional formal logic,

a cornerstone of Kants system, and for him, a "closed and '216 Hegel calls "the lifeless completed body of doctrine", 217 And Bergson, as v;e shall see, affirms bones of a skeleton".

the inability of the intellect to coue to grips vjith the

primordial flow of time.

Hegel s method of attacking this

problem

v;as

to invent a new logic, and a "concrete" doctrine

of judgment.

Bergson s method on the other hand is to ac-

cept the inadequacy of the intellect and to push the oppo-

sition between the intellect and concrete becoming to its

very limit.

Hegel does not repudiate the importance of Bergson not only rejects judgreality, but allows the signi-

judgment, but develops it.

ment as a method of

Icn owing

ficance of judgment itself to recede, and tries to place

the main emphasis of intellectucal thinking on the spatial

concept .

In developing a position, the seeds of which are

implicit in Kants theory of tine and inner sense, Bergson

finds it necessary to take Kants doctrine of judgment less

133

seriously, and to make of it a natter in which static concepts rather than active judgments are emphasized.

Instead

of accepting Kants doctrine of judgment and its applica-

bility to time, Bergson, in effect, denies it and makes it

far more artificial in its relation to the reality of be-

coming than it was in the philosophy of

Ivant.

Ue have

nov/

to examine Bergsons reasons for doing this.

V/e

have seen that it is essential to Kant*s position

that the categories should apply directly to time itself.

This necessarily involves the point that the cognitive re-

presentation of time should be legitimate.

In this regard

we find Kant, with typical candour, accepting on the one hand that the flux of inner sense is bej^ond the
.

\^

of

cognition, and on the other that time may be given a repre-

sentation in terras of concepts applicable to outer sense.

'Time is nothing but the form of inner sense, that is, of the intuition of ourselves and of our inner state. .'.And just because this inner intuition yields no shape, we endeavour to make up for this uant b3'analogies, v/e represent the time-sequence by a line progressing to infinity, in which the Pianifold constitutes a series of one dimension only; and xie reason from the properties of this line 'to all the properties of tine, i7ith this one exception, that while the parts of the line are simultaneous the parts of tirr.e are always succe-jsive. From this fact also, that all the relations' of tine allow of being expressed in an outer intuition, it is evident that the representation is 21g itself an intuition.
liniat

is involved in this spatial representation of

time?

First of all, there is the conception of both space

1S4

and time as homogeneous, made up of units, as it were, \;hich

are all the sane.
Kant, indeed, rejects the view that there

are ultimate simples, but nevertheless in speaking of space

and time as quanta continua he reveals the element of homo-

geneity in his coiiception of them.

3pacc and time are quanta continua . because no part of them can be given save as enclosed betvieen limits (points or instants), and therefore only in such fashion that this part iG itself again a space or a time. Space therefore consists solely of spaces, time solely of times. 219

The very fact that be conceives of points and instants

as limits implies that every determinate part of space is a

part of the same medium, space, and every determinate length of time is a part of time.
The homogeneity of these media

is necessary to Kant, othervjise he could never speak of the

"manifold (and) homogeneous in intuition in general".

220 It

is necessary, according to Kant, in order to obtain a con-

ception of space and time to construct these concepts,

and this, of course, involves intuition.
The only intuition that is given a priori is that of the mere form of appearances, spaaa and time. A concept of space and timej as quanta, can be exhibited a priori in intuition, that is, constructed, either"*in respect of quality (figure) of the quanta, or through number in their quantity only (the mere 221 synthesis of the homogeneous manifold).
'

But the manner in which Kant describes

hovtf

the

elementary conceptions of space and time are formed reveals

that in the formation of these concepts the successiveness

1^5

of time, itself, is cancelled, and left out.

Synthesis of

the homogeneous manifold, even thovigh it is always successive, for Kant, involves holding together in thought all

the parts which are successively synthesized.

V/hen I seek to draw a line in thought, or to think of the time from one noon to another, or even to represent to myself some particular nuiioer, obviously the various manifold representations that are involved must be apprehended by me iu thouj;ht one after another. But if I were always to drop out of thour.ht the preceding r'epre sent at ions , (the first parts of t'heline , the anteceaent parts of tne time eriod, or the units' i'n the order represented ) .and i d not reproduce them while'Tovancing t o those that follow t a complete representation v/ould never be ob tained none of the aoove-nentioned tI:on3ht's, not even the purest and most elementary rei esentations 222 of space and time could arise,
^

Bergson's analysis of the concept of number is

strikingly similar to that of Kant.

volves synthesis.

?or both, niomber in-

"Number", Bergson tells us, "may be de223

fined in general as a collection of uiiits, or, speaking

more exactly, as the synthesis of the one and the many".

Bergson also draws attention to the point Kant has made,
namely, that all the parts of the successive synthesis must be held together, as it were, simultaneously in
thou^^ht

For if we picture to ourselves each of the sheep in the flock in succession and separately, we shall have to do v/ith laore than a sin^^le sheep. In ornever der that the number shoiild go on increasing in proportion as v;e advance, v/e must retain the successive images and set them alongside of each of the new units

1^6

which we picture to ourselves: now it is in space that such a juxtaposition takes place and not in pure duration. In fact, it will be easily granted that counting material objects means thinking all these objects together, thereby leaving them in space. 221+
It may be thoui];ht that here Bergson has passed

rather hastily from the point that the units cannot be

dropped out of thouf^ht to the point that they must therefore be left in space.
But the passage from the one to

the other is not performed without argument, and in this

respect agaiai Bergson follov/s Kant,

For Bergson agrees

with Kant that mathematical concepts require an intuition.

Abstract numbers, as symbols, are not even thought, but
are merely expressions of number useful for reckoning.
It will be seen that we began by imagining e,g, a row of- balls, that these balls afterwards became points, and, finally, this image itself disappeared, leaving behind it, as we say, nothing but abstract number. But at this very moment we ceased to have an image or even an idea of it; we kept only the symbol which is necessary for reckoning and which is the conventional vjay of expressing number. For vre can confidently assert that 12 is half of 24 without thinking either the ntimber 12 or the number 24: indeed, as fai' as quick calculation is concerned, we have everything to gain by not doing so, 225

Kant constantly emphasizes the point that geometry

requires an intuition of space, and indeed cannot proceed

at all without this intuition.
It is Kants view that all

of mathematics requires an intuitive element; one cannot

arrive at the synthetic a priori propositions of mathematics by the mere analysis of concepts.

1^7 Pure mathematics, as synthetic cognition a priori , is only possible by referring to no ether objects than those of the senses. At the basis of their empirical intuition lies a pure intuition (of space and time) which is a priori . 226
'."e

have already seen that the schema of quantity

is number and that number, for Kant, always involves a

successive synthesis.

Even geometry, in dealing with lines,

planes, figures, etc. requires a successive synthesis.

I cannot represent to myself -a line, hoTf-zever small, v/ithout drawing it in thought, that is, generating from a point all its parts one after another. Only in this way can the intuition be obtained. 22?

The mathematics of space (geometry) is based upon this successive synthesis of the productive imagination in the generation of figures. 22^

Unfortunately, Kant here omits mentioning the

point that the synthesis must be held together in thought,

and that the units of the homogeneous cannot be dropped

out as they would be if the intuition v/ere solely successive.
V/e

cannot help wondering v/hat the character of this

product of Si^-nthesis is, and whether it would be correct

to call it spatial rather than temporal.
There is some

evidence that Kant, himself, did not think of it as temporal.

He tells us that we cannot conceive altei^ation

without having recourse to intuition,- "The intuition

required is the intuition of the movement of a point in 229 Kant seems to be saying that it is just the pespace",

culiar property of space that things in it can coexist .

which implies that in our conception of space/bhin^js are

'leld

together in the fashion he has said is necessary even

r.iost

for "the
time ,

elemoatary representations'' of space and

"space alone is determined as permanent, while

time, and therefore everything that is in inner sense is 230 in constant flux". Kant held that alteration is only con-

ceivable as alteration of substance, a permanent element

which persists through changes of state.

It s.^ems clear

that Kant thought of this permanent element as space itself, or rather the pemuixiynce of the real in space, which, of
course, iiaplies the permanence of space.
We have already

noted that the substance of outer sense is quite unlike

anything that could be said to be a substance in inner
sense.

Kant reaffirris this difference in a passage which

reveals not only that because of the perpetuity of the

flux of inner sense, we cannot obtain an imiuediate representation ox time, but also that we can obtain a mediate,
spatial representation 01 time because of the permanence
or "held-togetherness" of space. For in order that we may afteri;7ards make inner alterations likewise thinkable, vie must represent time (the form of inner sense) figuratively as a line, and the inner alteration through the drawing of this line (motion), and so in this manner by means of outer intuition make compretensible the successive existence of our-selves in different states. The reason of this is that .5.11 alteration, if it to be perceiv3d as alteration, presupposes something permanent in intuition, and that in inner sense no permanent intuition is to ?Z1 be met with.

1^9
Tho extent to
w'

ich intellectual operations preif at all, is a point v/hich we

suppose space, for

.'lant,

shall consiuer later,

ue wish here merely to draw atten-

tion to the point that Kant's figurative, spatial represenation of

tiiae

leaves out what he himself considers to

be the essential difference between space and time, namely,

the successiveness of time,

For althoUi:sh Kant holds that

we can represent time by a line, and x'eason from the properties of tnis line to the properties ox time, he also
says that the alteration of inner sense, and hence the

intrinsic character of time itself, can only be intuited

in motion*
iiere

a^ain, it is apparent that Kant is using

a term applicable to outer sense, iiameiy, alteration,

rather loosely with respect to inner sense,

inner sense
caiiiiot

rhe flux of

properly be called alteration, for as

Kant has vehemently asserted, there is nothing permanent

in inner sense to alter.
It
xiiay

seera

difficult to reconcile Kants affir-

mation of the permanence of spatial determinations, or of

permanence in and thi'ough spatial deteri/iinations, with his
view that all outer appearances are the result of a successive synthesis of the productive imagination.
It should

be noted, however, that there is nothing necessarily incon-

sistent in holding these

tv:o

positions together. The

190

successive s^-nthesis of the productive iraa,;ination, as a

transcendental condition of appearances, cannot, indeed,

be considered to be an event, ivhich has occiu'red, or does

repeatedly occur, a

viev;

which is associated with the opinion

that Kant is doing; a genetic psychology, a gross misinter-

pretation of him,
or outside of time.

Kor can it be considered to be nouKienal

Here
v.'e

agree v/ith Paton, who says:

. .1 do find it difficult to suppose. .that our minds are such that to them reality must appear, not only as a succession of changes in time, but as a succession of changes in time which must conform to causal law. There are two v/ays of avoiding this difficulty. One is to assert that the transcendental synthesis of \-jhich Kant speaks is a pre-conscious and noimenal synthesis which somehow constructs the whole physical world for us before we be^in to know it. For this view I can find no basis in Kant, nor does it seem to me to have the least plausibility as a metaphysical theory. 232

The transcendental synthesis must therefore be a

continuous process in time.

Only this, incidentally, v;ill

Iluiue.
IJox-/

suffice as a basis of a refutation of

it is not

inconsistent to suppose that in so far as time itself, as

a whole, is a

Permanent reality, i.e. an eternal process,

it should continue to synthesize quanta of space, which

would correspond to the permanence of time itself.

Accor-

dingly, the schema of substance, the permanence of the real

in time, would be a condition not derived from space, but

imposed on space, through the very continuity of the successiveness of the synthesis.
Thus the logical ground of the

191

principle of the permanence of substance would be expressed

in experience as the permanence of tisie itself, is quite in accord with what Kant says about it,
Porraanence, as the abiding correlate of all existence of appearances, of all change and all concomitance, Gxprosses tiiae in general, x'^'or change does not affect tiine itself, but only appearances in time. 233
/aid

this

In addition, there are

tv/o

reasons why this per-

menence of time must be expressed in outer sense rather

than in inner sense.
The first is that tiae itself can-

not be perceived.

Hence the permanence of tiue itself The second reason, and

cannot be imiediately presented.

perhaps the more fundamental one, is that the succession

or constant flux of appearances in time alone precludes

their :;iving a representation of the perTnanence of time

itself.

Obviously, the peruanence of

tiiae

itself cannot

be expressed or understood by means of appearances (re-

presentations) whic/i are in time alone, for the essential

of time is its successiveness. of all series.
It is the formal condition

Thus

-iihe

permanence of time as a condition

of the possibility of experience can only be knoim by means

of representations pertainin;- to another, different medium,

which is in itself not intrinsically successive.

But as the parts of space are co-ordinated with, not subordinated to, one another, one part is not the condition of the possibility of another; and unlike time, space does not in itself constitute a

192

-ovorthelesn the syiithesis of -ue maiiiJ?old partn of space, by means of which we apprehend space, is succesnlve, taking place in time ^^d coiitainiiig a series, 234
series:,
-

We can conclude, then, that the C0;:;nitive representation of space is a representation of space as a r.edium

in v/hich the parts are co-existent.

"For as its parts are

l.e

co-existent, it is an aggregate, not a series",

can

see also, that even though any magnitude whatsoever is the

result of a synthesis, the representation of that magnitude

must be as permanent, or at least as co-existing, heldtogether, in the representation.

Only tlirough the permanent does existence in different parts of the time-series acquire a magnitude which can be entitled duration. For in bare succession existence is always vanishing and recommencing, and never has the least magnitude, 236
Frorp

all this it is apparent that spatial concep-

tions do not and cannot represent the intrinsic nature of

time as
\ie

are aware of it in inner sense,

Kant has ad-

mitted this much in the imir.ediately foregoing quotation.

But it is just this difference betv;een space and time of

which Bergson takes advantage in his arguments against the

spatializing of time.
There are two kinds of multiplicity,

Bergson tells us, one kind applicable to space, and another

kind, radically different, applicable only to time.

193

Our final conclusion, therefore, is that there are tv;o kinds of multiplicity: that of material objects, to v;hich the conception of number is immediately appli-*cable; and the multiplicity of states of consciousness, which cannot be regarded as numerical v/ithout the help of some symbolical representation, in which a necessary element is space. 237
'

In admitting that in bare succession, there is ne-

ver the least magnitude, Kant has, in effect, admitted that

the concept of number applicable to outer sense in inapplicable to the states of inner sense.

For he makes clear

'233

that a necessary presupposition of coxmting is unity,

it is hard to see how there can be numerical unity where

and

there is not the least magnitude.

Both Kant and Bergson,

indeed, affirm that the ultimate ground of unity lies in

a pure act of the mind, and that the material on which this

pure act is brought to bear is a homogeneous manifold.

ViThat must first be givenwith a viei; to the a priori knovjledge of all objects is the manifold of pure intuition; the second factor involved is the synthesis of this nanifold by means of the imagination. But even this does not yield knov/ledge. The concepts which give unity to this pure synthesis, and which consist solely in the representation or this necessary synthetic unity, furnish the third requisite for the knowledge of an object; and they rest on the understanding. 239

For Bergson, the only kind of tinity is luiity intro-

duced by a simple act of the mind:

Nevertheless, by looking more closely into the matter, v^e shall see that all tmity is the unity of a simple act of the mind, and that, as this is an act of unification, there must be some multiplicity to 240 unify,
'

194
The unity wlilch the mind introduces implies, for
Berp'.son,

if not synthesis, at least the continuity of mul-

tiplicity.

You will never get out of an idea which you have formed anything which you have not put into it; and if the unity by means of which you make up your number is the unity of an act and not of an object, no effort of analysis will bring -out anything but unity pure and simple. No doubt, when you equate the number 3 to the sum of 1 1 1 nothing prevents you from regarding the units vjhich compose it as indivisible: but the reason is that you do not choose to make use of the multiplicity which is enclosed within each of these units. 241
Bergson agrees vjith Kant that
niuiibers

are reached

by a successive process of thought which reaches the final

result by going through a series of units.
The units see^

to be indivisible \7hile the process of synthesis is going

on, and indeed, because
vie

choose to regard them as sepaBut the

rate units for the purpose of synthesi25ing them.

final result is a unity in vjhich all discontinuity of the

synthesized xmits is merged in the unity of the synthesized

Again i if we form, the same nuraber v;ith halves,

with quarters, with any units whatever, these units, in so far as they serve to form the said number, will constitute elements v;hich'are provisionally indivisible j and it is alxvays by jerks, by sudden jumps, so to speak, that v;e' advance from one to the other. And the reason is that, in order to get a number. Me are compelled to fix our attention successively on each of the units of v/hich it is compounded. . ..Ind v;hen vje look at humber in its finished state, this union is an accomplished fact: the points 'have become lines, the divisions have been blotted out, the whole displays all the charac242 teristics of continuity.

195
In
speakixii';

of the synthesis iinplied in the addi-

tion of seven and five, Kant draws attention to the single

unified synthesis of each number itself.

But althout;h the proposition (7 f 5 12) is sjmthetic it is also only singular, oo far as we are here attending merely to the synthesis of the homogeneous (of units), that synthesis can take place only in one v;ay, although the employment of these numbers is general, 243

But the singular lonity of number is quite inappli-

cable to the perpetual flux of inner sense, for the units,

(if they can be called units) which flow throiigh inner sense

cannot be held together in a final sjmthesis.

They are,

as Kant says, dropped out of thoug:ht by virtue of the fact

that they are in time alone.

This, then, is the impasse

in which Kants theory of space and time remains, namely,

that while a recognition of the fact of becoming as exem-

plified in inner experience is necessary in order that Kant

should be able to speak of time as the form of inner sense,
it is quite impossible to see, in terms of Kants ovm ex-

planation of how the concepts of space,

are formed,
hov;

ti.

:,

;id

number,

the time of which we are aware as a form

of sensibility, can be represented in concepts al all.

This may be put in the form of a Bergsonian criticism in

the question: if time
terms,
hov;
_is

adequately represented in spatial

are space and time to be distinguished at all?

Tv/o,

As we mentioned above, in Chapter

Section One, it is

'

196

difficult to see how

tivo Luiique

things can have anything

But if space

in coraraon, and thereby be mutually understood.

and time are to be distinguished as

tv;o

different forms of

intuition, they must at least in so far as their intrinsic

properties are concerned, be quite unique.

And in this

case, for the reasons we have outlined, it v/ill not be pos-

sible to say that they can both be adequately represented

in concepts which have anything in common.

This, hov;ever,

is how, according to Kant, it must be; for in referring to

how we can have the phenomenal self as an object of consciousness, he illustrates this possibility by appealing
to the fact that time can only be represented in spatial

terns
Indeed, that this is how it must be, is easily shownif we admit that space is merely a pure form of by the fact that we the appearances of outer sense cannot obtain for ourselves a representation of time, which is not an object of outer intuition, except under the image of a line, vjhich tire draw, and that by this mode of depcting it alone could we knov/ of the singleness of its dimension; and similarly by the fact that for all inner perceptions we derive the determination of lengths of time or of points of time from the changes which are exhibited to us in outer things, and that the determinations of inner sense have therefore to be arran^^ed as appearances in time in precisely the same manner in which we arrange those of outer sense in space. 244

The illustration, however, does not improve Kants

argument, because both the phenomenal self and the repre-

senation of time have exactly the

safiie

relation to cognitive

197

representation.

They are indeed so closely associated

that to deal xvith them separately in this case is impossible.

But we have already seen the seriousness of Kant's
'^ven

ovm criticism of psycholosy.

if it is granted that

the phenomenal self can be an object of consciousness, it

must be admitted that as an object it is something absolutely different from the objects of outer sense, and cannot be
kno-vvn

by concepts applicable to them.

V/e

have also seen

how necessary it is to Kant's idealism to distinguish the

phenomenal self very sharply from the rest of the phenomenal

v;orld.

Kant, himself, realizes that he cannot by any means

admit the possibility of a materialistic psychology without

endangering his cardinal principle, the transcendental unity

of apperception,
axid

the associated forms of judgment.

In distinguishing between inner sense and pure ap-

perception, Kant has, in effect, opened up the possibility

of the two different lines of development followed by Berg-

son and Hegel respectively. Since he also made time the

sole form of inner sensibility, Bergson's theory represents

the result of emphasis on our awareness of the flux of in-

ner sense and the removal of this from the sphere of cognition.

For Bergson's main charge against Kant is that he

assumed imcritically that time nay be represented in spatial

terms, and that the flux of inner sense may be legitimate-

ly spatialized.

196

Kant's great mistake uas to take time as a homoseneous mediiim. He did not notice that real duration is made up of moments inside one another, and that when it seems to assume the form of a homogeneous whole, it is because it ;_:ets expressed in space. Thus the very distinction which he makes between space and time amounts at bottora to confusinj' time with space, and the symbolical representation of the ego with the ego itself. He thought that consciousness v;as incapable of perceiving psychic states othenvise than by juxtaposition, for^^etting that a medium in which these states are set side by side and distinguished from one another is of course space, and not duration. He was thereby led to believe that the same states can recur in the depths of consciousness, just as the same physical phenomena are repeated in space; this at least is what he implicitly admitted v/hen he ascribed to the causal relation the same meaning and the same function in the inner as in the outer v;orld, 245
V/h?.t

are Bergson's reasons for disagreeing v/ith

Kant?

They are, in fact, echoes of the very points we

have made in connection with Kant's recognition of the flujc

of inner sense,

Bergson points out that ",

.if time, as

iiBwediate consciousness perceives it, v/ere, like space, a

homogeneous medium, science vjould be able to deal v/ith it, 246 aB it can with space". This point is implied in Kant's
own criticism of psychology.

Because rverything in inner

sense is in constant flux, the categories cannot deal with

the appearances of inner sense.
But not only is empirical

science unable to deal with appearances in time alone,

there is no body of synthetic

a_

priori knovjledge, no pure

science, v/hich pertains only to the intuition of time. For

in spite of considerable shifting of ground, Kant was

199

unable to point to any body of synthetic a priori knowledge

which re.iuires

a pure

intuition of time, in the same un-

equivocal and plausible way in which it is arguable that

geometry presupposes a pure intuition of space.

Critique
,

In the

he affirms that kinematics, the general doctrine

of motion, is such a science, but it is apparent that .notion

requires not a pure intuition but an empirical intuition,

and the introduction of an empirical element into the science

destroys the very a priori element

contain.
In the Prole>?:omena
.

'.vhich

the science is to

Kant su^jgests that both the

doctrine of raotion and arithmetic require pure intuitions

of time.
Paton, who says that "the temporal science parallel 247 elucidates

to geometry is, at best, a trifle shadowy",

the difficulties which Kant has in trying to find such a

science.

... he has to bring in change and motion, but change and raotion are not v/holly free froqi empirical elements, and are not on the same footing as time and space. Furthermore the science of geometry takes account of space only, v;hereas the doctrine of motion must take accoiint of both space and time. Since time is the form of inner sense, a pure science of time should enable us to deal a priori v;ith inner states (not v;ith moving bodies), and should offer a basis for psychology rather than for physics. 24S
Paton adds in a foot-note that "The precise nature 249
of the 'doctrine of motion* is a further difficulty".
It seems clear that there is no such pure temporal science,

and those properties of time which Kant is able to state

rest on the very spatial analogy vdiich, in terms of his

200
own position, is hi^^hly questionable.
But it is not merely

the fact that Kant failed to find a pure temporal science

that is important here, although this in itself is significant,

V.'hat

is perhaps more pertinent is that Kant has

himself shoivn the impossibility of such a science in his remarks about inner sense and its constant flux.
The crux of the problem created by Kant's recogni-

tion of the flux of inner sense, and his criticism of

psychology as a science, rests in the relation betv;een

inner and outer sense.
It is Kant's view that time alone
IIow

is absolutely all-pervasive in the domain of sense.

how it is possible that this form of sensibility should

on the one hand be combined v;ith the other form of sensi-

bility, space, in one area of appearances, and on the other

hand should reserve to itself another area of appearances

which are non-spatial, is difficult to 'understand.

Kant is

emphatic on the point that "There is only one time in x;hich

all different times must be located, not as coexistent but 250 but he also seems to

as in succession to one another",

recognize "shbjective" time.

In ray own consciousness, therefore, identity But if I view of person is unfailin.^ly met with. myself frcn: the standpoint of another person (as an object of his outer intuition), it is this outer observer v;ho first represents me in time, for in the apperception t ime is represented, strictly speaking, only in me. Although he admits, therefore, the 'I', v;hich accorapanies, and indeed with complete identity, all representations at all times in m^r consciousness, he will draw no inference from this to the objective permanence of myself. For .just as the tirae in tvhich

201
theI observer oetG i:ie is not the time of uy own but ouserver oex^s me is_ noz zne oinie oi uj^ ovvn out or his sensibility so tVie identity vjTiTcTi is nee ^ssarily boiond up v;ith my coneciousness is not therefore boxmd up with his, that is, vjith the consciousness which contains the outer intuition of ray subject, 251
,

Kant has here sailed into very treacherous waters.

..ucstions arise,

for example, as to our knowledge of the

existence of other people's minds, and as to v;hether one

person's subjectivity may be an object for another person,
and so on.

Kant does not enli^-hten us about such raatters,

or even about the possibility of such knowledge.

This

passage, however, does reveal the difficulties v^hich arise

in connection with the relation between inner and outer

sense.

Briefly, we may ask: are we to suppose that the

(in the narrov/ meaning) is different

time of inner sense,

from the time of outer sense?

If not, by what right do

we suppose that the time v;hich is revealed to us as the

distinctive form of the flux of inner sense is identical

with the time of the phenomenal world as it is dealt with

in physics?

This, specifically, was a problem which was

of paraniount interest to Bergson.

He notes, first of all, that the
t-x.ie

o^

ov.osr

sense is markedly different from the duration of inner sense.

How, let us notice that when \ie speak of time we generally think of a homoseneous mediuiti in whicH our conscious states are rann;ed alonc;side one another as in space, so as to form a discrete multiplicity, VJould not time, thus understood, be to the multiplicity of our psychic states vjhat intensity is to certain of tham,- a sign, a sysibol, absolutely distinct from true duration? 252
.

202

Physical tirae, or measured time, the conception of

which is reached by just those spatial analogies which

Kant asserlis, is time with its peculiar character omitted. Bergson tells us that "II n^est pas douteux que le temps
ne se confonde d*abord pour nous avec la continuite de not re vie intlrieure "
,

and he is speaking here of the

conception of

tirae,
i.'e

the time which we employ to characterize

external things,

have already seen why, according

to Kant's principles, the form of inner sense, (in the

narrow meaning) is not representable in terms of number,

and magnitude, i.e. extensiveness, which are concepts ap-

plicable to coexistent space.

clude?

V/hat,

then, are we to con-

How does Kant manage legitimately to pass from the

time of inner sense to the time of outer sense, or, in other

words, to extend the narrow meaning of inner sense to the

wider meaning which includes the

In Bergson* s words:
" Comment

xi/hole

phenomenal \vorld?

pas sons -nous de ce temps

interieu au temps des choses ?"

plicit in Kant's theory of self to solve,
tirae

This is the problem im-

which Bergson has set

r.im-

Bergson' s solution consists in boldly affirming the

metaphysical primacy of the time of inner sense (dujration)

and in casting out raeasxu'ed or homogeneous time, as a spurious, (though admittedly useful) conception due to the

surreptitious introduction of space in the formation of the

203

conception.

He thereby develops Kant^s view that time is,

indeed, priiuary in the domain of sense, but rejects Kant's

view that time is a homogeneous raediun.

Bercson sees clear-

ly that the reco^^nition of the flux of inner sense makes

it quite impossible to conceive of time as a series, in

which events are placed as points on a line.

He admits

that tine is successive, but he arr;ues that the successive-

ness is not a.'.equately represented by spatial analogies.

.J real duration is what we have alv>iays . called time, but tirae perceived as indivisible. That time implies succession I do not deny. But that succession is first presented to our consciousness, like the distinction of a "before" and "after" set sice by side, is v/hat I cannot admit, 'i7hen we listen to a nelody v/e have the purest impression of succession we could possibly have,- r.n impression as far renoved as possible from that of simultaneity,- and yet it is the very continuity of the melody and the impossibility of breaking it up which makes that impression upon us. 255
,

Duration is not successive in a spatial sense, because such a conception of successiveness implies discrete

differences of state, which have some magnitude, and Kant

himself has recognized that there is not the least magnitude in bare succession.
Thus Kants acceptance of the theory
v?e

have called
tirae

possibility B lies alon^' the path toward a view of

as

pure becoming, in that since the categories must apply to

time itself, time must be a different medium from space*

But its sole difference from space consists in its

204

successiveness, wliich, as we
sist in the fact that the
ti;

!iave

seen, is thoiv'ht to con-

:e-series has an open end on

It is clear,

i;hich nev; events are constantly being added.

however,

tliat

it is the process of becoming at the open

end of the series to which the categories must apply if

they are to be determinative of appearances, because,
(a)

only in this way can it be shoivn that the future in some

respects must resemble the past, and

(b)

since the exten-

sive continuum of time is otherwise indistinguishable from that of space, a recognition of becoming would otherwise be

quite impossible.

Both possibilities A and B recognize

iiray,

becoming in the same

namely, by a conception of a
But A conceives this becoming as

projecting time-series.

a result of the movements of bodies, whereas B accepts

the fact of becoming and attempts to show that certain

principles govern the becoming itself.

But if it v;ere held

that these principles applied to the continuum of that which

has become , either one niust
aiirait

that there is no reiison

to believe that the future will resemble the past, or one

must conceive the whole time-series to be completely laid

out and thus cletemi^ed.

This latter position involves

the complete rejection of the significance of the temporal

distinttions made by the self, as well as the experience

of the becoming of things.

And this is tantamount to the

theory of the manifold, the diffici.aties of which we have

205

already exanined.

In any case, it is clear enough that

Kant did recosnizG the experience of becomins in his

remarks about the flux of inner experience.

Nevertheless, it is difficult to see
hox;

concepts

which are applicable to that which has becone can be

.

pplicable to time as pvre becoming.

In holdin:; that the

categories are schematised by means of time-determinations,

Kant implies that the forms of judgment are embodied in
time itself, VHiatever becomes must become according to

certain general determinative principles.

Thus he calls

the cate^^ories "rules of synthesis", which, as it v^ere,

guide the becoming of things along certain general lines.

This position bears a certain resemblance to the subjective

Qira

of ViJhitehead^s actual entities.

But v/hereas Whitehead

tells us that the actual entity freely chooses its subjective

aim from God, Kant does

iiot

tell us how the categories can

have an independent origin in the nature of the aind, and

can as well be embodied in time-determinations.
V/hitehead's

view may or may not be satisfactory, but it at least

attempts to reveal how the becoming of things is subject
to certain principles.
The difficulty lies in the independV/e

ent origin of the categories,

might perhaps hold that

thinking itself is a process, (in which case, as a process

for the whole of reality, it would be non -temporal, purely
logical, as it is with Hegel), and thus retain the essential

206

point that process is -uided or deterniiied by logical 256 categories. But it seems difficult to understand how a

temporal process must necessarily be subject to categorical determination if it is also held that these categories have an origin independent of the process.
V/e

may

well grant the necessity of dealing with the origin of

the categories and time, separately for purposes of exposition, but we cannot hold that the two are ultimately

distinct without making it hard to see how they ever get

together.
Kant^s Sopernican revolution not only requires

that they must get together, that is, that the categories

must be determinative of appearances through the medium

of time, but also requires that the categories must arise

independently of the realm of becoming entirely.

Other-

wise, he could never hold that they were prior to experience,

which is the locus of becoming.

It seems clear that Kants doctrine of the tran-

scendental imagination represents an attempt to meet this

very problem.

It is the imagination which performs the

successive synthesis, and it does so, according to Kant,

in conformity v/ith the categories.

But why it must do so

remains obscure.

Kant simply tells us that:

The'tv;o extremes ^ namely sensibility and tinderstanding, must stand xn necessary connection with each other through the mediation of this transcendental function ox imagination, because otherwise the former,

207

though indeed yieldinc appearances, v;ould supply no objects of empirical knowledge, and consequently no experience. 257
It is clear that Kant can only maintain this view

that ewpirical know led v;e

i_s

experience if he separates

sensibility, and hence time, from the deterrainative principles.

In order to show the possibility of synthetic

a priori knowledge, Kant has to found the categories on

an absolutely certain and

lui

questionably solid basis,

'le

did in fact found them on v;hat he took to be a solid basis,

namely, the forms of judgment of traditional logic, utilized

with appropriate changes.

In deriving the principles de-

terminative of appearances from a source qiiite apart from

the world of becoming, however, Kant created for himself

the problem of how to get the two back together again.

Hence, as well, the doctrine of the schematism, for as

Kant says, "The schema is in itself alv/ays a product of 25^ ima.-ination'' , But either the production of schemata by
the transcendental imacjination nust have the same apodeictic

certainty as the forms of judgment, in which case the in-

dependent origin of the categories is superfluous, or

there must be some further detennining relation betv/een
the production ox schemata and the logical f^inctions of
the understanding.
The obscurity of this relation re-

presents the seed of Bergson^s radical separation of the

intellect and tine.

II

20a

Bergson, in effect, asserts the insolubility of

this problem by declaring that the intellect falsifies

time in attempting to form a concept of it.
Pure duration

eludes conceptualization.

It eludes that very synthesis

which Kant tells us involves holding all the parts synthesized together, and not dropping them out, as they would
be in bare succession.

There is real duration, the heterogeneous moments -of which permeate one another; each moment, however, can be brought into relation with a state of 'the external xvorld vjhich is contemporaneous with it, and can be separated from the other noments in consequence of this very process. The comparison of these two realities gives rise to a S3nnbolical representation 01 duration, derived from space. Duration thus assumes the illusory form of a homogeneous medium, , . 259
'

Bergson argues, accordingly, that if there is such

a thing as a synthesis of time, it is not a synthesis which

can be conceived in terms of homogeneous irnits, but is a

synthesis which is qualitative rather than quantitative.

Space contains only parts of space, and at whatever point of space v/e consider the moving body, we shall get only a position. If consciousness is aware of anything more than positions, the reason is that it keeps the successive positions in mind and synthesizes them. But how does it carry out a synthesis of this kind? It cannot be by a fresh setting out of these same positions in a homogeneous medium, for a fresh synthesis would be necessary to connect the posii/e are tions with one another, and so on indefinitely thus compelled to admit that we have here to do with a synthesis x.'hich is, so to speak, qualitative j a gradual organization of our successive sensations, a unity resembling that of a phrase in a melody, 260
-

209
It ia, for Bergson, forever impossible to untler-

stand the reality of temporal process by means of concepts.

I-Ietaphysics must reject the sort of analysis which Kant

gives us, and resort to intuition.

Bergson says:

Of the real duration,

But still less coulc". it be represented by con cepts , that is- by abstract ideas, ivhether general or simple. Doubtless no image v;ill quite answer to the original fealing I have of the flowing of myself. But neither is it necessary for rae to try to express it. To him wiio is not ca] able of /^ivin,: hiuself the intuition of the duration constitutive of his being, nothing will ever give it, neither concept nor images. 261
It is apparent that Bergson, too, has accepted the

position we have called possibility (B), but instead of

grappling
vj-ith

the problem of how rational principles can

-ocess, he rejects the problem entirely.
hoi-;

be determinative of

If we are to understaad

the external world arises from

the basic temporal process, we must do so by raeans of this

metaphysical intuition.

If it is true of Kant, as Kemp-

Smith suggests, that an absolutely fundamental principle

of the Critique is that ail analysis presupposes a ore262 it is qqually true of Bergson viously exercised synthesis,

that all analysis falsifies the peculiar temporal oynthesis,

which we apprehend in dujration.

It follovjs that an absolute can only be given in an intuition while all the rest has to do with analysis. We call intuition here the sympathy by xvhich one is transported into the interior of an object in order to coincide with
,

210

what there is unique and consequently inexpressible in it. Analysis, on the contrary, is the operation which reduces the object to elements already knovm, that is, coiiiraon to that object and others. Analysing then consists in expressing a thing in terras of what is not it. 263
At the root of 3ergsons rejection of a logic of

temporal process is his conception of logical form as an

essentially static thing.

Concepts, for Bergson, and the

logical relations between them, represent the antithesis

of temporal process.
And for Kant, as well, the operations

of the understanding, were, as we have seen, completely

severed from all connection with time, after the period of

the Inaiogural Dissertation .
It is just for this reason

that it is all the more difficult to see the necessity of

the immanent logic of the production of transcendental

schemata by ihe transcendental imagination.

Bergson can

congratulate James for having cried out against the "block

universe", can share with James the viexv that future events
are aot determined, just because he denies a logical deter-

mination of the passage of time.

And because the reality

of time is known, as it is for Kant, (as a form of sensibility), through inner experience, Bergson can stress this

inner intuition as more important for metaphysics than

conceptual thinking.

At bottom, Bergson s theory of time

represents the results of carrying out the implications of

Kants distinction between inner sense and apperception.

211
In distin^uishinc between the two, Kant did not lay much

emphasis on the fact that inner sense is conscious inner

sense, in which we are directly aware of the reality of
change.
It is this fact v;hich Bergson seizes upon and
He follows Kant in separating this conscious-

develops.

ness from the mere logical identity of the subject, of

the thinking subject, and is thus led to the view that

this consciousness of change is prinary and quite inex-

plicable in terms of concepts.

Let the numerical identity

of the self be a logical fiction, a mere symbol imposed on

the flow of inner experience.

The consciousness of change

remains.

It follows that the self must be something quite

V/e

unlike a thinking being, or a logical subject.

thus

find Bergson, in effect, defining consciousness in term.s

of memory and anticipation, which in turn are associated
xvith the

temporal distinctions known to the self.

To create the 'future requires preparatory action in the present, to prepare what v;ill be is to utilize what has been; life therefore is employed from its start in conserving the past and anticipating the future in a duration in v/hich past, present and future tread one on another^ forming an indivisible continuity. Such memory, such 264 anticipation, are consciousness itself,
The idealistic element in Kant's thought necessi-

tates a distinction between the self and the material

world, which must be reflected in the character of the

phenomenal self.

His distinction between space and time

212
as forms of outer and inner sense, respectively, involves

the association of time vjith inner experience, and also

accomplishes the required separation of the phenomenal

self from the material world.
But he has further to sepa-

rate the logical functions of the understanding from time

and inner sense,

Kis Copernican revolution, hovjever, re-

quires that the categories should apply directly oo time.

This creates the problem of how this is possible in view

of the disparity betv/een inner sense and its form, time,

and the logical functions.
Kant is also faced v/ith the Thus

problem of how inner sense relates to outer sense.

Kant*s distinction betvjeen space and time serves several

purposes:
(a)

it allovjs him to separate the self from the

(b)

material world;

it allows a recognition of the immediate

()

apprehension of becoming;

it allows him to hold that

only time is all-pervasive in the domain of sense, which in

turn leads to
(d)

the doctrine of the application of the

cate;:^ories to time alone, and hence a complete ansv;er to

Hume.

But this necessary mixing of the temporal distinc-

tions made by the self with the view of space and time as
extensive, which is characteristic of the
tv;o

theories we
\i7hose

have called possibilities A and B, creates for Kant,

viev/s follow possibility B, two

major problems, which

augur a development in the direction of Bergson^s vievjs.

The
tvjo

points,

(a)

and (b) provide the seed of Bergsons

213

duree, a development stemminG from the distinction betv;een

inner sense in the narrow and wide meanings, and the problem

posed by the fact that inner sense in the

lies outside the pale of cognition.
The

narro;-;
tv;o

meaning

points, {c)

and (d) provide the seed of the separation betv/een the in-

tellect and time, a development from the distinction betv/een inner sense and the

faculty of apperception, and the

problem posed by the independent origin of the categories.

There remain, then,
tv/o

further questions,

l.hat

is there in Kants system which would suggest Bergson's

claim that time is the ultimate reality?

And v;hat is there

to su^igest that intellectual fimctions shoiild be associated

with space?

We must now attempt to determine what relation,

tv;o

if any, obtains betv/een Kant and Bergson on these

points.

214
3.

Primordial Time and the Spatializlng Intellect

The two points we are to deal with here are so

closely connected that

one another.

v;e

must treat them in relation to

For Bergsons assertion of the metaphysical

primacy of tine involves him in the view that there io an

intimate connection betvjeen space and the functions of the
intellect.
Indeed, the very method he takes in attempting

to answer the question of how inner and outer sense are

related consists in large part in shov;ing how it is that

the intellect breaks up the flow of time into discrete,

spatial parts.

In the nature of things, therefore, it is

natural that Bergson's solution to this problem does not

consist in a logical explanation, but consists rather in showing that a logical explanation of the relation between
space and time is impossible.
As we have pointed out in

Chapter Two, Section One, above, a theory which makes a

fundamental distinction between space and time has only a

limited number of possible ivays of explaining the relation
betvieen these distinct entities.

Bergson, as

v/e

have seen,

fully accepts the implications stemming from the point

that the diiratlon which is revealed in inner sense cannot
be adequately represented in concepts.

By this token, he

makes it impossible to explain the relation betv/een inner

and outer sense, and hence between space and time, in

215

rational terns.

His explanation therefore leans heavily

on an appeal to intuition.

The appeal to an intuitive i^rasp of tine and space

is not in itself, of course, wholly foreign to Kant*s

views, because time and space for him are, above all, pure

intuitions.

V/hen

Kant tells us that no man can possibly

265

answer the question of how in a thinking subject an outer

intuition is possible,
however, he reveals his conviction

that philosophy must be liraited to the rational explanation

of the possibility of knov/ledge.
V/here reason itself can

make no progress, as in speculations ^vhich extend beyond

the sphere of possible experience, reason must desist from

further inquiry.

But Bergson does not believe that philo-

sophy must stop there, namely, with the explanation of

the possibility of empirical knowledge.

Metaphysics may

proceed by means of intuition,- not indeed the intellectual

intuition which Kant thought would be necessary if there

were to be a knoivledge of things-in-themselves, but an intuition 'Which carries us into the heart of the duration

revealed in our own inner consciousness.

For, in order to reach intuition it is not necessary to ti\.nt;port ourselves outside the domain of the senses and of consciousness. Kants error After having proved by v;as to believe that it v;as. decisive arguments that no dialectical effort will introduce us into the beyond and that an effectever ive metaphysics would necessarily be an intuitive metaphysics, he added that v/e lack this intuition

216

and that this metaphysics is impossible. It would in fact be so if there were no other time or change than those v.'hich Kant perceived and which, moreover, we too must reckon with; . , ,266

Bergson goes on to say that there is another kind

of time than that which has been spatialized by the senses and

consciousness (which tends to spatialize time in the interests of action).

If we can undo this work of spatiali-

zation, we shall have a new kind of knov;ledge, an intuition

of real time.

Intuition doubtless admits of many de,2Tees of intensity, and philosophy many degrees of depth; but the mind once broiight back to real duration will already be alive with intuitive life and its knovjledge of thin,?;s will already be philosophy. Instead of a discontinuity of uoments replacing one another in an infinitely divided time, it will perceive the continuous fluidity of real time which flows along, indivisible. 26?

What relation has this view of time as "one iden'263

tical change which keeps ever lengthening",

to Kant's

theory of time as a form of sensibility, and a pure intuition?

Bergsons view seems to be a logical result of

Kant's in that Bergson carries out the implications of

Kant's admission that there is no substance which persists

through changes of state in inner sense.

Kant's theory of

time as a form of intuition presi^pposes that there may be

matter (of appearances) which can appear in this form.

But it seems clear that there can be no distinction between

matter and form in inner sense.

Since the appearances .

217

themselves, in inner sense, are not subject to catej;orical

determination, which provides for the formal features of

appearances, it is difficult to see how, in
tliis

area of
Kant,

appearances matter can be distinguished from form,

himself, supports this conclusion by saying that in inner

sense no determinate intuition is to be met v;ith.
But if

it is not possible to distinguish between time as a forra

and the appearances which permeate one another in the flux

of inner sense,

v:e

seem to be forced to identify the form

with the flux itself, or rather, to speak more accurately,

to say that time is not a form at all, but is the indivisible change in inner sense itself.

Bergsons theory of

time, then, in this respect is a logical development of

that of Kant.
Thus Bergson*s intuition of time as duration is
an intuition with a concrete content, but a content which
is not broken up in itself into differentiated parts.

Time,

for Bergson, is definitely not an abstract entity. It is

not an abstract becoming, but is

itself.

unique, specific change

The trick of our perception, like that of our intelligence, like that of our language, consists in extracting from these profoundly different becomings the single representation of becoming in general, undefined becoming, a mere abstraction which by itself says nothing and of which, indeed, it is very- rarely that we think. To this idea, -always the same, and alv;ays obscure or unconscious, we then Join, in each

21^

particular case, one or several clear images that represent states and which serve to distin :uish all becomings from each other. It is this composition of a specified and definite state with change in general and undefined that vie substitute for the specific change. 269
But why should Bergson equate this concrete time

with ultimate reality?

'.Tiat

is there in Kant that might

suggest this further development?

In

Are we to suppose that

this case Bergson's

t''o::^y

does not represent a logical

development from that of Kant?

e shall try to show here

that If this further point does not represent the only

logical development from Kant, there are at any rate many

reasons which suggest that at least Bergson's contention

represents one consistent line of developraent from Kant,

Je

have already seen that inner sense, for Kant,

shares in one respect, the role of things-in-themselves,

in that the appearances of iuier sense are not subject to

the categories, and must be considered paradoxically enough,

to be, technically speaking, beyond all possible experience.

Although these appearances are outside of possible experience in Kant's sense of the terra, they are certainly not

psychologically impossible,

B'or

we are immediately aware

of them, and Kant himself must have held that we xvere

immediately axvare of them or he could never have pointed

to the fact of the flux of inner sense as a point hostile

to the rational doctrine of the soul.

If one were to

identify this heterogeneous flux with time itself, as

219

Bergson doos, (since it is not an abstract form), one might

be lad to the view that in
oiir

immediate apprehension of
If the

it, we were in direct relation with reality itself.

mark of the ultimately real is something which is in itself

j^uite

apart from our v/ays of knoiving things, and we are

_ls,

directly av;are of it as it

and not by

r.ieans

of our

categorical ways of knov/ins, not, that is, as appearance,

we must admit that we are ipso facto aware of the ulti-

mately real.
requirements.

The flux of inner sense corresponds to these

It is not subject to the categories.
Ive

do

not knov; it by msans of the categories.

Yet we are

immediately av;are of it.

It is not a mere form, because

it has an indubitable content.

It must, therefore, be

something ultimately real.

Hence, also, as we have

pointed out before, it is misleading to speak of appearances

of inner sense.

There is a further reason, however, for contending that time is an ultimate reality, and this as well stems

from the fact that we are immediately aware of change in

inner sense.
to meet it.

Kant has seen this objection and attempted

Mis statement of it and his reply to it are

as follows.

Against this theory, which adi:dts the empirical reality of time, but denies its absolute and transcendental reality, I have heard men of intelligence so unanimously voicing an objection,

220
that I must suppose it to occur spontaneously to every reader to whom this way of thinking is unfamiliar. The objection is this. Alterations are real, this bein,2; proved by change of ovir own representations even if all outer appearances, together with their alterations, be denied. ICow alterations are possible only in tirae, and time i;3 therefore something real. There is no difficulty in meeting this objection. I grant the whole argument. Certainly time is something real, namely, the real form of inner intuition. 270

But Xants reply does not touch the core of the

difficulty, as Paton points out.

I'Cant takes the contention to be that we are (or inner states) as changing in time, and he has no difficulty in showing that we are equally aware of extei*nal objects as in space. Space and time are on precisely the same footing, and if one is only the form of appearances, so also is the other. The objector does not, hov^ever^ mean that he is axvare of his ideas, or inner states, as in tirae. He means that his awareness is also in tirae, or is a temporal succession. His contention is that tinie r.-ys a very special reality, because it is implied, not merely in what he is aware of, but in his ax/areness of it. l/Vhat he is av/are of may be mere appearance: his awareness of it cannot be mere appearance, but must be absolute reality* And if this is so, time must be, not merely empirically, but absolutely, real. 271

avjare of our ideas

Paton goes on to supply for Kant a more effective

answer to this more stringently phrased objection.

His reply

consists in saying that even if it be granted that time is

implied in our av/areness of appearances, we must be aware

of our awareness of these appearances, and if it is supposed

that this avt/areness, (of awareness) is also in time, we

embark on an infinite regress.

At any stage of this

221

infinite re-ress, Kant can contend that time is merely the

empirically real

foria

of what we are aware of, and not This is a highly incenious

implied in awareness itself.

reply, and must be examined with care.

First of all, we note that the thing which Pat on,

(and Kant) have to prove is that time is not implied in

awareness, that is, in consciousness itself.

Does Patons
The bubble of

reply suffice to show this?

l/e

think not.

the objection is immediately deflated v/hen it is pointed

out that even if an indefinitely extended series of avrare-

nesses of avjareness is involved, this might very well

occur in time.
In fact, it ivould be a rather strange
tiiae.

thing if our awareness of anything did not take some

Thus in speaking of an infinite regress, Paton has merely

dragged in an irrelevancy, and has not shoxm that tine is

not implied in consciousness itself.
It is this point, namely, that we are immediately

aware ox duration in our inner experience, and that this

duration is inextricably bound up with our consciousness that Bergson takes as central to his philosophy.
'272

He tells

us that "The existence of xvhich we are most assured and

which we

knox-*

best is unquestionably our own",

and he

follovjs Kant in making time a condition of our experience

of this existence.

But unlike Kant, he refuses to separate

222

time from the content of our consciousness, the conscious-

ness we have of our own existence.

Consciousness becomes

virtually synonjinous with duration itrslf

The more we succeed in making ourselves conscious of our pro-;rQSs in pure duration, the more v;e feel the different parts of our being enter into each other, and our whole personality concentrate itself in a point, or rather a sharp edge, pressed against the future and cutting into it unceasingly. It is in this that life and action are free. 273
It is to be noted that Kant's theory of space and

time does not allov; one to apply the saae kind of argument
to prove the absolute reality of space.

For the essential

point in the above argument is that time is alvmys involved

in consciousness.

Space for Kant, hovjever, is not alv;ays

involved in consciousness, for tine alone is absoluteJ.y

primary in the domain of sense.

Space, according to Kant,

As Paton re-

is the condition of outer appearances only.

marks, "the fact remains that space is said to be the con274 It is dition of some human experience, but not of all".

conceivable that there might be consciousness without the

presence of space in it, but it is not conceivable, on

Kant's principles, that there should be consciousness

which did not involve time.

It is only because Kant se-

parates the contents of consciousness from the form that

he can maintain that time is only empirically real. But

since time is, for Kant, associated with the undeniable

223

reality of inner consciousness, rather than the external

world, his view
sui:;;:;e3ts

that the reality of consciousness

is necessarily bound up with the reality of time, and

vice versa.

This is Bergson's doctrine and is, therefore,

a logical development of Kant's.

But this suggests also that time may be only empi-

rically real and not even transcendentally ideal.

Bergson

accepts this result, but elaborates a new doctrine of

v/hat

empiricism sho\ild mean.

Traditional empiricism, he

argues, does not start with what is fundamental, but with

what is already derivative and broken up, like a set of

ready-made garments "which will suit Peter as well as Paul 275 Bergbecause they do not shov/ off the figure of either",
son's empiricism is different, and is as v;ell the true

metaphysics.
Ilais vji empirisrae vrai est celui qui se propose de serrer d'aussi prSs que possible 1' original lui-

The empirically real in this sense, then, is also

the absolutely real, and time is by the same token the

ultimate reality.

Hence we must conclude that the relation

between the theories of time in Kant and Bergson in this

respect is one of a natm'al and logical development.
In view of Bergson's acceptance of time as ultimate

reality, xvherein does his explanation of the relation between

224
space and time consist?
lie

It is apparent that the only way

can explain this relation is by somehow deriving space

from tine.

And the only

v;ay

he can communicate

liis

expla-

nation of how it is that conscious beings have outer in-

tuitions of space is to attempt to lead us into an intuitive

grasp of this derivation of space from time.

fact, the course which Bergson takes.

This is, in

He invites us first

to make the effort of acquiring an intuition of duration.

Let us seek, in the depths of our experience, the point where v;e feel ourselves most intimately within our own life; It is into pure duration that we then plunge back, a duration in v/hich the past, alv;ays moving on, is swelling unceasingly v;ith a present that is absolutely new. But, at the same time, we feel the spring of our will strained to its utmost . 277

By this means we can have an intuition of reality

itself, and although a complete grasp of it is beyond hu-

man powers,

Vie

can attain to the intuition of duration in

If we wish to attain the intuition of
ive

varying degrees.

space and matter, hov;ever,

must reverse this effort

and proceed in the direction of a relaxation of the creative

tension.
But suppose v;e let ourselves go and, instead of acting, dream. At once the self is scattered; our past, v^hich till then was gathered together- into the indivisible impulsion it communicated to us, is broken up into a thousand recollections made external to one another. They give up interpenetrating in the degree that they become fixed. Our personality thus descends It coasts around it conin the direction of space. 27^ tinually in sensation.

225
Bersson, in effect, agrees with
i.cnt tiiat

the intel-

lect, and hence pure concepts, do influence or have an effect

on
tiiiie,

but he points out that this is merely a practical

exi,::^ency.

Bergson^s acceptance of concrete change as time

brings the whole question of the relation between space and

time down from the abstract level of pure concepts influ-

encing the pure form of sensible intuition, to the level of

a nat\iralistic explanation of what is involved in our per-

ception of natter and space.

Thus the problem v;hich in

Kant^s terms reraains insoluble becomes explainable in psychological and biological terms.
But even here psychology

and biology are dealt \jith in terras involving liberal doses

of 3ergsons doctrine of intuition, and so it becomes

questionable whether he is not doing metaphysics throughout, a metaphysics designed to deal with that question

concerning the relation between inner and outer sense, which

Kant refused to answer.
But because Bergson has taken the

step of considering time to be the 'ultimate reality, his

explanation of how inner and outer sense come to be related

in one experience, and
hovj

change in general is possible,

rests on a metaphysical explanation of how matter and intelligence arise out of the basic flov; of time.
This explanation, how^ever, carries him far beyond

anything involved in Kant's position on space and time.

It

226

must certainly be said that this aspect of Ber^son's thought

does not represent a logical developnent from Kant's position, or perhaps, that of anyone else.

For Bergson's

acceptance of time or concrete process as ultimate reality

does not necessarily involve the viev; that pure concepts

must be derived therefrom,

shoxv

I/e

have V/hitehead's example to

Plotinus,

that a different development is possible.

v/hom Bergson resembles in some respects on the explanation

of natter, of course, represents the antithesis of Bergson

on this matter, as Ber^'Son himself points out.

More generally, the relation that we establish in the present chapter between "extension" and "detension" resembles in some aspects that which Plotinus supposes. . .v/hen he makes extension not indeed an inversion- of original Being, but an enfeeblement of its essence, one of the last stages of the procession. , . Yet ancient philosophy did not see v.'hat consequences' v;ould resvilt from this for mathematics, for Plotinus, like Plato, erected mathematical essences into absolute realities. Above all, it suffered itself to be deceived by the purely/ superficial analogy of duration v;ith' extension. It treated the one as it treated the other, regarding change as a degradation of immutability, the sensible as a fall from the intelligible. Whence. . .a philosophy vdiich fails to recognize the real function and scope of the intellect. 279

Thinking along these lines, Bergson can, with a

2o0
measure of truth accuse Kant of out-and-out Flatonizing,

All resemblance and development in this respect, then, must

be rejected,

nevertheless, Bergson 's conceptions of matter

and space do resemble those of Kant in content, so that in

examining their views in relation to one another, we can

227
see the point where Bergson's development breaks off and

takes a direction quite opposed to Kants position,

Bergson begins his investigation of the concept

of matter from a phenomenalist standpoint.
He asks us to

put aside all philosophic preconceptions, all latent meta-

physics and look out upon the v/orld of outer sense merely
as it is,
i.hat

do

vje

find there?

V.'e

find ima^;es, and it

is in ima^^es tliat Bergson discovers the answer to the

question "'what is natter?". The

Sian

unaccustomed to the

theories of philosophy, Bergson tells us, would be greatly

astonished to be told either that objects

vjere

only mental

existents, or on the contrary that they v/ere things quite

unlike the objects he perceives.

Common sense, then, seems

to have grasped a fact of prime importance, namely, the

self -existent reality of images as images.

an ai^gregate of images.

Thus matter is

As such, it may be treated as a

system obeying its

ov;n

laws quite apart from perception,

or it may be regarded as containing a privileged image,

namely, that of the body, which conditions all other images,

modifying them in quixotic ways.

The former represents the

position of realism, the latter of idealism; in one, the

mind is an accident, in the other science and its stable

lav/s

of nature is an accident.

"But, for both parties, to

231

perceive means above all to knov;".

228
Perception, for Bercson, is a fxmction of the living organism, and dependent thereon for its scope and richness,
,

contention that

fev; v/ould

dispute.

For Bergson, however,

this fact is the basis of important metaphysical conclusions,

one of which he states in the form of a general law: "Per-

ception is master of space in the exact measure in v/hich 282 action is master of time". But what does this mastery
involve?
It involves the active powers ox a certain image,

the human body, or an animal body, as a centre of action

amid the totality of images which make up the universe.

Pure perception, that is, perception which is considered

apart from the conditioning of memory which accompanies all

actual perception, consists in the action of the body image

whereby v/hatever part of other images which may be detached,

is detached as the representation of these other images.

Thus the representation v/hich appears in consciousness in

perception is not a thing apart from the aggregate of images

v;hich constitute matter; it is among them, in the relation

of a part to a whole.

But the perceptual image is not,

therefore, identical with the object, because the body can-

not lay hold of all of the object image. It can only master

what it can master.

Perception appears, 'then, only as a choice. It creates nothing; its office, on the contrary, is to eliminate from the totality of images all those on which I can have no hold, and then, from each of those

229

which

I retain, all that does not concern the needs of the image which I call iny body. 2&'3

Thus, perception is not, as realism and idealism

take it to be, pure knov7ledp;e . which, as such, has a wholly

speculative interest.

It is rather the action of living

centres of indeternination upon other ina^es in the totality

of matter.
In other xvords let us posit that system of closely-linked images which v;e call the material world, and imagine here and there, centres of real action, represented by living matter: what we mean to prove is that there must be, ranged round each one of these centres, images that are subordinated to its position and variable with it; that conscious perception is bound to occur. . . 2^4

But opposed to pure perception, that is, radically

different from it in kind, is pure memory, a thing of the

spirit.

Bergson rejects the Ilobbesian theory that memory

ilor

is weakened or decayed sense,

in the brain.

are meuiories stored up

Pure perception is a mere extreme v;hich,

if it were ever actual, v;ould be instantaneous; but our

actual perception has duration, and the difference is due

to memory, the v;elling-up of the past into the present.
In concrete perception m.emory intervenes, and the subjectivity of sensible qualities is due precisely to the fact that our consciousness, vjhich begins by being only memory, prolongs a plurality of moments into each other, contracting them into a single in2^5 tuition,

Thus every concrete perception is "a synthesis,

made by memory, of an infinity of pure perceptions which

230 2g6
succeed one another".
Here again we have 3ergsons doc-

trine of the qualitative synthesis, quite unlike that due

to pure concepts and a transcendental iinity of appercep-

tion.

Perceived objects are indeed the result of a synv/ay,

thosis, just as Kant, in a different

and according
It is for

to different principles, thought they were.

ji5rsson a

synthesis which holds together the results of

These pure per-

a succession of pure perceptions in time.

ceptions are fragments detached from the totality of

images, parts of the totality of images v;hich makes up
the material v/orld, for "images outrun perception on every

side".

But the agency v/hich performs this synthesis is

not thou;:ht but memory, and the synthesis is a qualitative

one not a quantitative one.

Thus the multiplicity of ob-

jects in the natural world of experience is brought about

not by ordering principles which stem from pure concepts of the understanding, but by the practical exigencies of
life and the agency of memory.
folloviTS

Thus, Bergson, like K.ant,

the requirements of the theory we have called

possibility B, but unlike Kant the means ivhereby he explains objects in space is partly psychological and

biological,

iievertheless, perceived objects in space

are the result of principles or agencies which act through

the medium of time.

The principle or agency in Bergson s

case is, of course, pure memory.

This raises the question

231
Oi to ;/hat extent in Bergson's system material objects are

distinct from one another in nature.

Bergson tells us that "That there are, in a sense,

multiple objects, that one man is distinct from another

man, tree from tree, stone from stone, is an indisputable 2gg fact". But in reality these objects are not so distinct

as they appear in our perception, for "to perceive consists in condensing enormous periods of an infinitely

diluted existence into a few more differentiated moments

of an intenser life".

To perceive means to immobilize,

Distinct,

and all this is done in the interests of life.

sharply separated bodies are marked out and distinguished

only in the practical interests of life and action.
It is difficult to say to what extent bodies in

the exteraal v;orld have an independent existence, for Bergson, if indeed, they have any.

Matter as an aggregate of

images is "an existence placed half-way betv;een the * thing* 290 and the representation*". It is difficult to under-

stand exactly what Bergson means by this spatial metaphor. Perhaps the best way of grasping his meaning is to approach
it from the standpoint of distinctness itself.

The 'thing*

for Bergson, in its clear-cut distinctness, is an unreality, a fiction of the intellect, employed for purely

practical purposes.

The

representation*

as a purely

232

psycholocical entity, a sensation, v;hich is absolutely

inextensive, is
dichotomy,
tlie

other extreme of this unsatisfactory

real is soniewhere betv/een the
tv;o

./hat is

in distinctness.

That which is j:iven, that which is real, is something intermediate betvjeen divided extension and pure inextension. It is what v;e have termed extensive . Extensity is ti.e most saliant quality of perception. 291

Space as an absolute, honojeneous

i.eaiuia emer^^es

in Bergson's treatment as not lexically prior to niaterial

things, but posterior to then, "like an infinitely fine

network v;hich

v/e

stretch beneath material continuity in

order to render ourselves master of it, to decompose it 292

according to the plan of our activities and needs".

It is a mistake to regard this homogeneous space as
a thin^, as a legitimate object of speculative interest,

when at bottom abstract space is "nothing but the mental

293

diagram of infinite divisibility",

vdiich has arisen as

a fiction of the imagination in the interests of the

needs of life.
There is much similarity
bet^ifeen

Bergson's account

of space and that of Kant, although, of course, in Kant,

as regards the genesis of space,
*u'e

find no equivalent to

Bergson's vitalism.
v;ith

Bergson, indeed, explicitly agrees

Kant in regarding space as a form of sensibility.

233

So we have assvuned the existence of a homogeneous Space, and with Kant, distin:;ui3hed this space from the natter which fills it. i/ith him we have adnitted that homogeneous space is a "form of our sensibility", . . 294

Bergson's professed agreement v;ith Kant, however,

is somewhat misleading. For, as his later writings

abundantly show, Bergson agrees with Kant only in what

Kant denied, namely, a knowledge of the ultiraately real 295 by means of concepts. ..Ithou^^h Bergson agrees with
Kant to the extent that they both, in different v;ays,

follow the theory we have called possibility (B), it is

apparent that he rejects the v;hole Kantian point of view,
v/hen he derives the intellect

itself from primordial

duration.

He agrees that Kant's Transcendental Aesthetic

"appears to have established once for all that extension

is not a material attribute of the same kind as the 296 others". But Bergson's conception of this "form of

sensibility" proves to be radically different from Kant's.

He reviews the three metaphysical possibilities which

Kant had conceived: "either the mind is determined by

things, or things are detei*rained by the mind, or between
297

mind and things

v/e

must suppose a mysterious agreement",

but he also proposes a fourth possibility, which he takes

to be the correct one.

234
.ilia .^liieiTiativo conLix:;os. -ix-oo ox all, in resardin- the intellect as a special function of the mind, essentially turned tov;ard inert matter; then in saying that neither does matter detemine the form of the intellect, nor does the intellect impose its form on natter, nor have matter and intellect been regulated in re^^ard to one another by vje i-noiJ not what pre-established hannony, but that intellect and matter have progressively adapted themselves one to the other in order to attain at last a common form. This adaptation has . moreover , been brou'ht about quite"riaturallyV because it is the sane inversion of the sa..ie ..lovoiient 'vjiiich'"creates at orxco the inteTlectuality of in'ind and t li e materiality of thin;^s . 29^
'

This passage follows after Bergson's explanation

that for Ilant, "space is given as a ready-made form of our

perceptive faculty
we see neither
hovj

a veritable deus ex machina, of

Ix-

which

it arises, nor i/ny

is

vrtiat

it is

299

rather than anything else".

3o

x-te

can only conclude that

v/hether or not Bergson wovild stand by his early agreement

with Kant that space is a form of sensibility, his final

conception of space is quite unlike that of Kant. definitely not a priori in Kants sense.
It is

V/hether Bergson e

theory of space fulfills the same function as Kant*s,

namely, of justifying the applicability of geometry to the

physical v;orld, is highly questionable,

that his theory does this.

Bergson believes

, there is this about it (space) that is , , remarkable that our mind, speculating, on it with its own powers alone, cuts out in it, a priori , figures whose properties we determine a prTori ; experience, vjith v/hich we have not kept in^touch, yet follows us throiigh the infinite complications of our reasonings and invariably justifies them. That is the fact. 300 Kant has set it in clear light,

235
But in spite of this wide divergence between these

two thinkers, their theories of the concept of homogeneous

space and its relation to the space of sensible experience

bear a marked similarity.

The absoluteness and infinitude

of space, for example, is in Kant's viev; a mere idea, the

work of reason.
To assume an absolute space, that is, one which, because it is not material, ccn be no object of experience as ^iven for itself, means assuming: somethin^j tirhich, neither in itself nor in its consequences (notion in absolute space), can be perceived, for the sake of the possibility of experience, v;hich nevertheless nust always exist Vidthout it. Absolute space is in itself nothing and no object at all, but signifies merely everi^ other relative space that I can at any tine c:.. criv outside the given space, and that I can extend bts/ond each given space. to infinity; one that includes the ,;;riven space and in vjhich I can assume it as moved. But since I have the enlarged, although still material space only in thought, nothing I abstract is known to me of the niatter indicating it, from this, and it is conceived, therefore, as a pure, and absolute space, . 301 non-empirical

The conception of absolute space is reached then

by reasoning from given empirical spaces.

Absolute space

is the logical limit of the expansion of finite spaces.

It is just in this fashion, Bergson argues, that Newton,

Euler and others arrived at the notion of absolute space,

A place could be absolutely distinguished from another place only by its quality or by its relation to the totality of space: so that space would become, on this hypothesis, either composed But to finite of heterogeneous parts or finite. space vje should give another space as boundary, and beneath heterogeneous parts of space we should imagine an homogeneous space as its foundation: in both cases it is to homogeneous and indefinite space that XT/e should necessarily retvim, 302

236
Similarly, as micht be expected, the divisibility

of homoseneous space is another feature conraon to the

doctrines of both Kant and Bergson.
divisibility is
a
/uid

for both, this

function of the intellect.

The whole of matter is niade to appear to our thought as an immense piece of cloth in which we can cut out v:hat we will- and sov; it together again as v^e please. Let us note, in passing, that it is this power 'that v;e affirm vihen we say that there is a space . that is to say, a homogeneous and empty medium, i'nfir.ite and infinitely divisible, lending itself indifferently to any mode of decomposition v;hat soever, A medium of this kind is never perceived; it is only conceived. 303

Kant tells us that "the space that is filled by

304
matter is mathematically divisible to infinity",
further, that "mathematics can indeed.

and

.rest in the

certain possession of its evident assertions of the in305 finite divisibility of space". At the same time, Kant is
at pains to point out
tliat

this infinite divisibility of

space is not a property of space as thing-in-itself , but

is merely due to the possibility of conceiving in thought

the infinite divisibility of a given space.

Thus we can only say of phenomena, the division of v^hich 2oes on to infinity, that there exist so many of the parts of the phenomenon, as we give of them, that is, as far as we can ever subdivide. For the parts . as belon:;in to the existence of a phenomenon exist only in thou;;ht . namely , i n their division itself . 306
^:

ilatter, for Kant, although not described in terms

of images, is similar to Bergson*s view of matter in that

237
it is sensed, or felt,- matter is phenomenal. It is de'307 fined as the "movable in space", and its sensible charac-

ter is rsvealed in Kant^s reinarks in which he makes the

distinction between matter and form.

Matter J in contradistinction to fom, i';ould be that -which in external intuition, is an object of feeling, and consequently the properly erapirical of sensible and outvjard intuition, because it cannot be given at all 'a priori . In all experience something must be felt, ""and this is the real of sensuous intuition. 30S
Thus, for both Kant and Bergson the divisibility

of space and the matter, as sensed, which fills it, is

not a property of

i/hat is

actually sensed, but is introtlie

ducec^y neans of the intellect throwjh

mathematical

conception 01 the infinite divisibility of space as an

object of thought.

Their reasons for holding this in

are different.

each case,

hox-^ever,

For Kant the reason

consists in the point that space as an object of thought

is different, in some manner vjhich Kant does not fully

explain, from the pure intuition of space, qua intuition.

His vievj seems to be that a given erapirical space is

potentially divisible, and this is due, no doubt, to the

fact that as a continuous quantity it has been snythesized.
The pure intuition of space in so far as it is divisible

provides the basis of division v/hich in itself is a mathe-

matical construction.
potentially divisible,

Hence although a given space is

v;e

cannot say it has an infinite

236

number of parts, becauso it is thus divisible only so far

as we divide it.
On the other hand Bergson maintains, as
vie

have

seen, that matter is somewhere between pure inextension

and infinite extension.

As such it is extensive but has

not got all the properties which pertain to homogeneous

space,
Ilatter arises from the degradation of the extra-

spatial into spatidlity, but the process does not go on to

absolute conpletion.

The intellectual idea of pure space

"is only the schema of the limit at which this movement

would end",

Spatiality admits of degrees, and "matter

extends itself in space without being abso3.utely extended 310 therein". Thus, in Bergson 's view, there are two kinds
of space, or rather different degrees of spatiality,- the
one exemplified in matter, a kind not carried to complete

spatiality,- the other a thing of the intellect, infinite,

indefinitely divisible, homogeneous.
This exhibits the

main difference between Kants view of the relation of the

intellect to space, and that of Bergson,
There is no evi-

dence to suggest that Kant thought that spatiality could

admit of degrees.

The sharp distinction between inner and

Hev'erthe-

outer sense, in any case, vjould preclude this,

less, the
tvio

kinds of spatiality in Bergson 's philosophy

are related in exactly the same way as are Kants relative,

empirical space, v/hich is felt or sensed, and the absolute

239
space which is the logical limit of the conceptixal expan-

sion of finite spaces.

In each case, the absolute space

is regarded as the liiiit tov;ards x/hich the other tends.

But the character of the tendency is different,

'..'ith

Bergson, the tendency from inextension to absolute spa-

tiality, and also the notion of something which lies betv/een these limits,

is highly obscure. In the absence of

onlj'-

the requisite intuition, one can

fall back on the

logical conception of continuity, which, indeed, Bergson

himself exploits in attempting to cominiinicate this doctrine,

liay

we conclude that Bergson is perhaps tinwitting-

ly introducing the concept of space nere, and that his

doctrine represents the ghost of Kant*s logical expansion

of spaces?
In this case, we might be entitled to conclude

that Bergson s doctrine of space, and even its derivation

from the non-spatial is not so very different from Kant's.

But here, unfortujiately, we are faced with a point which
can never be decided by rational argument.

Bergson thinks that the intellect is characterized

by a latent geometry, v/hich makes it operate the

does.

v;ay it

He goes so far in this direction as to suggest that

deduction itself is due to this latent geometry, which

'311 and that from this "is immanent in our idea of space",

latent geometry arises logic.

In the face of this inter-

pretation of logic, all comparison between Kant and Bergson

240

must come to a halt.

Ber^son is perspicacious enovigh to

see that "from the point of view of the intellect, there

is a petitio principii in makin^j geometry arise automati312 It is, of cally from space, and lo^^ic from geonetiT-".
course, just because Bergson does not assuiue the point of
viev; of the

intellect that he differs so radically from

Kant.

Thus, Sergson^s development of some aspects of

Kant's thought ends in a view which, for Kant, would pro-

bably represent "the chicanery of a falsely instructed

reason''.

241

MOTS
1

Gottfried Ilartin, Kant 3 I-letaphysics and Theory of Science . trans. P.G. Lucas (Kanchester: The I-iaiichester University Press, 1955), p. 11.
<

Iramanuel Kant, Critique of Pure'^leason y trans. Norman Kemp Smith (London: Ilacmillan and Co., Ltd., 1953), p. 6^, Subsequent references to this vork will be designated "Critique", and both the page nuinber of the translation and that of the first and second editions of the original work v/ill be given.
3

Cf, 17, Meinecke, "Die Bedeutung der nichteulvlidischen Geometrie", Kantstudien XI, I9O6. H.J. Pat on, Kant < Metaphysics of Experien ce, (London: Ooorge Allen & Unv.'in, Gottxried T art in, Kant's Ixetaphysics ond Theory of 1936) . Science .

Donald C. .Jilliaras, "The llyth of Passage", Joumr.l of Philo sophy . XLVIII, no. 15, July (1951), p. 457.
'

A. D. Lindsay, The Philosophy of Bergson ^ (London: J.M. Dent & Sons, 1911 J, pp. 2-3.

The predominant -historical influence on Bergson seems to have been French, although the direct and indirect influence of Gerr,ian philosophy cannot be ignored. In this regard see Ben-Ami Scharf stein. Roots of Bergson ^s Philosophy ^ (Ke\J York: Columbia University Press, 1943
)

7
raent of

Perhaps the most notable difference between the measiirespace and the measurement of time, is that spatial lengths can sometimes be compared, side by side, in sensuous intuition, v/hile it is impossible to take one "chunk" of time, and place it beside another.

242

Ernst Cassirer, Substance and Fimction and Einstein *s Theory of llelativltv . trans. ..'illj.nn flurt^a .'^^wn^py anrl Marie CJoilins 3v;abey (Chicago-London: The Open Court Publishing Company, 1923), p. IO5. (Italics not in text).
9

John Alexander Gunni The Problera of Time . (London: George Allen & Unwin, 1929), p. 3?!.
10 Ibid., pp. 373-374.

11
Of. Samuel Alexander, Space. Time and Deity . London, 1934.

12
H, I-iinkovtfski , "Space and Time", in II. A. Lorentz, A, Einstein, -H. Ilinkov/ski, and H. Weyl, The Principle of Rela tivity , trans. W. Perrett and G.B. Jeffery, v;ith notes by A. Sommerfield (London: Metliuen & Co., 1923), p. 75.

13

George Garaow, One. Two. Three. Viking Press, 1947), p. o^.

.

.Infinity .

(New York: The

14 This is a view v/hich Kant, himself, combats. Cf, Critique . ("A" refers to the page nianber of pp. 8O-0I, A39, 356. the first edition, "B" to that of the second).
, .

15

Ethics . I, Def, 3.
16 Ethics . I, Prop. 3. 17 Cf . The Problems of Philosophy . (London: Williams & Nor19l2); also Mysticism and Lor; ic (London: Longmans, Green c: Co., 191^); also Proceedinj^s""of the Aristo telian Society . (1910-11).
r,ate^ Ltd..

IS John Locke, An Essay concerning Human Understanding , ed. Oxford Oxford University Press, A. S.'Pringle-Pattison 1924), Bk. II, Ch. 15, Sec. 12, p. 121.
( :

243

19

Alfred Horth v/hitehead, Science and the riodern l/orld . (Cambridge: Cambridce University Press, 1925 J, Cheap Edition p, 63,

20
Ibid., p. 62,

21 Ibid., p. 62.
22

Ibid ., p. 70.
23

Alfred North Whitehead, Adventures of Ideas, (Cambridge; Cambridge University Press, 1933 i, p. 197.
Alfred North Whitehead, Science and the
Ibid ., p. 156.
26
A, A, Robb, The Geometry of Time -and Space . (Cambridge: Cambridge University Press, 1936) P 19.

Modem World

p,15o.

27

Ibid ., p. 23.

2g

'
-

Broad . Scientific Thoiiarht . Trubner & Co., 1923), p. 217.

CD.

(London: Kegan Paul, Trench,

29
Ibid., p. 55.

30
Jolin

Locke, 0. cit .

3k, II, Ch. 15, 3ec. 1, p. 115.

31

Ibid., II, 15 I Sec, 9, p. Ho. It should be noted that for Locke "ti le" is a term used for measured duration. The latter is the basic idea, and the forrrxer is derived therefrom by measurement. Locke calls space "expansion" to distinguish it from extension, "which by some is used to express this distance only as it is in the solid parts of

244 matter". But Locke is not entirely consistent in this usage, since, in the same chapter, (II, 15), in which he makes this distinction, he reverts to usin^j "extension" as synonyijious with space, Cf, Sec. 9,
32 Ibid., II, 15, Jec, 5, p. 117.
33

CD,
34

Broad, Scientific Thou"-ht

p,

53,

CD. Broad, An Examination of IIcTa.f-;;art*s Philosophy . (Cambridge: Cambridge University Press, 1933), II, ^69.
A, A, Robb, o. cit
, .

35
pp, 6-7

36
Ibid ,, p. 20,

37

Richard Taylor, "Spatial and Temporal Analogies and the Concept of Identity", The Journal of Philosophy . LII, Ho, 22, Oct. (1955), p. 59^::

38
Ibid ,, p, 600,

39
Bertrand Russell, Our I^owledge of the gxtemal 1/orld . (Chicago: The Open Court Publishing Co,, 1929), p. 254.

40

CD.
^^
Ibid.

Broad, Scientific Thoup^ht

p. 61.

p.

5^.

42
Taylor, o. cit .
43
.

p.

601.

Ibid., p. 601.

44

Ibid., p. 602.

245
45
C.D. Broad, Scientific Thou;?:ht . p. 57.

46
Cf. Taylor, o.

cit .

pp. 610-612.

47

Nelson Goodman, The Structure of Appearance . (Cambridge: Harvard University Press, 1951), p. 301,
46
Ibid ., pp. 300-301.

49
C.D. Broad, Scientific Thow^ht . p. 57.
50

Ibid ., p. 59.
51
'

'

Donald G. ..illiaras, "The Sea Fi^ht Tomorrow", in Ilenle, Structure, llethod and Leariin,"; . Essays in Honor of' Henry I-I. Sheffer, (llew York: The Liberal Arts Press, 1951), p. 304.
52

Goodman, 02. cit ,

53

p. 302.

H.R, King, "Aristotle and the Paradoxes of Zeno", in The Journal of Philosophy . ILVI (1949), p. 662.

Ibid ,, p. 667.

Bertrand Russell, kysticism and Logic . (London and New York: Longmans Green & Co., 1918), p. S4.

56

This is the viev; taken by Goodman, who remarks, "Strangely enough it turns out not that time is more fluid than- (say) space but rather that time is more static". 0. cit . . p. 300,
57
<
.

Cf. Hobbes, Leviathan, Pt. I, Gh. I; also De Corpore

246
56

Critique f p. I9O; AI5I, BI9I.

59

Immanuel Kant, De'Iiundi Sensibilis atgue IntelltTibilis Forma et Principiis . trans. Jonii Iianayside in Kant ^3 Inau^-ural Dissertation and Sarlv Writinps on Space . (Chicago and London: The Open Goui-t Publishing Co., 19^9), p. 45.
Ibid,, p.

247
73

Critique , p. 204; A169, B211,

74
Ibid,,

24S
fi9

Ibid., p. 192; A155, B194.

90

Ibid ., p. 166; B154.

91 Ibid., p. 136; AlO?.
92

Ibid., p. 63; A23, B37

93

Ibid ., p. 77; A33, 349.

94
Ibid.
,

p. 56; B20.

95

Ibid ., p. 131; A9S-99.

96

Ibid ., p. 133; AlOl.

97

Ibid., p. 356; A3g6-337.

98

Ibid ., p.
99

346';

A372-373.

Ibid ., p. 355; A3g5.

100 Ibid., p. 6g; A23, B37. 101

G, F. Stout, A Ilanual of Psychology .

6th Ed. p. 496.

102 Critique , p. 255; B291.

103 Ibid., p. 136; AIO7.

104 Ibid ., p. ^7; B67.

249
105
stout,
oj3,

cit.

p, 496.

106
F. C. Bartlett, 'Troblems in the Psychology of Temporal Perception", in Philoaophy . XII (1937), p. 461.

107

Ibid ., p. 463.
IQg

Kant's theory of space and time, however, has been profoundly misunderstood by some psychologists. For example, ary 3oax''t*3 comment on a passage from the Transcendental ^esthetic "Time cannot be a priori in the Kantian sense, because knowledge of it both develops in the individual and appears to vary in- different cases", is, of course, irrelevant. (11. 3turt, The Psycholo/^y of Time , p. 10.
'

109
Critique . p. ^7\ A49, 366.

110 111

Ibid ., pp. 72, SO; A2g, B44; A39, 356,

Ibid., p. b7; B66.
112

Ibid ., p. 77; A34, B51.

113

Ibid ., pp. 346-347; A371.

114
Ibid., p. 351; A37g.

Ibid ., p. 351; A37S.

116
Ibid., p. 346; A373.

117 lie

'

'

Ibid ., pp. 239-240 ; A220, 3267. Ibid ., p. 166; B154.

250

119
Criticuai p. l65; B15-. "In so far as imaelnation is spontaneity^ I sometimes also entitle it the productive imagination , to distinsuish it frora the reproductive iinagination, v;hose synthesis is entirely suoject to empirical laws, the lav;s, namely, of association. .The reproductive synthesis falls within the domain not of transcendental philosophy, but of psychology". Kmpirical la\;s,of course, presuppose the transcendental unity of apperception, the transcendental act of imasination, and the categories. But the analogy concerning the presuppositions of the re<productive sjrnthesis is not entirely exact, as psychology, for Kant, is not a science. See beloxv p. I6l et_ seq .
-

120
C ritique
. .

p. 167; 3x^,.

121 "Taken to.'cether, the analogies thus declare that allappearar^ces lie, and must lie, in one nature". Critique . p. 237; A216, B263.
122

Critique ., p. 209; A177, B220.

123

Ibid ., p. 133; AlOl.

124
Ibid ., p. 226; A200, B2l,3.
125
Ibid.

251 131
Ibid ., p. 352; A379.
132 Ibid., p. 77; A33, 349-50.
133

Ibid ., p. 192; A155, B194.

134
135

Ibid ., p. 330; A343, B401,

'

Pat on, o. cit .

I, 64.

136
137

Ibid .. I, 65.
Critique . p. 369; B407.
13fi

Ibid ., p. 22; B xvii.

139
Ibid ., p. I7O; BI59-I6O.

140

David Hume, A Treatise of- Human Katurej ed. Selby-Bigge' (Oxford: The Clarendon Press, 1896), Bk I, Pt, 3, Sec. XII, p. 134.
141
142
-

Pat on, OP. cit ., I, 556.

Critique , p. 165; A145, B1S4.

143

Ibid ., p. 212; AlSl, B223-224.

144
Ibid ., p. 236; A217, B264.
145
'

Ibid ., p. 237; A215, B262.

252

146
23?-; A210, B255. Kanr^^-^r^';: ^erke sd.-J. H. German text: Imanuel Kant, o5...^Gliche von Kirchmann, (Ileidel-

147 Critique, p. 232; A210-211, B256.

14^
Ibid., p. 261} A241, B300.

149
Ibid., p. 251 J A233, 32S6.

150
Ibid*! P 236; A215, B262.

151
Ibid .f p. 196; Al60, B199.
152

Ibid., p. l&k;
153

Alii.3,

Blg2.

Ibid., pp. 166-167; 3154.

154
Ibid ., p. 167; B154.

155
Ibid .y p. 112; A7g, B104.

156
Ibid., p. 204; A170, B211-212.

157
Ibid., p. 295j A291, B347.

153
Ibid., p. 63; A24, B3f^-39.

159
Ibid ., p. 193; A163, B203.

160

Ibid ., p. 199 J A163, B204.

253

161
Ibid., p. lo7; B154.

162
Ibid,, p. 167; 13154-155.
163

(Kants footnote).

Ibid ., p. 167; B154-155. (First italics not in text).

164
Ibid ., p. 167; B155.

165
Ibid ., p. ira; A13 9-140, B169-170.

166
Ibid ., p. 183; A142, BlCl.

167

'

Paton, . cit.

II, 7^.

I6g
Critique
,

p. 376; B420.

169 This is a paradox because the metaphysical theory of the manifold is supposed to accoiint for all contents of the universe v/ithout residue . In vjhat does the individualioy of the consciousness, which is liiTiited to certain finite regions at certain times, and yet continues the same in di fferent regions at other times, consist?
-

170
Critique , p. 136, AIO7.

171
Henri Bergson, Creative Svolution . trans i Arthur i^itchell (New York: The Kodenn Library, Random House, Inc., 1944),
p. 3.

172

Critique , p. 291; A265| B341.

173
'

Ibid., p. 29O; A2g3, 3339.

174
Ibid., p.

.,

2^55;

A274, 3330.

254
175

Ibid ., p. 166; 3153. 176

Bergson, Great l\re Svolutlon . pp. 5-6,
177
Critique, p. 353; A3gl.

17^

Ibid ., p. 209; A177, 3219.

179

Ibid ., p. 204; /a70, B211,

ISO
Ibid ., p. 67; A23, B37.
131
Ibid., p. 75; A31, 347.

1S2

Ibid ., p. 343; A373.

133

Ibid ., p. 232; A270, B326.

134
IrananuGl Kant, Iletaph^sical Foundations of Natural Science trans. Smest -ieifort Bax in Kant*s Prolagomena and riGtaphysical Foundations of HaturaT" Science . (London: George Bell and Sons, 1391) i p. 14^. References to this xvork^vill be hereinafter designated "Bax".

135
Bax, p. 137.

136 Ibid ., p. 222. 137

C ritique
,
.

p. 332, A3 47, B405.

133

Ibid ., pp. 352-353; A331.

139
Ibid., p. 255; B291.

255

190
Ibid.
p. 34^; A3 64.

191
Ibid.
!-

3^Jf

Ji.364.

192

Ibid
193

p. 214;

Alg3, 3226.

Ibid.

194
Ibid
>

p. 255; B292

195
Ibid,
p. 254; 3291.

196
Ibid,
p. 353; A3fil.

197
Bax, p. 221,

19s
Grioique . p. 373; 3415.

199
Bax, pp. 140-141.

200
Ibid .
.

p. 141.

201
Ibid ..

202
^By nature, in the empirical sense, we imderstand the connection of appearances as -regards their existence according to necessary rules, that is, according to lav;s.' There are certain laws which first make a nature possible, and the3e lavjs are a priori" . Critique , p. 237; A216, 3253,
203

Critique . p. 167; B154. (Kant repeats this point several times throughout the Critique
)

256

204
Bax, pp. 220-221,

205
Ibid .

206
Ibid .
207
-

'

Henri Ber^son, 'Introduction to I-ietaphysics" , trana, Mabelle L. ikidison in The Creative I_Iind (iJev; York: The Philosophical Library, 194-6 J, PP. 191-192.

20S
Ibid ., p. 191.

209
Critique , p. 359; A392-393.

210
Ibid ., p. 355} A3S6-3g7.

211
Henri Bergson, Tine and Free Will ( Essai sur les donn^ea iimaediates de la conscience trans. F.L. Fogson (London: George Allen ^ Unwin, 1910 j, p. 100.
)

212
Bax, p. 141

213

Henri Bergson, "The Perception of Change", trans. I^abelle

L. Andison, in The Creative Mind , p. 173.

214
Bergson, Time and Free Will
,

p.

122.

215
Ibid ., p. 106.

216
Critique . p. 17; B viii.

217
G.W.F. Hegel, The Science of Lo,?:ic . trans. .H. Johnston and L.G. Strut hers, 2nd Ed., (London: George Allen & Unwin, 1951), I, 39.
'

257

21g
Critique , p. 77; '03, B49-50.

219
Ibid ., p. 204; A169, B211. 220
Ibid ., p. 196'} A162, B203. (Kants German "Kirn ist das Bevj us G t> se in de s I .annigralt ijs;en Oleic hart i,:en in der Anschauun.a; Ifoerhaupt * . >" v.'ould seem to be less av<rkwardly

rendered as ". .consciousness of the fold in Intuition in general. . .''

221
Ibid,, p. 5ol; A72C, 374^.

hor.;or;eneous

mani-

222

Ibid ., p. 133; A102. Italics not in text.

223

Bergson, Time and Free V/ill . p. 75

224
Ibid., p. 77.

225
Ibid., p. 7S.

226
Inunanuel Kant , ProlegiCiHena to any ?ut\ire Metaphysics trans. Paul Carus, lieprint Zd, (Chicago and London: The Open Court Publishing Co., 1949), Sec. 11, p. 36.
-

227

Critique , p. 198 j A162, B203.

Ibid ., p. 199; A163, B204.
229
Ibid., p. 255; B292.

230
Ibid ., p. 255; B291.

231
Ibid., p. 255; B292.

232

Paton, 0. cit., II, 279-2g0.

25S
233

Jritique , p. 214;

MB},

3226.

234
Ibid., p. 38C; A412, B439.

235
Ibid ., p. }B; A412, B439.

236 237

'

Ibid ., p. 214,

..j.o3,

3226.

Bergaon, Time and

j'ree

uill . p. S?.

23a
Critique , p. 112; A7S, BIO4.

239
Ibid .

240
Bergson, Time and Free Will , p. SO.

241
Ibid ., p.
242
6'1.

Ibid ., pp. 62 -S3.

243

Critique , p. 199-200; AI64, B205.

244
Ibid ., p. I6g; BI56.

245
Bergson, Time and Free VJill . p. 232

246
Ibid.
,

p. 234.
' '

247

'

Pat on, o. cit .. I, 129.

24d

Ibid., I, 12a.

259

249
Ibid . 250
Critique , p. 217; Alfig, B232,

251
Ibid ., pp. 241-242; A362.
252

Ber^son, Tirae and Free Will

253

p.

90.
-

Henri Bergson, Purge et 3imultaneite . 2nd Sd. enlarged, (Paris: Librairie Felix Alcan, 1923 ;, Ch. 3, p. 54.

254
Ibid ., p. 55.

255
Bergson, "The Perception of Change", in The Creative Mind, p. 176.

256
This, in itself, has its difficulties, as for example in Hegel's transition from the Absolute Idea to Nature.

257

Critique . p. I46; A124,

Ibid ., p. 1^2; AI40, B179.

259
Bergson, Time and Free V/ill . p. 110.

260
Ibid., p. 111.

261
Bergson, "Introduction to Ketaphysics" in The Creative Mind, p. 195

262

Norman Kemp Smith, A Comiaentary to Kant's Critique of Pure Reason, 2nd 3d. rev. co enl., (London: Ilacmillan, 1930),
p. 96.

260
263

Bergson- "Introduction to Metaphysics" in The Creative

Ilind . p. 190.

264
Henri Jerijson, "Lire and Consciousness" in Mind-Snergy . trans. II. l/ildon Carr (London: Ilacmillan, 1920), p. 13.

265
Critique
.

p. 359; A3 93.

266
Bergson, "r^.iloso rir:al Intuition" in The Creative

Mind , p. 151.
267

Ibid ., p. 150.

26g
Ibid .

269
Ber^son, Creative gvolution . pp. 330-331.

270

'

Crioique . p. 79; A36-37, B53.

271
272

Paton, 02. cit .. I, lo2,

Bergoon, Creative ijlvolution . p. 3.

273

Ibid ., p. 220.

274
275

'

Paton, o. cit .

I, 146.

Ber^iSon, "Introduction to lietaphysics", in The Creative Hind , pp. 206-207.

276
Ibid ., p. 206. French text:- "Introduction a la Mltaphysique" in La Pensee et Le Iiouvant . Oeuvres Completes d' Henri 3ergson, (Geneva: Editions Albert Skira, 1946),
p. l^g.

261
277

Dergson, Creative Svolutioii pp. 21?J-219, f

27s Ibid ,, pp. 220-221. 279 Ibid ., p. 230, 2S0

Bergson, "Introduction to Iletaphysics", in Creative Mind , pp. 232-233.

281
Henri Bergson, llatter and Memory , trans. Nancy Margaret Paul and ;v. Scott Palmer (London: George Allen Ez Unwin. * 1911), p. 17.
2g2
Ibid., p. 23.
263

Ibid ., p. 304.

2^4
Ibid ., p. 21.

2^5
Ibid ., p. 292.
26-6

Ibid ., p. 236.

2^7

Ibid ., p. 305.
Ibid ., pp. 277-278.

269
Ibid ., p. 275.

290
Ibid .
.

p. III.

291
Ibid .
.

p. 326.

262

292

Ibid ., p. 305.
293

Ibid ,, p. 273.

294
lioi'^^jpon,

Tine and Free

dUl

p. 236.

295
"That is vjhat kaiit orou^ht out so clearly and that, it seems to ne, is the greatest service he rendered to speculative philosophy. He definitively established that, if uetaphysicG is possible, it can be so only throiigh an effort of intuition", "The Perception of Change" in The Creative Mind , p. I65. Thus Bergson agrees with Kant denial of the ability of Reason to knov; ultimate reality. But, of course, Bergson does not agree xvith Kants denial of a metaphysical intuition.

296
Bergson, Creative Svolution . p. 223.
297

Ibid ., p. 225.

296
Ibid ., pp. 225-226.

299
Ibid., p. 224.

300
Ibid ., p. 223.

301
302

'

Bax, p. 151.

Bergson, Matter and Memory , p. 256,

303

Bergson, Creative involution , p. 172.

304
Bax, pp. 176-177.

263

305
Ibid ., p. 179.

306
Ibid ., pp. l50-lgl. Italics not in text,

307 Ibid ., p. 150.

30g
Ibid ., p. 151.

309
Bergsoa,
./re^-oivu

'

.liVoiution ,

p. 221.

310
Ibid ., p. 223.

311
Ibid ., p. 230.
312

Ibid ., p. 232,

264
BIBLIOGRAP^fY

W.G, Only those

TTforks

referred to in the text are listed.

PRIiViAiTi

ouoi i^-JO

Bergson, Henri Oeuvres Conpletes . Geneve: Editions Albert Skira, 1%6.

-. Buree et Siinultan^itl. Paris: Librairie Felix Alcan, 1923.
Time and Free 17 ill . trans. F, L. Pogson, London: George Alien &. Unwin, 19IO

., Creative Evolution , trans. Arthur Mitchell, New York: The Modem Librai?y, Randon House, 1944.
l-latter anc. I'lemory , trans, Ilancy liargaret Paul , and VJ, Scott Palmer, London: George Allen & Unwin, 1911.
.

iiind>;Energy, trans.
192^5

H,

Wildon Carr, London:

Macmillan,

-. The Great iye Fdnd trans. Ilabelle L. Andison, New York: The Philosophical Library, I946,
An Introduction to Metaphysics , trans. T.E. Hiilmej iutro. Thomas A. Goudge, New York: The Liberal Arts Press. 1949.

Kant, Imiaanuel oSfbmtliche Werke . F^-d. J. H, von Kirchmann, Heidelberg: Georg Jeiss, 1891.
- Kritik der relnen Vernunft . (Yol. I of aboveJT
Iv'orks

as

H.

, Kleinere Schriften zur Ijaturphilosophie . Sd. J. von Kirclmann, Berlin: L. Iieimann, 172,

Critique of Pure Reason , trans. Norman Kemp Smith London: Ilacirdllan Ik Co., 1953. th,

265

- Metaphysical Foundations of Matural Science , trans, .Ernest Belfort Bax, London: Geor/ie Bell Tc Sons, IS9I.

. rrole;?:oraena to any Futui'e rietaphysics . Ed.& trans. Paul Uarus, Chicago and London: Open Court Publishing Co., 1949.

- r.ant*3 Inau.^ural Dissertation and garly Uritin^a on Space , trans. Jolin Ilandyside^ Chicago and London: Open Court Publishing Co., 1929.

SSCOKDARY SOUx^CSS
Books
Alexander, Sai/mel. Space. Time and Deity . London, 1934,
Broad,. C, Di , Scientific Thought . London: Kegan Paul, Trench, Trubner k Co., 1923.
An Examination of IcTa.'^rgart^s- Philosophy . Cambridge: Cambridge l^ni varsity Press, 1933.
I

Cassirer,3mst. Substance and Fimction andlSinstein^s

Theory of Relativity , trans. V/illiam Curtis 3-.;abev and Marie Collins Sivabey, Chicago and London: Open Court Publishing Co., 1923.
Einstein, Albert, H. I1inko*^w/ski, ?i.A. Loi'entz, i.nd H. IJeyl. The Principle of lielativity . trans. W. Perrett and G, B. Jefferyi ;vith notes by A, Soimaerfield, London: Ilethuen & Co., 1933.
Gamow, George. One . Two . Three . Vikins Press, 1947.
. .

Infinity .

Ilexir

York: The

Goodman, Kelson, The Structure of Appearance . Cambridge: Harvard Univex'sity Press, 1951.
Gunn, John Alexander. The Problem of Tine . London: George Allen So Unwin, 19W. Hegel, G.V/.F. . The Science of Logic . trans. v',H. Johnstor and LiG. Struthers , 2nd Tild , , London George Allen & Unwin, 1951.
:

266
..enle,

(ed,). ^triicbure. I.ethod and Leaning . New York: The Liberal Arts PrusG, l^^l.

Hobbes, Thoiaas, Leviathan .

-. De Corpore ,

HuTie,

David. A Treatise of Jlirnan Nature . ed. Selby-Bigge, Oxford: The Clarendon Press, 1S56,

Lindsay, A.D. , The Philosophy of Ber/:3on . London: J.M. Dent So JonsV I9II.
Locke, Jolrin, An jJ^Gsay concerning!: II\jiaan Understanding: , ed, A.3, Pringle-Fattison, Oxford: Oxford University Press, 1924.
I'lartin,

Gottfried. Kant^s I-Ietaphysics and Theory of Science Manchester: I>Ianchester University Press, 1955.

<

Pat on, H.J, , Kant*s Metaphysics of Iilxperience . London: George Allen ^ Unwin, 1936.

Robb, A. A,

Garabridt!;e

The Geooetry of Time and Space . Cambridge: University Press, 193 &.

Russell, Bertrand, The Problems of Philosophy . London: V/illiams & Worgate, ~I912.
i

Mysticism and Lo^ic . London: Lon^nans, Green

F.c

Co., 1918.

Our KnowledF:e of the Sxternal World . Chicago and . London: Open Court Publishing Co., 1929.

Scharf stein, Ben-Ami. x^oots of 3er;^son*s Philosophy . Mew York: Golumbia University Press, 1943.
Smith, Noman Kenp. A Commentary to Kant ^s Critique of Pure. Re ason . 2nd Sd. rev. ic enl. London: Mcmillan

Spinoza,

St hies .

Stout, G.F.. A 1-lanual of Psychology . 6th Sd., London: Tutorial Press, 1938.

267

Sturt, liary , The r-sychology of Time . London: Kesan Paul, 19ZT.

Vifhitehsad,

Alfred North . Science and the Modem Uorld, Cambridge: Cambridge University Press, 1926.

- Adventures of Ideas . Cambridge: Cambridge Univarsity Press, 1933.

Articles
Bartlett, F.C. .- "Problems in the Psychology of Temporal Perception", Philosophy . XII (1937).
King, H.H, . "Aristotle and the Paradoxes of Zeno", The Joixmrl of Philosophy . XLVI (1949).
Iieinecke, W,-. "Die Bedeutung der nichteiil:lidischen Georaet rie " , Kantstudien XI , ( 1906 )
V/illiams, Donald G.
.

Philosophy .

}[L\T:II

"The Myth of Passage" Journal of (1951).

Taylor, Richard . "Spatial and Temporal Analogies and the Concept of Identity", Journal of Philosophy . LII (1955^.