Bradleys Dialectic

CO CO

g]<OU 164679 g
~

OSMANIA UNIVERSITY LIBRARY

Call No. 111.5/C56B Accession No. 23495

Author Church, R.^ r .

Title Bradley 's dialectic

This book should be returned on or before the dale last marked below.

Bradley's Dialectic

By the same Author

A STUDY IN THE PHILOSOPHY OF
MALEBRANGHE

HUME'S THEORY OF THE UNDERSTANDING
AN ESSAY ON CRITICAL APPRECIATION

BradleyY Dialectic

by

Ralph Withington
Church

D.Phil.(Oxon.)

Sometime Lecturer In Philosophy at

Balliol College

Associate Professor oj Philosophy in
Cornell University

LONDON

George Allen & Unwin Ltd

First published in 1942

All rights reserved

Printed in Great Britain by
Billing and Sons Ltd.* Guildford and Esher

Preface

THE main aim of this work is two-fold. The primary object-
ive is to bring out in somewhat simple terms the essential
character of Bradley's dialectic. To that end, criticism of
the dialectic is, for the most part, confined to a chapter oh
some basic difficulties in it, in order that excursions in
criticism may not unduly complicate exposition of the
central doctrine. Perhaps it ought to be said in passing
&Eat the difficulties considered in that chapter are not the
only ones with which Bradley's dialectic seems to me to be
infected. They were selected for emphasis because some
selection was unavoidable and these difficulties seem to
have been too generally neglected.

A parenthetical few words may be called for as to the
sense in which the term " dialectic" is here used. In some
quarters "dialectic" has been given a perverse significance.
Thus it is frequently used to mean what is meant by "verbal".
A line of argument that is considered to be no more than a
matter of words is dismissed as being "dialectical". There
would appear to be no etymological justification for this
usage. In its most radical sense, "dialectic" means what is
meant by "elucidation". A dialectic is a method or way of
elucidation. The history of philosophy alone makes it
clear that there are diverse methods of elucidation. But
Hegel, and Bradley after him, claim there can be only one,
the dialectic of relational essence; what Bradley calls "the
relational way of thought". Bradley's dialectic is his method
of elucidation.

It is rather difficult to avoid the unpalatable conclusion

5

Preface

that those who, by virtue of temperament and training,
are foreign to any form of systematic Idealism, find it easier
to call down criticisms on Bradley than to follow him. And
that is no more than natural, so long as the reader of a
philosophy of Hegelian origins insists that it must square
with Aristotelian logic. Hegel, and Bradley after him,
repudiate the logic of contradictories and seek to replace
it with the dialectic of contraries, which they call the true
logic. Bradley denounces the Law of Identity as being a
tautology and therefore inane. He then identifies the
contradictory with the contrary, thus to proscribe the Law
of Excluded Middle.

Now anyone who has the patience to examine this repu-
diation of the Law of Identity will see that it is no mere shift
in doctrine. Rather it is a radical innovation in principle.
For consider: on this view there may be no contradictories,
as the contradictory is formulated by the Law of Non-
contradiction. The Law of Identity being repudiated, the
contradictory is identified with the contrary. That is why,
in a word, there is on this view a middle term between any
two co-opponents. And Bradley assumes, as would a thinker
out of the tradition from which he largely derives, that
Appearance, or everything short of Absolute Reality, is in
process. Thus the middle term between any two differences
is a moment of mediation in process, not a self-identical,
static being. This moment of mediation Bradley calls rela-
tion; and the terms mediated he calls qualities.

For several reasons this repudiation of the Law of Identity
and the identification of the Contradictory with the Contrary
entail the consequence that the identity of no matter what is
"relational". The theory of relational identity is the burden
of the neglected chapter on Relation and Quality in Appearance
and Reality. That neglected chapter of Bradley's "meta-
physical essay 55 is the subject of the first chapter of the
present work. In that chapter it is brought out that quality

6

Preface

and relation mutually contribute to constitute the identity
of each other. And this conception of relational identity as
the essence of the internality of relations in Bradley is con-
trasted with each one of the ten senses of ''internal relations"
which have been made out by Dr. Ewing. This is followed
by a brief chapter on Bradley's treatment of Space and
Time, thus to afford an illustration of how Bradley applies
the dialectic of relation and quality to two categories of
Appearance.

At the end of the chapter on Relation and Quality ', Bradley
writes as follows, "The reader who has followed and has
grasped the principle of this chapter, will have little need
to spend his time upon those which succeed it. He will have
seen that our experience, where relational, is not true; and
he will have condemned, almost without a hearing, the
great mass of phenomena." Accordingly, after reviewing
Bradley's treatment of Space and Time, further considera-
tion of the contrariety with which Appearance is infested
"everywhere and always", as that is elucidated by Bradley
in the remaining chapters of Part I, is omitted. Thus
chapter III of the present work is concerned with the
opening chapters of Part II the Part of Appearance and
Reality in which emphasis is laid on degrees of Reality.
Bradley's arguments to his criterion of Reality, and the
ways in which the dialectic of relation and quality yields his
monism, are the main topics of this chapter, which leads us
in chapter IV to the Internality of Thought and Reality.

In this chapter Bradley's conception of the nature of
thought is considered in some detail. To that end, the work-
ing of thought in and through stages of judgement into more
and more comprehensive degrees of Reality is emphasized
in order to bring out the cardinal sense in which thought
and Reality are internal to each other. In order that this
conception of thought may be grasped, we must see the
source of thought as being what Bradley calls "the this"

7

Preface

and "the mine", or virtual immediacy. For in Bradley's view,
thinking develops out of virtual immediacy into "the ideality
of the finite" by virtue of the inherent self-transcendence of
all psychical fact. Indeed, the " ideality" of the finite is the
inherent self-transcendence of the finite, in so far as that self-
transcendence is ideational in character. Moreover, the self-
transcendence of psychical fact is seen to be the very process
of fission in sentience by virtue of which differentiation or
relation occurs. Thus, since thinking is the ideality of the
finite, or moment of differentiation in experience, thinking
is relational. The consummation of a process of thinking
in judgement is the fulfilment of a thought. This is achieved
in a synthesis of the initial content of the thought within a
context of relations and qualities which differ from those
wherein the thought thus consummated in judgement had
its inception.

These considerations bring us to the coherence theory of
truth and Reality, which is the main concern of chapter V.
The nature of error as the relative privation of coherence
in judgement, and the sense in which error is both unreal as
a privation and real as some (however slight) degree or
other of coherence, are brought out as a preface to Bradley's
theory of truth.

On that theory, a judgement is more or less true as it is more
or less self-coherent. And it is self-coherent to the degree
to which it is internally related within the systematic Whole
that is reality. The criterion of truth is comprehensiveness.
The more comprehensive the scope of the qualities and
relations that are the content of a judgement, the more fully
the judgement is true.

The criterion of degrees of truth is likewise the criterion
of degrees of Reality. An Appearance that is more com-
prehensively self-fulfilling and self-fulfilled than another
appearance is the more real. Questions as to the explicit
sense in which judgements and appearance are more or

8

Preface

less coherent bring us to the principle of identity in difference
which is the topic of the next chapter.

This principle arises out of the identification of the con-
tradictory with the contrary, between whose terms there is
a middle term. This middle term is the moment of media-
tion, differentiation, or partial identity between and in the
terms thus mediated. The principle of identity in difference
is the principle of the dialectic of relation and quality. For
the moment of mediation the middle term is the moment
of differentiation that is relation. And the terms mediated
are the qualities that are thus related.

Since Appearance is everywhere and always relational,
and the Absolute may not be a term in relation and remain
Absolute, Bradley's relational dialectic poses a question
as to what may be the relation of Appearance to Reality.
That question is examined in chapter VII, on the relational
and the Absolute, which is followed by a chapter on some
basic difficulties in Bradley's dialectic. These difficulties
converge on the radical point in principle that, for Bradley,
identity is relational (in his sense of the term), not absolute
or tautological, as in A is A. And that brings us to the second
main objective of this work.

That aim is, in a word, to point out the disjunction between
identity as relational, and identity as an absolute A is A.
Since identity may not be both relational and absolute:
since we may not both affirm and deny the Law of Identity:
this is truly a disjunction, not a pair of alternatives.

It is a necessary condition of any coherent understanding
of the dialectic, and of the concrete universal which it
elucidates, that the notion of relational identity, or identity
in difference, be borne in mind. For unless it be understood
that by contradiction Bradley means what (to a non-
Hegelian) is meant by contrariety, only bemused irrelevan-
cies, such as Dr. Broad's monstrous assertions about Bradley's
theory of relations, can result. And moreover, unless that

9

Preface

cardinal point in principle be taken into account, a necessary
condition of an understanding of the relational way of thought
in any of its expressions, whether dialectical, political, or in
economics, is thus ignored.

Something strangely like a creed of exclusive party
loyalty seems to have infected students of philosophy in
some quarters. The doctrinaire bigotry thus engendered is not
the less deplorable because it professes to be intellectual.
As a consequence of this single-minded devotion to non-
Hegelian modes of philosophy, otherwise sensible men,
who have avowedly steered clear of Hegel, take his name in
vain with some abandon. These same men apparently
sit down to read Appearance and Reality upon the thoughtless
assumption that its major doctrine could only have originated
on either or both sides of Didcot Junction. The misunder-
standings that result from such cavalier procedure have
made Bradley, in these same quarters, a synonym for non-
sense.

Yet the dialectic that Bradley calls "the relational way of
thought" formulates a mentality an almost hopelessly
romantic way of imagination and evaluation that is abroad
today in force, and we badly need to take account of it.

This commentary was commenced some eleven years ago
during a memorable and very pleasant sojourn at Balliol.
Since theto, it has been recast several times, and each time
the more severely limited in scope and detail in an endeavour
to arrive at a comparative simplicity of treatment. The
reader will find that Bradley's theory of relations is re-
iterated at certain junctures in the course of the com-
mentary. In view of the basic significance of that theory
in Bradley's dialectic, some repetition of it in certain
contexts was deemed advisable.

The work owes more than could be even indicated to the
Seminar of the late Professor H. H. Joachim on Hegel, and

10

Preface

on Bradley, which it was my good fortune to attend through-
out two years. It owes no less to the tutorial skill and kind-
ness of Mr. J. D. Mabbott of St. John's. And it owes much
to the benevolent ministrations of the late Professor J. A.
Smith, whose frequent alarums and excursions over the
perils lurking and working in the writings of his friend
Bradley would have aroused the curiosity of a Bourbon.
An acknowledgement of gratitude in this regard could only
be so inadequate as to be virtually meaningless.

Students at Harvard, in the Seminar on Metaphysics
with which I was charged during a visiting Lectureship in
1935, gave me many helpful criticisms of parts of the essay
as it stood at that time. In that connection, I am par-
ticularly indebted to Dr. Milton Gross, now of Columbia
University.

As in many other ways, I am gratefully and deeply
indebted for critical instruction in systematic Idealism to
discussions over a period of years with my colleagues,
Professor G. Watts Cunningham and Dean George H.
Sabine, as well as with Professor W. R. Dennes, of the
University of California. These friends were also kind enough
to criticize the work in manuscript form.

I wish to express my gratitude to the Oxford University
Press for their kindness in allowing me to make extensive
quotations from Bradley's works.

R. W. G.

SANTA BARBARA,

December, 1940

II

Contents

CHAPTER PAGE

Preface 5

I The Dialectic of Relation and Quality 15

II Space and Time 56

III The General Nature of Reality 62

IV The Internality of Thought and Reality 81

V The Coherence Theory of Truth and

Reality 94

VI Identity in Difference 109

VII The Relational and the Absolute 119

VIII Some Basic Difficulties 133

IX Relational and Absolute Identity 161

Index i 88

"But I clearly recognize that, if
Not-A were taken as a pure negation,
no compromise would be possible.
You would then have to choose be-
tween the axiom of contradiction and
the dialectical method."

F. H. BRADLEY

CHAPTER I

The Dialectic of
Relation and Quality

IN the course of the first three chapters of his "metaphysical
essay". Appearance and Reality, Bradley reaches a conclusion
that is fundamental to his entire metaphysics. This conclu-
sion, moreover, is valid for, and is variously illustrated by,
the several arguments which constitute the succeeding
chapters of Part I of that essay. "The reader who has followed
and has grasped the principle of this chapter (Relation and
Quality), will have little need to spend his time upon those
which succeed it. He will have seen that our experience,
where relational, is not true; and he will have condemned,
almost without a hearing, the great mass of phenomena." 29 *
Bradley arrives at the principle that is expounded in the
chapter on relation and quality by way of an examination
of the distinction between primary and secondary qualities,
and that between substantive and adjective; two of "the
many ways in which we try to understand the universe."
However fundamental, or however footless for such an
attempt these ways be deemed, is a matter of but little
moment, in Bradley's view. "From whatever point we had
begun we should have found ourselves entangled in the
same puzzles, and have been led to attempt the same way
of escape." This way of escape is afforded by the principle

* Unless it is otherwise indicated, all numerical references are to
Appearance and Reality, and edn., Oxford University Press.

15

Bradley's Dialectic

of the internality of relations. Bradley arrives at this
principle in chapter III of Part I. In Part II of Appearance
and Reality the internality of quality and relation turns out
to be the Janus-faced principle at once of Appearance and
of degrees of Truth and Reality.

Bradley takes it that the distinction between primary and
secondary qualities is made in order to reconcile changing
appearances with a belief in a changeless reality. This
distinction is " easily disposed of." For those who make it
admit that qualities appear only as conditioned in their
existence. How, then, may we expect to say anything at all
about qualities alleged to be unconditioned, as are those
designated "primary" and " independent"? To call the
changing and relative qualities "appearance", in contrast
with a reality that is alleged to be permanent and absolute,
is to accomplish nothing at all: for nothing is actually re-
moved from existence by being labelled "appearance".
"What appears is there, and must be dealt with; . . . ," 12
Moreover, extension, taken as devoid of secondary quality,
is inconceivable: it "cannot be presented, or thought of,
except, as one with quality that is secondary." 14 The alleged
reality of extension, as something independent of the
qualities called secondary, is thus seen, Bradley thinks, to
derive from a distinction in theory that is repelled by fact.

Then again, the line of reasoning by which the qualities
called secondary are alleged to deserve the name holds
with no less force of those qualities that are distinguished
as being real in their own right. This means, to give the
main point here but one illustration, that the example of
the square tower which looks round at a distance, illustrates
the conditional character of perceived extension quite as
well as the example of the piece of wax illustrates the in-
constancy of the temperature, odour, and colour of wax.
Taken as a serious attempt at even a partial view of experi-
ence, any doctrine on which anything like the distinction

16

The Dialectic of Relation and Quality

between primary and secondary qualities would be a valid
distinction between reality and appearance can hardly
survive even a light attack of Berkeleian criticism.

The distinction between the substantive and the adjective,
to which Bradley turns next, also is found to be of no avail
as a satisfactory way of interpreting our experience. " We
may take the familiar instance of a lump of sugar. This is
a thing, and it has properties, adjectives which qualify it.
It is for example, white, and hard, and sweet. The
sugar, we say, is all that; but what the is can really mean
seems doubtful." 18 A lump of sugar plainly is not any one,
nor yet any set, or group, of its qualities. It is neither
sweetness, taken alone, nor is it that quality and whiteness,
and hardness, etc., taken in bare conjunction. 16 ' 17 The
view that "the secret of the thing" 16 lies in the co-existence
of these adjectives "in a certain way", 16 wholly fails to dis-
cover any real unity existing in and throughout them.
Either the nature of a lump of sugar is exhausted in the
sum of the co-existing adjectives which are in question, or
that nature involves a substantive which is their persisting
unity. Yet where, or how, is this substantive to be found ?
A substantive would be that which may not be a predicate.
And whatever appears, it would seem, may be predicable
of something else.

So much as this may suffice to indicate vyhy Bradley
concludes here that "we can discover no real unity existing
outside these qualities, or again, existing within them." 16
He proceeds to suggest an alternative to any further search
for a satisfactory conception of substance. On this alternative,
the qualities which constitute a thing are held together, not
by a substance in which they would inhere, but rather by
their relations. "One quality, A, is in relation with another
quality, B. But what are we to understand by 'is' ?" 17 Surely
not that "being in relation with B" is (identical with) A.
Bradley does not here distinguish between the "is" of

17 B

Bradley's Dialectic

identity and the u is" of predication. For where there is
no degree of identity in the constituent terms of a judgement,
those terms stand together in bare conjunction. As so
arranged, the terms would be related quite externally, the
one to the other, were such an external relation possible
at all. Since, as we shall find Bradley urging in detail, any
external relation whatever can only fail to relate its alleged
terms, the "is" of predication, taken as a (verbal) sign of
conjunction, is ruled out by the rejection of all merely
external relations.

Yet, in the judgement, "A is in relation with B", we surely
do not mean that "being in relation with B" is A. Neither
do we mean that "being in relation with B" is different from
A. "And we seem unable to clear ourselves from the old
dilemma. If you predicate what is different, you ascribe to
the subject what it is not\ and if you predicate what is not
different, you say nothing at all." 17 Any conception of the
thing on which it is a substantive complex of related qualities
will thus fail to elucidate both the way in which the qualities
may be predicated of relations, and relations may be predi-
cated of qualities.

The way to resolve this dilemma, it may be suggested,
is an evident and simple one: let us regard the relation of
inherence as an external relation. In thus taking "is" to be
the sign of a relation that is separate from, or independent
of, its terms, we cease to regard that relation as being in any
sense inherent in the terms it is said to relate. "Let us abstain
from making the relation an attribute of the related, and
let us make it more or less independent. 'There is a relation
C in which A and B stand; and it appears with both of them.'
But here again we have made no progress. The relation G
has been admitted different from A and B, and no longer
is predicated of them. Something, however, seems to be
said of 'this 5 relation C, and said, again, of A and B. And
this something is not to be the ascription of one to the

18

The Dialectic of Relation and Quality

other. If so, it would appear to be another relation, D, in
which C, on one side, and, on the other side, A and B,
stand. But such a makeshift leads at once to the infinite
process." 17 ' 18 On this alternative, the phrase "stand in"
takes the place of the term "is". Now we say not, "A is in
relation with B," but rather, "A and B stand in relation C."

And so the question is raised as to the meaning of the
phrase "to stand in" as that phrase is used in this state-
ment. For, to say that A and B "stand in" a relation G,
would seem to say something about A, and B, and C. "And
this something is not to be the ascription of one to the
other." 18 For, by hypothesis, any conception of a relation
as inhering in its terms is excluded on this alternative. Still,
it may be urged, the fact is that the phrase in question adds
nothing to the meaning of the statement "A and B stand in a
relation C." For that statement refers to no more than the
relational complex A C B. That C relates A and B is a
matter of fact which neither can nor need be explained.

On such a view of the matter as this, the relation G would
be conceived of as quite independent of its terms. For to say
that those terms (or any others) stand in that relation (or in
any other), we are told, is not to say anything about A, C, and
B: it is to say merely that we have A, and G, and B: where
"and" is the enumerative, not the conjunctive "and". This
being so, C is not even conjoined with A and B. It is quite
independent of A and B. And as thus independent, the
so-called relation C might as well be enumerated with
other independent relations, such as R, and R'. For, as
independent, the relation G requires no terms that it may
be a relation. To such a relation, terms would be accidental:
it could have no terms at all and still be a relation; for it is
something that has its being independently of all terms.
Yet this something is merely verbal. For to speak of a relation
that is without terms is to speak of a relation that relates
nothing, and so is not a relation at all.

19

Bradley's Dialectic

But why take it, one may ask, that the relational complex
A C B is exhaustively described in and through an enumera-
tion of its constituents, regarded as merely independent?
To be sure, it may be said, A, and G, and B are independent
of one another: yet they are so only in the sense that each
one is what it is in its own right. The term A is A absolutely,
rather than relatively; and that is all it means to say that
A is A independently of all else. The intrinsic self-identity
of A, of C, and of B, is not their separation from one another.
For in the relational complex A C B you have A and B
related by C. In other words, in the enumeration, A, and
C, and B, the "and" is the conjunctive, not merely the
enumerative "and".

Such an objection only raises all over again the question
which we have been trying unsuccessfully to evade. For we
must either deny or else affirm that this conjunctive "and"
contributes to constitute the relational complex in question.
If our attitude be that of denial, we are left with the bare
complex A C B, whose constituents can only be enumerated.
But if, on the other hand, we affirm that A C B are con-
joined, then they are conjoined somehow, arid not merely
compresent.

Moreover, to say that the constituents of the relational
complex A C B are conjoined, is to say something more than
that there is the relational complex A C B; for it is to say
something about those constituents, viz., that they are con-
joined. "Something, however, seems to be said cf this rela-
tion C, and said, again, of A and B. And this something is not
to be the ascription of one to the other. If so, it would appear
to be another relation, D, in which C, on one side, and on
the other side, A and B, stand. But such a makeshift leads at
once to the infinite process. The new relation can be pre-
dicated in no way of C, or A and B; and hence we must have
recourse to a fresh relation, E, which comes between D and
whatever we had before. But this must lead to another, F;

20

The Dialectic of Relation and Quality

and so on, indefinitely. Thus the problem is not solved by
taking relations as independently real." 18 The relation C,
regarded as quite independent of its terms, A and B, is then
viewed as being merely compresent with them. As thus
independent, C does not require A and B, or any other
terms, that it may be a relation. Hence C would be a relation
were it without any terms at all. And yet a "relation" that
relates nothing is surely no more than a mis-use of the word.

The attempt to .take relations as real independently of
their terms, while at the same time their "connexity" is held
to be their being joined with their terms, also is futile. For
either this "being joined with" is something more than the
independent relation, or else we are back where we were
before. Yet it is precisely this "something more" for which
the notion of independent relations cannot provide. And
so, in a foredoomed attempt to supply the deficiency,
another independent relation is posited in the relational
complex. But this too can only fail to relate; hence the infinite
process, and "we are forced to see, when we reflect, that
a relation standing alongside of its terms is a delusion". 18
The attempt to understand experience through the notion
of the thing and its qualities breaks down, no less when the
thing is taken as a set of qualities related by an independent
relation, than when a thing is taken to consist of qualities
inhering in a substance, of which no positive definition,
or description, can be given.

In view of the fact that Bradley's argument to his rejection
of external relations has been taken for and quoted from
as his positive theory of relations (as, for example, by Dr.
Broad and Dr. Ewing), it may be well at this juncture to
point out that such is not the case. The conclusion of the
argument in question is negative, and the text of it is part
of the chapter on Substantive and Adjective, not of the
following chapter on Relation and Quality, in which the
basic content of Bradley's theory of relations is brought out.

21

Bradley's Dialectic

Bradley thinks it will be "evident" that the problem dis-
cussed in the chapter on substantive and adjective "really
turns on the respective natures of quality and relation." 21
The notion of a substantive and its adjectives proves to be of
no avail: that of a relational complex raises the question,
what, if not a relation independently real, may relate the
qualities of a complex ?

At the outset of his elucidation of the meaning of "rela-
tion" and "quality" for his essay in metaphysics, Bradley
foreshadows his main conclusion in that regard. "Our con-
clusion briefly will be this. Relation presupposes quality,
and quality relation. Each can be something neither together
with, nor apart from, the other; and the vicious circle in
which they turn is not the truth about reality." 21 In follow-
ing Bradley's main arguments to this conclusion, it may make
our course the easier if we remind ourselves that Bradley,
in his Logic., is constrained to reject any view of identity as
absolute, and to adopt a theory of identity in difference.
That conception of identity is to be the topic of a subsequent
chapter. Nevertheless, the fact that Bradley denies the Laws
of Identity andTT^iPContradiction, as those laws are under-
stood on an Aristotelian Logic, is fundamental to the nature
of his dialectic. For that reason, it may be well to consider
the point, albeit briefly, at this juncture.

In the Logic, Bradley writes: "The principle of Identity
is often stated in the form of a tautology, C A is A 5 . If this
really means that no difference exists on the two sides of the
judgement, we may dismiss it at once. It is no judgement
at all. As Hegel tells us, it sins against the very form of judge-
ment; for, while professing to say something, it really says
nothing. It does not even assert identity. For identity without
difference is nothingat all. It takes two 15 makeTEe^ same,
and the least we can have is some change of event in a self-
same thing, or the return to that thing from some suggested
difference. For, otherwise, to say 'It is the same as itself

22

The Dialectic of Relation and Quality

would be quite unmeaning. We could not even have the
appearance of judgement in 'A is A', if we had not at least
the difference of position in the different A's; and we can not
have the reality of judgement, unless some difference actually
enters into the content of what we assert."*

Any genuine judgement, we are then told, will assert
unity in diversity, not the barren identity of the tautology
"A is A". Judgement may not exist in the absence of either
the differences or the unity. Unless the different constituents
of a judgement are in some sense united in it, there is no
judgement but, at best, an association of ideas. And unless
it be differences that are said to be in union there is no judge-
ment, but rather the utterance of mere tautology, and so
no movement in thought at all.

If the formula of judgement may not be A is A, no more may
it be A is B. For in this latter form, we assert (it is assumed)
that A is identical with, or strictly the same as, B. Therefore
we are confronted by a dilemma: on the one alternative we
assert about A nothing at all; on the other, we say that A
is what it is not viz., B.

The error that sustains this dilemma and from which it
derives may be brought to light by considering the nature of
the contradictory. "We have to avoid, in dealing with Con-
tradiction, the same mistake that we found had obscured
the nature of Identity. We there were told to produce
tautologies, and here we are by certain persons forbidden to
produce anything else. C A is not not-A' may be taken to
mean that A can be nothing but what is simply A. This is,
once again, the erroneous assertion of mere abstract identity
without any difference."! That assertion is erroneous
because it rests upon the assumption of pure negation. In-
deed, that false assumption is at the basis of the entire
erroneous matter. For the assumption that A simply is not
B brings us ineluctably to the conclusion that A is A, irrespec-
* Logic, and edn., p. 141. f Ibid. p. 146.

23

Bradley's Dialectic

live of its relations. And that conclusion entails the above-
mentioned dilemma.

For this reason, among several others, there can be no
place in logic for the notion of mere negation. "The con-
tradictory idea, if we take it in a merely negative form, must
be banished from logic. If not-A were solely the negation
of A, it would be an assertion without a quality, and would
be a denial without anything positive to serve as its ground.
A something that is only not something else, is a relation
that terminates in an impalpable void, a reflection thrown
upon empty space. It is a mere nonentity which can not be
real.' 5 * In short, mere negation is groundless verbiage.
Every significant negation presupposes a positive ground.
We can not and do not deny a predicate of a nothing; rather,
we deny it of a subject on the ground that this subject pos-
sesses a quality which is incompatible with the predicate
of our negative judgement. \x

Thus, for Bradley, the negative judgement does not express
bare otherness, or mere negation, between the terms of
which there would be no middle ground. Hence Bradley
denies that the logical form of the contradictory, within
which no middle term is possible, and the logical form of the
contrary, within which a middle term is always at least
possible, are distinct; and he largely identifies them.
"But then this positive ground, which is the basis of negation,
is not contradictory. It is merely discrepant, opposite, incom-
patible. It is only contrary"*

The notion of the contradictory, as that term is customarily
understood with reference to the square of opposition, must
be banished from logic. The contradictory means very largely
what is meant by the contrary. No rule that would compre-
hend the scope of the contrary could be formulated. For
"contrary opposition is indefinitely plural. The number of
qualities that are discrepant or incompatible with A, can-

* Ibid., p. 123.
24

The Dialectic of Relation and Quality

not be determined by a general rule. It is possible of course
to define a contrary in some sense which will limit the use
of the term; but for logical purposes this customary restric-
tion is nothing but lumber. In logic the contrary should be
simply the discrepant."* The contradictory is in the main
one with .the contrary, and that is simply the discrepant.
Thus, wherever Bradley writes of a relational situation as
being self-contradictory he means not what a reader steeped
hi the Aristotelian tradition would erroneously take him to
mean; rather, he means that the terms and relations in
question are respectively contraries, and that by virtue of
these contrarieties the relational situation i self-discrepant.
It may help to avert another misunderstanding if, at this
juncture, we pause to ask how we are to take the term
"appearance", as used by Bradley. Is it the name of a static
veil of qualities and relations which stands between us and
the really real ? Now it will be recalled that there is a tradi-
tion on which what is in becoming is appearance and appear-
ance is what is in becoming. Readers of Bradley will recall
how he writes of an ' 'infinite process" in relations; of a
"principle of fission which conducts us to no end," 26 of
relations that "break out", and "fall between" qualities in
appearance; and of a "what" being "in collision" with
another such. Again, appearance is "taken up into" the
eternal Absolute. These samples of Bradley's idiom could
easily be multiplied ad nauseam. If one were to take it that,
for Bradley, "appearance" is the name of a static veil, one
would have to take this idiom as mere metaphor. In the
static there may be neither process, infinite or otherwise;
nor fission, whether endless or not. The static excludes
whatever might "break out", or "fall" no matter where, or
be in a "collision". And to take Bradley as writing meta-
physical irrelevance; would be to beg the question of the
meaning of his writing altogether. Appearance is not static:

* Ibid., p. 124.

25

Bradley's Dialectic

rather appearance is in becoming; in it there is, Bradley
finds, an infinite process > an unending fission, in and through
which relations do* break out and fall between qualities. As
contrasted with the eternal, self-identical Absolute, appear-
ance is what is becoming.

On the view that identity is absolute, the question as to
how self-consistent judgements about becoming may be
made is, to be sure, an old one. If A is A absolutely (rather
than relatively), it is hard to see how A could become Y.
For in the course of the process A, Ay, Ayy, . . . Y, there
would finally be a point at which A is no longer A, and is
not yet Y. On a view of identity as absolute, there could be
no middle term in and through which the contradiction in
judgements about change might be sublated. In Hegel's
smaller Logic the initial middle term is becoming; i.e. the
synthesis of being and non-being. * 'Being is the notion
implicit only": no case of mere being is to be taken alone,
but only as the contrary of non-being. The resulting alterna-
tion in thought between being and non-being discloses itself
as a process of becoming determinate, and so we find these
primary contraries taken up into the category of Becoming.

Bradley not only rejects Hegel's "ballet of bloodless
categories", he finds the dialectic of contraries incompetent
to achieve a self-consistent elucidation of the two contraries
that are for him the characteristics of Appearance everywhere
and always; viz., quality and relation. Now, for reasons to
which we proceed, should we attempt to illustrate Bradley's
conception of relation by thinking of a relation defined as
a universal that requires at least two particulars for its
illustration, our attempt could only be irrelevant. And were
a sense-quality, thought of as self-identical, taken to illustrate
Bradley's view of quality, this again would be a mistake.
For self-identical relations and qualities may not become;
they are what they are. Therefore they may not be taken
as illustrations of Bradley's view of relation and quality.

26

The Dialectic of Relation and Quality

On that view, "quality" is the name of any moment of
experience wherein immediacy is dominant and differentia-
tion is recessive. The main point here may be illustrated
by the process of cell-fission. If we attend mainly to the
new cells that are thus emerging, what is then most immediate
or qualitative will be dominant in our experience. If, on the
contrary, our attention emphasizes the mediation or differen-
tiation thus in process, then relation will be dominant,
while quality is recessive.

Thus conceived of, "Qualities are nothing without rela-
tions. In trying to exhibit the truth of this statement, I will
lay no weight on a considerable mass of evidence 5 ': 21 all that
evidence, indeed, which goes to show how qualities are
varied in fact by changes in their relations. "But I will not
appeal to such an argument, since I do not see that it could
prove wholly the non-existence of original and independent
qualities." 21 ' 22 Instead of an appeal to extensive matters of
fact, Bradley proceeds to offer a demonstration a priori.

Any attempt to arrive at a relationless quality by abstrac-
tion could only fail. The process of abstraction is a process
of differentiation, and so that process, like any other, is
relational.

Again, we should be thwarted were we to "appeal to a
lower and undistinguished state of mind, where in one feeling
are many aspects. ... I admit the existence of such states
without any relation, but I wholly deny there the presence
of qualities". 22 For these felt aspects are not qualities if they
are quite undifferentiated, and if they are differentiated,
then, by that very fact, they are related. "In short, if you go
back to mere unbroken feelings, you have no relations and
no qualities. But if you come to what is distinct, you get
relations at once." 22 Where there are no distinctions in
feeling, there are no qualities: where there are qualities,
there are distinctions; and, by that very fact, relations.

To this, Bradley thinks it will be answered that though

27

Bradley's Dialectic

the process of distinguishing qualities be relational, still that
process of making these distinctions is not essential to the
qualities thus distinguished. It will be acknowledged, Bradley
thinks, that "as you say, what is different must be distinct,
and in consequence, related." 22 ' 23 But, it will be objected,
the relation that is the process of making distinctions does
not belong to the qualities that result. They are differentiated
by relations which depend in no wise on the respective
characters of the qualities they relate. And so these qualities
and their relations are in no sense determined or conditioned
by each other. Moreover, these qualities, though different
from each other, are different intrinsically, each one in its
own inalienable right.

For such reasons as these, it may be urged that relation-
less qualities may exist in abstraction. For the process of
abstraction, even though it be relational, is not essential
to the quality abstracted. And the quality itself is not altered
by being abstracted. "But such an answer depends on the
separation of product from process, and this separation seems
indefensible. The qualities, as distinct, are made so by an
action which is admitted to imply relation. They are made
so, and, what is more, they are emphatically kept so. And
you cannot ever get your product standing apart from its
process. Will you say, the process is not essential ? But that
is a conclusion to be proved, and it is monstrous to assume
it." 23 The force of this reply begins to be felt as Bradley goes
on to develop his argument that the being of quality implies
relation.

Bradley assumes that a difference is a relation and that
a relation is a difference. His main point here is that where
there are different qualities, there are qualities related by
their differences. "For consider, the qualities A and B are to
be different from each other; and, if so, that difference must
fall somewhere. If it falls, in any degree or to any extent,
outside A or B, we have relation at once. But, on the other

28

The Dialectic of Relation and Quality

hand, how can difference and otherness fall inside ? If we
have in A any such otherness, then inside A we must dis-
tinguish its own quality and its otherness. And, if so, then
the unsolved problem breaks out inside each quality, and
separates each into two qualities in relation. In brief,
diversity without relation seems a word without meaning." 24
Either the difference between A and B, in virtue of which
they are distinct, "falls outside A or B," thus to relate them;
or, that difference ' 'falls within" A and B respectively, thus
to differentiate each one of them within itself. On the latter
alternative, the moment of relation would break out within
A, and within B. Hence the difference that differentiates A
and B must fall "outside", or "between" 24 them; and thus
"we have relation at once." 24

For exstmple, in the very incipience of a process of cell-
fission the incipient differentiation "must fall somewhere".
And it falls "outside" or "between" the incipient cells in
the observable sense that it is their differentiation. As soon
as we notice this, then (on the assumption that differentia-
tion, or difference, is relation), we are aware of relation at
once. For we are then aware of the difference in virtue of
which the incipient qualities are differentiated, or related.
Were a quality without relations in this sense of "relation",
it would be in no wise different from anything else, and so
would fail to be a quality at all.

To be sure, no fixed line between a differentiation, and
what is thus differentiated, can be drawn and maintained in
experience. For any relation, and any quality, will be in
process. As the incipient qualities become more and more
determinate, their differentiation alters; and, as their
differentiation becomes the more marked, the qualities
alter. "Hence the qualities must be, and must also be related.
But there is hence a diversity which falls inside each quality.
Each has a double character, as both supporting and as
being made by the relation." 28 In order that a quality may

29

Bradley's Dialectic

be distinct, it must be differentiated from other qualities.
This differentiation is no separate relation: rather, it con-
tributes to constitute what it differentiates. Thus, in so far
as A i quality, A is not relation; and yet, that it may be
distinct, A must be both itself and its differentiation. "A is
both made, and is not made, what it is by relation; and these
different aspects are not each the other, nor again is either A.
If we call its diverse aspects a and a, then A is partly each
of these. As a, it is the difference on which distinction is
based, while as a it is the distinction that results from
connection. A is really both somewhere together as A
(a a). But (as we saw in chapter II) without the use of a
relation it is impossible to predicate this variety of A. And,
on the other hand, with an internal relation A's unity disap-
pears, and its contents are dissipated in an endless process of
distinction. We, in brief, are led by a principle of fission
which conducts us to no end." 26 Without a relation, A
would be undifferentiated, and so would be nothing at all.
With a relation, A is at once the a that is differentiated, and
the a that is the differentiation. Thus "A is partly each of
these". Neither a nor a is the other, "nor again is either A";
for oc is what is differentiated, while a is the differentiation.
And this differentiation, a, is essential to the a that it
differentiates.

That is why "A is both made, and is not made, what it is
by relation. ... It may be taken as at once condition and
result, and the question is as to how it can combine this
variety. For it must combine this variety. For it must com-
bine the diversity, and yet it fails to do so". 26 A must be
at once a, the aspect differentiated, and a, the aspect of
differentiation. Without a there is nothing differentiated,
and so no quality: without a there is no differentiation and
so nothing at all.

A is at once itself a, and its differentiation a: and this
identity of A, which is a, implies its differentiation, which is

30

The Dialectic of Relation and Quality

a. Thus the quality A is the moment of immediacy a, and
the moment of mediation a, by which that moment of
immediacy is differentiated from other qualities and
relations. In brief, that A may be at all, it must be at once
what falls within itself, and what falls between itself and
all else. Thus A both is itself, and is transcendent of itself.
No quality, then, will be self-sufficient or self-contained;
no quality will be self-identical. For that in virtue of which
a quality is distinct will fall beyond that quality. Yet, at the
same time, that differentiation will be essential to that
quality. For without that differentiation, the quality would
not be differentiated, as it is differentiated, and thus would
not be the quality it is. Hence no quality is wholly self-
consistent; and that means, for Bradley, that no quality is
wholly intelligible by the relational way of thought that is
ineluctably ours.

Thus, "we have found that qualities, taken without
relations, have no intelligible meaning. Unfortunately,
taken together with them, they are equally unintelligible". 26
So far, we have noticed that relations without qualities
would be relations that relate nothing, and so fail to be
relations at all. And we have noticed that qualities without
relations would be undifferentiated, and so would be not
many, but one. That there might be a plurality of qualities
without relation is impossible. Yet, taken together with its
relations, we have seen that no quality is wholly self-
consistent, or intelligible.

The same difficulty appears when experience is "taken
from the side of relations. They are nothing intelligible,
either with or without their qualities". 27 As relations apart
from qualities are a delusion, so together with quality they
are in no finite context completely intelligible. For that a
relation may differentiate its terms, it must * 'penetrate
and alter" them, and so be implicated in their respective
natures. Yet, that this differentiation may not disappear

3 1

Bradley's Dialectic

altogether, it must "fall" to some extent "between" the
qualities which it differentiates, and so relates. A relation
thus involves within itself a contrariety: it must be at once
implicated in and transcendent of its qualities. And so
"again we are hurried off into the eddy of a hopeless process,
since we are forced to go on finding relations without
end." 28 Then again, and on the other hand, in so far
as a relation or differentiation is implicated in its quali-
ties, it does not fall between them. In this respect a differen-
tiation fails to be a relation at all. Yet in so far as a
differentiation falls between qualities, it is outside them
both, and so again fails to relate them. 27 ' 28

Hence no moment of differentiation may be absolute.
In a process of fission, no absolute distinction is to be found
between the moments of quality and the moments of
differentiation. There is "a diversity which falls inside each
quality. Each has a double character, as both supporting and
as being made by the relation." 26 Likewise, each relation
has a double character, as both supporting and as being made
by its terms. Qualities taken without relations or as absolute,
and relations as absolute or without qualities, are both,
Bradley finds, inconceivable. Yet qualities, taken as moments
of immediacy that are at once made by and essential to their
differentiations, are in no case absolutely self-identical;
for they are differentiated by their relations. And so their
respective identities are relational, not absolute. Likewise,
relations taken as moments of differentiation that are at once
implicated in and transcendent of their qualities, are in no
case absolute (or separate) relations, for they involve within
themselves that infinite regress in relational identity which is
the principle and content of degrees of truth and reality. *

In a posthumously published essay, Relations, Bradley
reverts to his point about relations being less than wholly
self-consistent and intelligible. "A relation (we find) holds
* Collected Essays, pp. 634-5.
32

The Dialectic of Relation and Quality

between its terms, and no term (we find) can itself be or
become a 'between. 1 On the contrary, in order to be related,
a term must keep still within itself enough character to make
it, in short, itself and not anything diverse. And again, while
the relations are not the terms and the terms are not the
relations, neither the terms nor the relations can make that
whole, in which nevertheless we find them. For the terms
and the relations (we have seen) cease as such to exist,
unless each maintains itself against whatever is not itself
but is outside. And the attempt to find the required unity
and totality in the terms and the relations taken somehow
together must end obviously in failure."* The attempt in
question can end only in failure because neither the terms
of a relational situation, nor their relation, may be self-
consistent, or self-identical. For each one of the two terms
will be a moment of immediacy, while, at the same time, it
will transcend itself in being continuous with the differen-
tiations by which these terms are differentiated.

Likewise, these differentiations will be internal to, or
confluent with the terms they relate; while, at the same time,
they will fall between and thus transcend them. This is
why "relation both is and is not what may be called the
entire relational situation, and hence in this respect contra-
dicts itself." A relation is the whole relational situation in
the sense that it is what differentiates the qualities of that
situation. These qualities, were they differentiated otherwise,
would be different qualities. Conversely, were the qualities
of the situation different, their differentiation also would
be different. As thus determining the qualities of the situa-
tion, and so its own character as a differentiation, "a relation
to be actual cannot itself be less than all and everything that
makes the entire relational fact".| For it is in virtue of the
relation that the qualities which it differentiates are the
distinct moments in process which they are. In a case of
* Ibid., pp. 634-5. t Ibid., p. 636.

33 c

Bradley's Dialectic

cell-fission, the qualities which are being differentiated are
confluent with their differentiation; as, likewise, that relation
is continuous with those qualities. At any two points in the
process of fission, this relation may be marked off from its
qualities, and its qualities thus will be marked off from their
relation; but no such distinctions may be final for theory.
For the distinctions in question would themselves be differen-
tiations. These relations would break out, on the one hand,
between the moments of fission which we had marked off
as qualities, and, on the other, between what of the process
would then be marked off as the differentiation of those new
qualities. This "what" thus would become a quality in its
own right: for it would itself be differentiated by the distinc-
tions in question. Plainly such distinctions might be multi-
plied indefinitely within a process of becoming. Since to make
distinctions ipso facto is to relate, there can be no limit to the
relational constituents of a relational situation.

Yet "This on the other hand must be denied. For a rela-
tion is not its terms, but, on the contrary, it is between them.
And though the terms may 'enter into the relation 9 , yet, if
they were nothing beyond it, they obviously would no longer
be terms."* In this sense, relations pervade and determine
the character of that partial whole; and thus they may be
said to be not less than "all . . . that makes the entire
relational fact." Still, in no case is the differentiation identical
with the differentiated. The qualities differentiated are to be
sure continuous with their relation. Yet that either quality
or relation may be at all, each must be distinct, y 7

That there may be qualities at all, distinctions must occur
or be made, and no distinction in appearance may be
ultimate. Once the differentiation is marked off, it is itself
thus distinguished in being singled out; and hence between
it and the initial qualities, fresh relations break out with
their qualities.

* Ibid., p. 636.
34

The Dialectic of Relation and Quality

No distinction, no differentiation in process, may be
self-sustaining or absolute. For a differentiation will be at
once continuous with its qualities and distinct from them.
In virtue of that distinction, and no matter where in the
process it may break out, or where it be discriminated, a
fresh relation, with its own qualities, then appears. This new
relation, though continuous with its own qualities, neverthe-
less is different from them. And so, again, there appear fresh
relations with their new qualities.

The notion of independent relations, we have seen,
yields an "infinite process" in relations that do not relate:
the dialectic of relations which differentiate their terms
exhibits a process wherein neither the differentiation itself,
nor the quality differentiated, is absolute in its own right,
or self-identical. Hence the conclusion that no relational
situation is wholly self-consistent, or intelligible.

Thus we may come to see that "Every relation (unless our
previous inquiries have led to error) has a connexion with
its terms which, not simply internal or external, must in
principle be both at once".* A relation must be internal
to its terms in the sense that, as their differentiation, it con-
tributes to constitute their qualitative character: yet, if it is not
to disappear altogether, a relation must, to some extent, "fall
between" its terms. And in that sense a relation is at once
internal and external to the qualities which it differentiates.

It is sometimes said that the "internality of relations"
means for Bradley simply this: any alteration in a relation
ipso facto is an alteration in its terms, and any alteration in
either term ipso facto is an alteration in the relation and in
the other term. To be sure, it would be difficult (without
achieving flat irrelevance) to take Bradley as meaning less
than this. Since a relation is what differentiates its qualities,
those qualities are what they are in virtue of that differentia-
tion. That is why any alteration in the relation ipso facto

* Ibid., p. 641.
35

Bradley's Dialectic

is an alteration in the qualities thus differentiated. And
since the differentiation is what it is in virtue of the qualities
which it differentiates, an alteration in either of these quali-
ties ipso facto is an alteration in their differentiation, and in
the other quality.

Yet, if this is the least that may be said in this connection,
so much is far from enough. For such a bare statement
of the matter might be taken badly amiss. It would allow
one to think that for Bradley relations are "absolutely and
merely internal"; i.e. exhausted in the character of their
terms, as is the philosophical relation of resemblance for
Hume. A difference or a resemblance which was absolutely
internal would be wholly intrinsic to the different, or the
resembling, terms themselves. A, in being A, would be dif-
ferent from B; and B, in being B, would be different
from A, not in virtue of any differentiation or process
that partially fell between them, but rather in virtue of the
intrinsic character of A and of B. Yet in being intrinsically
what they are, A and B are self-identical absolutely; and
their being different is not a differentiation at all: rather,
their being different is exhausted in the respective beings
that are the intrinsically different qualities A and B.

The several ways in which Bradley elucidates his theory
of relations, in the course of the dialectic of Relation and
Quality, surely ought to suffice to make it clear that this is
not his view of the matter. Moreover, in his unfinished essay
on Relations, Bradley is at some pains to tell us what he
considers to be the meaning of "external relation", and
what he does not mean by the phrase "internal relation".
"What should we mean (I will ask first) by a relation asserted
as simply and barely external? We have here, I presume,
to abstract so as to take terms and relations, all and each,
as something which in and by itself is real independently.
And we must, if so, assume that their coming or being
together in fact, and as somehow actually in one, is due in

36

The Dialectic of Relation and Quality

no way to the particular characters of either the relations or
the terms. From neither side will there be anything like a
contribution to, or an entrance into, the other side or
again to, or into, that union of both which we experience
as a relational fact. Undeniably the fact is somehow there,
but in itself it remains irrational as admitting no question
as to its 'how' or 'why'. Or, if you insist on a reason, that
would have to be sought neither in the terms nor the rela-
tion, but in a third element once more independently real
and neither affecting, nor again affected by, either the
relation or the terms. This, I suppose, is the way in which
relations have to be understood, if you take them as external
merely and also as ultimately and absolutely real."* As so
conceived of, external relations would merely coexist with
their terms. Moreover, that such a relation should coexist
with any terms at all, rather than simply exist without
terms altogether, would be a merely fortuitous happening.
Then again, this so-called relation might as well be taken
along with other termless relations, once the absurdity of a
relation without terms is admitted. Yet the relational situa-
tion, A C B, is a unitary fact. And when A and C and B are
taken respectively as external to each other, they do and
may afford no explanation of the unity in which, neverthe-
less, they exist. Any resort to an intermediary, itself external,
by which to relate A and C, and B and C, fails to explain
the unity and leads at once to an infinite regress in relations
that fail to relate anything at all.

Those of a positive turn of mind may object that in all
this there is no unity that requires explanation. They may
urge that the unity of A C B is exhausted in the coexistence of
A and C and B. There are the terms; there is the relation;
they coexist; and that is all there is to it. Yet, on any such
view of the matter, the relation C still might be taken without
any terms whatever. And to speak of such an entity as a

* Ibid., p. 642.

37

Bradley's Dialectic

"relation", would surely be to mis-use that word. "What
(I ask next) should, on the other hand, be meant by a
relation viewed as absolutely and merely internal? You,
I presume, still in this case would continue to take the terms
each one as, so far, in and by itself real, and as independent
absolutely of any whole that could be said to contain them.
And you would go on to attribute to the particular character
of the terms, as so taken, some actual relation or relations
which you find, as you say, to fall between them. Something
like this, I suppose, is or ought to be meant by a relation
which is asserted to be real ultimately and internal merely."*
Such a view of relations Bradley rejects as "ludicrous"
when ascribed to his view of the matter.

One fundamental reason for this rejection derives from
the theory of relations that is explained in chapter III of
Appearance and Reality. There we find that a relation differen-
tiates its terms and thus contributes to constitute their
character; as in a process of fission, the growing differentia-
tion is confluent with the developing cells thus related.
Therefore, when there is alteration in the relation, there is
ipso facto alteration in the cells or qualities that are differen-
tiated by the relation. For it is in virtue of this differentia-
tion that the qualities thus related are respectively what
they are.

A quality would be "ultimately and absolutely reaP'j
were its nature quite self-contained. Such an entity would
be what it is in virtue of itself alone : and so it would be itself
absolutely, without relation to anything beyond its own
absolute nature. If two such entities were posited, it would
follow, in Bradley's view, that any relation between them
would be wholly exhausted in the respective characters of
the two qualities posited. "Relations would be merely
internal if, the terms being taken as real independently, each
in itself, the relations between them (as a class, or in this or
* Ibid., p. 642. t

38

The Dialectic of Relation and Quality

that particular case) in fact arose or were due merely to the
character of the terms as so taken."* This sentence is quoted
from a note on the point in question made by Bradley for
an earlier draft of his essay on relations. It expresses briefly
Bradley 's lengthier statement of what he means by "a
relation viewed as absolutely and merely internal". To that
lengthier statement he adds: "The idea, I would add, that I
myself accept any such doctrine as the above seems to myself
even ludicrous. And to whom, if to any one, it should be
attributed in fact, I will not offer to discuss. In any case,
to assume it as the necessary alternative, when the mere
externality of relations is denied, is (I submit) an obvious, if
perhaps a natural, mistake. "t

The mistake here consists in taking the terms of a relation
as if those qualities were real absolutely, each one in its
own right. Such a view is in flagrant opposition to the
doctrine that is made out in the chapter Relation and Quality.
To be a quality at all, is to be differentiated by a relation
which thus contributes to constitute the quality which it
differentiates. The identity of that quality implies this
difference. But not this difference alone. In the relational
situation A C B, the relation C differentiates A and B. That
relation is implicated in their respective identities, and those
qualities A and B imply that difference: were it otherwise,
they too would be different. And the qualities A and B do not
exhaust the content of experience. They will differ from
qualities D, E, . . . n, as well as from the relation by which
these other qualities are differentiated among themselves,
and from all else. And so with B, D, E, . . . n: hence
"identity implies difference".

It may be objected that this well-known dictum is but

a truism, almost obvious and no less empty. To be sure,

identity implies difference: for A must be numerically

different from other items, in order that A may be an item

* Ibid., p. 665. t Ibid., pp. 642-3.

39

Bradley's Dialectic

at all that could be identified by attention as that item.
Such an objection is not wholly irrelevant, for it affords an
example of a relation that is "merely internal". The items in
question are different simply in virtue of themselves alone.
Thus their difference from each other is merely internal
to each one of them.

Again, if "resemblance" be taken to mean any case of a
qualitative identity distributed in enumerably different cases
of itself, then we have another illustration of a merely
internal relation. For the resemblance here is the qualitative
identity which is exhibited in two cases of itself. Were
numerical difference, and resemblance, in this sense of the
term, regarded as relations, the terms of those relations would
be self-identical. For the numerically different items are
what they are in virtue of each respective item itself; and
the several cases of resemblance are qualitatively identical
absolutely, without reference to anything beyond themselves.
Moreover, were it suggested that all of the relations of which
a quality might be possessed are "merely internal", then the
identity of that quality could only be absolute, not relational.
Since it would contain within itself all of its relations (so-
called), that quality would be an entity quite in its own
right: its identity would be not relative to anything beyond
itself, but absolute.

What is absolutely self-identical is changeless and may not
become. Yet Appearance is not changeless: it is in process of
becoming. Now were we to attempt to view the differentia-
tions within a process of fission as being merely internal
to their terms, we could only succeed in ignoring altogether
the becoming there in process. For such an attempt would
end in seeing the terms as self-sufficient, each one within
itself, and so without the existence of any differentiation
or relation between them. Such terms would be self-
contained; their identity not relational, but absolute; and
their becoming in any sense would be impossible. And thus

40

The Dialectic of Relation and Quality

we see that "Every relation (unless our previous inquiries
have led to error) has a connection with its terms which,
not simply internal or external, must in principle be both
at once".* Relations must fall to some extent or to some
degree between the qualities of a process. Otherwise those
relations would fall wholly within their qualities, and so
they would fail to connect them. If such relations were
regarded not as differentiations at all, then they would be
"merely internal"; as is the philosophical relation of
resemblance for Hume; and so they would be in no wise
distinct from the respective characters of their self-identical
terms. If, on the other hand, the relations which a mistaken
opinion allowed to fall within their qualities, still were taken
to be differentiations, then fission would break out within
each quality, and the old issue would be raised all over
again. 24 - 26

A relation, taken as a differentiation, will differentiate
qualities, and so will fall to some degree between them.
A relation will be at once internal and external; the former
in so far as it contributes to constitute, or is confluent with
what it differentiates; the latter, in so far as it falls between
these qualities. Thus a relation will be at once involved in
and transcendent of its terms. Likewise, a quality, in being
differentiated, will be involved in its differentiation, while,
in being a quality differentiated, it will transcend its relation.

And so a relation will have a being proper to itself only
in so far as it transcends, and so is not, the very terms without
which it is not a relation at all. Hence the contrariety, or
"contradiction", which Bradley finds in the nature of
relation. Likewise, a quality is immediate or qualitative
only in so far as it transcends the very differentiations without
which it would not be a quality at all. Thus, alike in its
mediations or differentiations, and its immediacies or quali-
ties, experience is found self-discrepant, or incoherent.

* Ibid., p. 641.
41

Bradley's Dialectic

Yet this incoherence is in no case absolute; for nowhere is
an appearance self-identical. The identity of every quality,
we have seen, implies its difference from all else. Any experi-
ence is thus related with all different experiences. In virtue
of the unrestricted internality of relations and qualities,
experience is a systematic whole within which there may be,
and are, degrees of coherence in truth and reality.

II

The characteristic nature of Bradley's theory of relations
can be emphasized, perhaps, by comparing it with certain
other views about relations which have been, and are,
called "internal relations". The admirable elucidation of
ten senses in which the phrase "internal relations" either is or
may be used which Dr. Ewing gives us in his critical survey
of Idealism, affords a convenient and comprehensive text
for comparison. Moreover, sweeping and acute as is that
elucidation, it would seem unduly to neglect what is, for
British Idealism, the most germane sense of the phrase in
question. "Bradley's actual argument against relations,"
Dr. Ewing says, "I need not discuss at length because it has
already been answered by several different writers, and I
do not think I have anything really fresh to add."* Yet the
fact that Dr. Ewing considers this one of Bradley's arguments
to have been refuted, does not explain to us what Dr. Ewing
takes to be Bradley's own theory of internal relations. He
does say that "Bradley is commonly included among the
supporters of internal relations". But of which one, if any,
of Dr. Ewing's ten senses of "internal relations" Bradley is a
supporter, we are left to find out for ourselves.

Now, though it be found that Bradley's theory of relations
is radically different from any of the senses of "internal
relation" with which Dr. Ewing is concerned, that contrast

* Idealism, p. 147.

42

The Dialectic of Relation and Quality

in theory is not and is not to be taken as a criticism of Dr.
E wing's own account of the matter. Dr. Ewing gives the fol-
lowing statement of the first one of his ten senses of the
phrase "internal relation". "In the first place it is sometimes
said that all relations are internal in the sense that they all
Tall within the nature of the related terms'. This definition,
while etymologically more justifiable than most, involves
many ambiguities and confusions. If we mean by the nature
of a thing its full nature and do not include only its essential
characteristics, this may be interpreted as including every-
thing which is predicable of it and so all its relational charac-
teristics without exception, however unimportant they may
seem. We may then say that any relation falls within the
nature of both or all the terms that it relates, if by this is
meant simply that, whenever r relates A to B, A has the
characteristic of standing in the relation r to B and B has
the characteristic of standing in the converse relation to A;
but this will not carry us very far. It is an important fact
about relations that no instance of a relation can occur as a
self-subsistent entity, but only in conjunction with terms
which possess the characteristic of standing in that relation;
but if by 'nature' be meant essential nature it does not
follow, at least without further argument, that relations
fall within the nature of either or both related terms."*
Dr. Ewing goes on to point out still more ambiguities in this
sense of the phrase in question. So much as this, however,
may suffice for purposes of comparison. It is, presumably,
clear that relations which "fall within the nature of the
related terms" would be what Bradley calls "merely internal
relations". We have seen that Bradley rejects this view, on
which relations would be merely internal, as a relevant
interpretation of his own theory of relations. The distinction
between the essential and the accidental characteristics of a
thing's nature, which Dr. Ewing goes on to make, is excluded

* Ibid., p. 119.

43

Bradley's Dialectic

on Bradley's view of the matter. The identity of A implies its
difference from all else: the question of the importance, or
the unimportance of many, if not almost all, of these dif-
ferences for a given step in practical or theoretical life, is
irrelevant to the wholly relational identity of no matter
what. In this first sense of the phrase "internal relations",
such relations would "fall within the nature of the related
terms." Therefore they would be very like the merely internal
relations which Bradley rejects. And as emended by a dis-
tinction between essential and non-essential relational
characteristics, this definition is no less incompatible with a
doctrine on which no relations are non-essential to the
identity of any moment of appearance, or degree of Reality.

There is, moreover, a difference in principle between this
sense of the phrase "internal relations", and Bradley's view
of the matter. As an illustration of this, let us consider the
following texts: "The contradictions which he (Bradley)
alleges seem to arise through supposing that a relation must
be treated either as a quality or as a third term. For in the
former case it will qualify but not relate its terms, and in
the latter case it needs a fresh relation to link it to each term
and so ad infinitum." Thus, Dr. Ewing finds the ground of
Bradley's arguments to the conclusion that any relational
situation will be incoherent to some degree or other, in one
of two alternatives, both of which fail to treat relations as
relating anything. This would seem to say that Bradley
ignored the question at issue; for he is alleged to have been
concerned solely with entities (to call them that) which
were not, and could not be, proper relations at all.

Dr. Ewing may think that the notion of relations treated
as qualities is both self-evident in its bearing on Bradley,
and transparently false on any view. His procedure at this
point, however, is not that he goes on to justify his assertion
of that notion as an alternative ground for Bradley's theory
of relations, but rather that he proceeds at once to advance

44

The Dialectic of Relation and Quality

an alleged refutation of Bradley's rejection of independent
relations. Thus, with regard to the point that such a "rela-
tion" can only fail to relate its terms, Dr. Ewing writes:
"One might similarly argue that it was impossible to tie two
things together with string because you would need another
piece of string to tie the string to each object and so on for
ever. The argument would be valid if each piece of string
were so defective that it broke in the middle; similarly
Bradley's objection would be valid of relations if and only
if they did not fulfil their function of relating. Only then
would they need another relation to do the relating for
them. But in that case they would not be relations."* Dr.
Ewing's piece of string may well be a good illustration of a
relation regarded as being a third term. But Dr. Ewing's
opinion that Bradley's argument to the conclusion that any
notion of such independent entities leads to an infinite
process of relations that fail to relate, "would be valid if each
piece of string used were so defective that it broke in the
middle; similarly Bradley's objection would be valid if and
only if they did not fulfil their function of relating", is
difficult to accept. In his first sentence here, Dr. Ewing
appears to be saying that relations, as separate entities or
third terms, fail to be relations because they break in the
middle, like pieces of defective string. In his second sentence,
Dr. Ewing asserts that "similarly Bradley's objection would
be valid if and only if" relations failed to relate. The two
sentences together would appear to say that, as a relation
will fail to relate if it is like a piece of string that breaks in
the middle, so "similarly Bradley's objection" will be valid
only on the assumption that relations fail to fulfil their
function of relating. Thus, whereas in the first of these two
sentences, Dr. Ewing is concerned with relations as separate
entities, in the second of them he refers to relations without
qualification. Yet he says that, as he argues in his first
* Ibid., p. 147.
45

Bradley's Dialectic

sentence, so "similarly* 5 he may argue in his second sentence.
But that would mean that the relations referred to in the
first sentence are similar to those spoken of in the second.
And, indeed, the two sorts of relation are similar: for like
relations as third terms, the relations on which the validity
of "Bradley's objection" is alleged to depend are relations
which fail to relate.

Thus, immediately after the sentences in question, Dr.
Ewing writes: "only then would they need another relation
to do the relating for them. But in that case they would
not be relations."* It would seem that Dr. Ewing takes
"Bradley's actual argument against relations", and "the
contradictions which he alleges", to rest mainly, if not
altogether, on a version of Bradley's argument to the rejec-
tion of separate relations. These "relations", we have noticed,
may not relate anything, and so they fail to be relations at all.
Yet this negative argument surely is not the substance of
Bradley's "actual argument against relations" as real in
their own right. Were it so, Bradley would hardly have
found it necessary to follow up his argument showing the
impossibility of separate relations (which falls in chapter II
of Appearance and Reality), with the elucidation, in chapter III,
of the contrariety that is of the essence of the relational
situation. This positive dialectic depends on no assumptions
about relations which do not relate, but rather on the charac-
ter of relations which do relate the qualities they differentiate
and thus contribute to constitute.

Directly after his statements last quoted above, Dr. Ewing
goes on to offer a refutation, not of anything in chapter III,
but of the notion which he has asserted to be the basis of
Bradley's actual argument against "relations"; viz., that of
relations which do not relate anything. "If, in A and B, r
is really a relation, it relates A and B itself and does not
require new relations to connect it to either, for the relational

* Ibid.
46

The Dialectic of Relation and Quality

characteristic of standing in the relation is not itself a rela-
tion. To say that, because A is related to B by r, A must
stand in the relation r to B, and therefore must be charac-
terized by (i.e. stand in the relation of 'having as charac-
teristic' "to) the relational characteristic of standing in the
relation r to B, and be characterized further by having the
characteristic of being characterized by the relational charac-
teristic of standing in the relation r to B, is only to say the
same thing over again in different words, so that the so-called
different relations which are supposed to constitute the
infinite regress are really only more and more cumbersome
ways of expressing the same relation."* However sound this
may be as a view about relations and terms that are self-
identical, as a difficulty for Bradley's elucidation of the
relational situation it is inane. For, as stated by Dr. Ewing,
this view offers no explanation of how it is that r relates A and
B. To say that r relates A and B because r is a relation, and
that it is the nature of a relation to relate, is to be un-
reflectively assertoric. And that is not quite what Dr. Ewing
says. He says that r "relates A and B itself and does not
require new relations to connect it to either, for the relational
characteristic of standing in the relation is not a relation".*
Thus in A r B, A has the relational characteristic of standing
in r to B, and (that) B is characterized by the relational
characteristic of standing in the relation r to A. "But what
are we to understand here by c is' ?"

And that is Bradley's initial question at the outset of his
arguments to show that the meaning of the statement,
"one quality, A, is in relation with another quality, B", is
not altogether explicit. We do not mean that "being in
relation with B" is something different from B. Hence the
dilemma of predication; a dilemma which may not be
resolved by any distinction between the "is" of predication,
and the "is" of identity. Thus we are forced to abandon the

* Ibid.
47

Bradley's Dialectic

attempt to predicate relations of terms, or terms of relation.
"Let us abstain from making the relation an attribute of the
related," Bradley goes on to suggest, "and let us make it
more or less independent. 'There is a relation C in which
A and B stand; and it appears with both of them.' But here
again we have made no progress. The relation G has been
admitted different from A and B, and no longer is predicated
of them. Something, however, seems to be said of this
relation C, and said, again, of A and B. 5 ' 18 In Dr. Ewing's
view, what is thus said about A is that A is characterized by
the relational characteristic of standing in C to B; and B is
characterized by the relational characteristic of standing
in C to A. Yet on this view the dilemma of predication
remains. And the step which Bradley takes beyond his futile
attempt to regard relations as attributes is not, by Dr.
Ewing's view of the matter, even begun. Thus the explana-
tion of the connexity of relations, by the notion of relational
characteristics which characterize the terms which stand in
those relations, does not so much as tend to refute Bradley's
elucidation of the self-contradictory character of the rela-
tional situation, for it does not begin to follow through his
reasoning in that regard.

In Dr. Ewing's immediately subsequent paragraph in
this connection, he writes: "The same fallacy in a subtler form
appears in Bradley's argument that each term 'has a double
character, as both supporting and as being made by the
relation', 26 so that these two aspects will be again related
and so on ad infinitum. The distinction, so far as I can under-
stand it, is between A .as abstracted from the relation and
A as related. But what is the relation between these two
aspects of A? Simply that the second includes besides the
other characteristics of A the characteristic of standing in the
relation, while the first does not do so. But this will not
generate a second distinct relation between them and so on
ad infinitum, unless it is assumed as before that A must have

48

The Dialectic of Relation and Quality

a relation to the characteristic of standing in any particular
relation and that this relation must in turn be related to A
by another relation."* Here again, the theory of relational
characteristics is made to explain the connexity of relations.
Since, however, that theory leaves unanswered Bradley's
initial question about the relational situation, it can hardly
bear on Bradley's actual views in the matter. We have
noticed that a quality has a double character in that any
quality is itself and transcends itself. The quality A is itself
in so far as A is not the relations by which it is differentiated
and made itself. Yet without those relations A could not be
differentiated, and so A could be nothing at all. Hence,
that A may be differentiated, it must transcend itself in the
relations in virtue of which it is A. And there is no absolute
line between the moment of immediacy that is the quality A,
and the moments of mediation by which A is differentiated
by its relations. Wherever the emphasis in feeling may dis-
tinguish immediacy from mediation, quality from relation,
there a distinction is made. This distinction will be a fresh
differentiation in appearance; a differentiation which will
contribute to constitute the qualities thus differentiated.
Thus we may find a relational regress in any relational
situation.

Dr. Ewing has explained that, in his work on Idealism,
"I did not say of which of my senses of 'internal relations'
Bradley is a supporter because Bradley denied the reality of
relations and therefore can not have held that relations
really were internal in any of my senses or in any other
sense" t Thus Dr. Ewing reaffirms the statement, made in
his Idealism^ that "Bradley regards relations as unreal. ..."
One would have thought it common property that Bradley
denies reality to relations solely in the sense of the term

* Ibid., pp. 147, 148.

t Journal of Philosophy, Vol. XXXII, No. 10, p. 273.
.: P. 123.

49 D

Bradley's Dialectic

"reality" in which the Absolute alone may be real. To say?
flatly that Bradley "denied the reality of relations", is tojj
ignore Bradley's theory of degrees of reality. No relational
situation is ultimate reality: but any relational situation will
be real to the degree to which it is self-coherent and compre-
hensive.

Dr. Ewing has explained also that he did not consider
Bradley's theory of relational identity in connection with his
own account of internal relations, because that account
threatened to become unwieldy.* And therein lies the
difference in principle between the internal relations with
which Dr. Ewing is concerned, and Bradley's theory of
relations. In Dr. Swing's treatment of the matter, the
identity of a relation, and of a term, is absolute, not rela-
tional, as is the case in Bradley's view. This is to say simply
that such relations, and such terms, are taken by Dr. Ewing
to be what they are, not in virtue of their respective dif-
ferences from all else, but in virtue of what they severally
and respectively are in and of themselves alone. As thus self-
identical absolutely, and not relationally, such relations will
be internal in senses that can only be as different from
Bradley's view of the matter as are absolute and relational
identity. This difference, moreover, is that of a disjunction:
the identity of A may not be both absolute and relational.

This same difference in principle is again illustrated by
the second of Dr. Ewing' s senses of the phrase in question.
"The second meaning of 'internal relation' is 'a relation
essential to its terms.' This sense can easily be derived
from the first if we interpret 'the nature of a thing' more
strictly so as to exclude all characteristics which seem super-
ficial." f Here again, and throughout Dr. Ewing's discussion
of this sense, relations and terms are regarded as self-
identical each one in its own right. It is only to be expected,
therefore, that we find in this discussion no explanation of
* Jour, of Phil., XXXII, p. 273. f Ibid <> P- I22

50

The Dialectic of Relation and Quality

"being essential to" as meaning that relations contribute to
constitute their terms; and that, in so far as they do so,
relations transcend themselves as relations.

Dr. Ewing's third sense need not detain us. In this sense
of the phrase in question, internal relations are implicitly
reduced to qualities;* and that is a "definite error".f
It is repudiated as an error by Bradley. "Fourthly, the
internal view of relations may be taken as simply asserting
the fact that relations involve some kind of genuine unity
between their terms." As Professor Laird says, J "Ultimately,
. . . the question is whether a relation between things can
describe a genuine connection or unity between the things.
If it can, there is no mystery; for the fact, ex hypothesi, is
intelligible. If it cannot, such relations do not relate and
are unintelligible if they pretend to do what they cannot
do." It is plain that on Bradley' s theory of relations, this
view of the matter is not ultimate at all. For it offers no eluci-
dation of how and why it is that relations relate their terms.
A mere set of facts is not even its own description, much
less any analysis which might explain the "genuine unity"
of internal relations. For our main purpose in this connection,
however, it is enough to notice that, with regard to the rela-.
tions and the terms here in question, nothing is said which
so much as suggests that their identity is relational.

This is also pretty plain in the case of Dr. Ewing's fifth
sense of the phrase in question. "Professor Laird also criti-
cizes strongly a certain interpretation of the internal relations
view, which we may take as our fifth sense. 'It is possible
to assert with some confidence,' he says, 'that if A has some
relation r to B, it is not only logically conceivable that A
and B retain their character unmodified in the relation;
but it is logically inconceivable that they should not do so.

* A. G. Ewing, Idealism, p. 122.

| Ibid., p. 122.

J Knowledge, Belief and Opinion, p. 214.

51

Bradley's Dialectic

Relations hold between terms, and form or express a tie
between those very terms. Thus, in the propositions '3 is greater
than 2' and '3 is greater than i', one and the same 3 occurs
in both propositions, not a 3 modified by its relation to 2 in
the first instance and a different 3 modified by its relation
to i in the second instance. Either the whole relational way of
regarding things is mythopoeic, or this identity of terms must
be preserved."* <c Clearly he is right (Dr. Ewing continues)
in holding that no tolerable view of relations can be incom-
patible with the fact that the same term may stand in
different relations, and he is also, as I contend elsewhere,
right in holding that it cannot be true that all relations alter
or modify their terms, if by this is meant that they cause a
change in their terms. If A and B are to stand in a relation
at all they must first have a certain character of their own,
and this character is not made by the relation in question.
Relation!?, we may say then, are all external in the sense
that any relation presupposes a certain character in the
terms related which is itself not due to and not modified or
constituted by the relation in question. But for all that it
might still well be the case that the relation was internal
in the different senses, discussed later, that its character
followed from the character of the terms, and that the terms
could not be the same if the relation were different."!
On a logic of contradictories or of absolute identity, it would
seem to be plain that, as Professor Laird says, it is "logically
inconceivable" that A and B should be modified by r, or
by anything else. For, as intrinsically self-identical, A and B
are changeless, and therefore not to be modified. For the
same reason the number 3 of Professor Laird's first proposi-
tion would be qualitatively identical with the 3 of his second
proposition.

* Knowledge, Belief and Opinion, pp. 78, 79; cf. also Cook Wilson's
Statement and Inference, Vol. I, p. 71.
f Idealism, pp. 125, 126.

52

The Dialectic of Relation and Quality

In Bradley's view, however, finite identity is not absolute
but relational. Therefore, on this view, it is logically incon-
ceivable that A and B should not be modified by their rela-
tions. For it is in virtue of those differences that A and B
are differentiated from all else; and it is thus that their
respective identities imply their difference from all else.
This would have to be true likewise of the number 3 in
Professor Laird's first proposition and the number 3 in the
second proposition of his illustration. Since the two contexts
are different the relations are different: hence the meaning of
3 in "3 is greater than 2" will be different from the meaning
of 3 in "3 is greater than i".

Again, if the view of identity as relational be left out of
account (and we have noticed above Dr. Ewing's reason for
not considering that view in connection with his discussion
of internal relations), one may agree "that any relation
presupposes a certain character in the terms related which
is itself not due to and not modified or constituted by the
relation in question". It would be impossible that they should
be unmodified in any respect by the relations which differ-
entiate them. For those terms would be what they are in
every respect in virtue of their relations by which they were
differentiated.

Dr. Ewing derives his sixth sense from a phrase taken
from Bosanquet, who "defines internal relations as relations
grounded in the nature of the related terms". It might be
expected that any elucidation of this definition would be
carried out in the light of Bosanquet' s doctrine of the concrete
universal. Yet Dr. Ewing says of the phrase in question:
"By this must presumably be meant that their presence
depends on and is determined, either causally or logically,
by characteristics of their terms. This is undoubtedly true
of some relations, e.g. the mathematical relations, also
similarity and difference. That 7 is half 14 or that something
blue differs in colour from something red can undoubtedly

53

Bradley's Dialectic

be deduced from the intrinsic nature of the terms/' In view
of these illustrations, and of what Dr. Ewing says about them,
it would appear that he understands internality in this sense
to be the "mere" internality of intrinsic similarity and.
intrinsic difference. This view of internality is, presumably,
of fundamental importance on a logic of contradictories;
but it is, as we have noticed above, ruled out by Bradley's
conception of identity as relational.

Of the remaining four of Dr. Ewing's ten senses, the seventh
is that in which a relation is said to make a difference in its
terms. A relation is internal "in this sense if it is such that
both of the terms could not have been what they are without
the relation holding between them".* In the eighth sense
in question, a relation is internal if it is such that from "a
knowledge of one term and the relation in which it stands to
the other term", the second term may be inferred to possess
necessarily a certain characteristic "other than the charac-
teristic of standing in the relations in question".")" A relation
R is internal in the ninth sense "when A could not exist
unless B existed andwas related to it by R". J And a term that
is not only causally but logically dependent on its relation
to the other term and vice versa is related internally in Dr.
Ewing's tenth sense.

It may be said that since, for Bradley, relations differen-
tiate the qualitative character of their terms, no terms could
be what they are had they been otherwise differentiated.
Nor could A exist as the quality it is without its being
differentiated from B by R. It is also the case that, for Bradley,
any term is logically dependent for its qualitative character
on the relations by which it is differentiated from all other
terms. This can hardly mean, however, that Dr. Ewing's
seventh, ninth and tenth senses are equivalent to Bradley's
theory of relations. Though on that theory no quality could

* Ibid., p. 131. t N>id. 9 p. 135.

$ Ibid., pp. 135-6.

54

The Dialectic of Relation and Quality

be or exist as the quality it is were its relations different in
any respect, and though this dependence of quality on rela-
tion and vice versa is logical in being the nature of implication
(as opposed to "linear inference"), this is so because
relations contribute to constitute the identity of their quali-
ties. Thus the internality of relations in Bradley's sense
requires that relations be viewed as the moments of differen-
tiation in a process wherein the identity of each moment
implies its difference from every other. This would be to
assume a view of identity on which no single term or relation
could be repeated unaltered in diverse contexts; a view of
identity that nothing in Dr. Ewing's analysis of relations
would seem even to suggest.

55

CHAPTER II

Space and Time

BRADLEY assures the reader of Appearance and Reality that
once he has grasped the main argument of the chapter on
relation and quality, "he will have little need to spend his
time upon those which succeed it". For he will have seen that
our experience, everywhere and always, is self-discrepant to
some degree or other. Nevertheless, it may serve to illustrate
the main conclusions of chapter III if we consider the ways
in which Bradley finds that those conclusions about relation
and quality are exhibited by his conception of space and time.

At the outset of his very brief chapter on that subject,
Bradley explains: "The object of this chapter is far from
being an attempt to discuss fully the nature of space or of
time. It will content itself with stating our main justification
for regarding them as appearance. It will explain why we
deny that, in the character which they exhibit, they either
have or belong to reality. I will show this first of space." 30
To that end, Bradley proceeds to show that space must be
more than relational, and that space may be no more than
relational, in nature.

This "puzzle", as Bradley calls it, he sets forth anti-
thetically, (i) "Space is not a mere relation. For any space
must consist of extended parts, and these parts clearly are
spaces. So that, even if we could take our space as a collection
it woufd be a collection of solids. The relation would join
spaces which would not be mere relations. And hence the
collection, if taken as a mere inter-relation, would not be
space. We should be brought to the proposition that space

56

Space and Time

is nothing but a relation of spaces. And this proposition
contradicts itself." 31 Space must be more than relation, for
relations must have terms. And space must consist of extended
parts. These will be homogeneous solids, not relations at all.
Thus the relation in question, if it were the very nature of
space, would relate spaces which are not mere relations. To
take space as being no more than relational would be to
take it as consisting of relations and so of relations without
terms. Once the terms related are taken into account, it is
seen that space cannot consist of a relation of relations.
For that relation would not relate spaces.

Therefore, space must be more than a relation. (2) "But
space is nothing but a relation. For, in the first place, any
space must consist of parts; and, if the parts are not spaces,
the whole is not space. Take, then, in a space any parts.
These, it is assumed, must be solid, but they are obviously
extended. If extended, however, they will themselves consist
of parts, and these again of further parts, and so on without
end. A space, or a part of space, that really means to be
solid, is a self-contradiction. Anything extended is a collec-
tion, a relation of extendeds, which again are relations of
extendeds, and so on indefinitely. The terms are essential
to the relation, and the terms do not exist. Searching without
end, we never find anything more than relations, and we
see that we cannot. Space is essentially a relation of what
vanishes into relations, which seek in vain for their terms.
It is lengths of lengths of nothing that we can find."

"And, from the outside again, a like conclusion is forced
on us. We have seen that space vanishes internally into
relations between units which never can exist. But, on the
other side, when taken itself as a unit, it passes away into the
search for an illusory whole. It is essentially the reference
of itself to something else, a process of endless passing beyond
actuality. As a whole it w, briefly, the relation of itself to a
non-existent other. 5 ' 31 ' 32 Space is no more than relational

57

Bradley's Dialectic

because space consists of parts and these parts consist of
parts, and so on indefinitely. Thus space consists of the
differences by which it is internally divided without limit.
The notion that space consists of extended parts is an illusion
that springs from the assumption that the parts of space are
solid. Rather, those parts consist of divisions that are them-
selves divisible indefinitely. Those divisions or differentiations
are relations. That is why, "Searching without end, we never
find anything more than relations; and we see that we cannot.
Space is essentially a relation of what vanishes into relations,
which seek in vain for their terms". 32 This is so because space
is divided into parts that are divided into parts and so on
without end. Thus space consists of divisions or differentia-
tions, and they are relations. Hence we have it that space
is no more than the relations that are the parts of which
space consists.

And, when we attempt to understand space not as a
matter of parts, but as a unit, we are brought to much the
same conclusion. "We have seen that space vanishes intern-
ally into relations between units which never can exist." 32
Yet, as we attempt to consider space as a whole, we find that
it cannot be one. For consider, either space itself has bound-
aries, or else it has none. On the first alternative, space has
boundaries. They are its limits, and they differentiate it from
all that is not space. These boundaries, then, relate space
to whatever is different from space. Therefore, the non-
spatial terms of those relations are not spatial, and so we have
spatial relations that lack terms at one end. On the second
alternative, space has no boundaries: it is not differentiated
from anything else. On both alternatives, then, space is
"the relation of itself to a non-existent other". 32 The whole
of space with boundaries posited is differentiated by relations
which, at one end, have no terms. Thus the posited bound-
aries fail to bound; for they carry with them no termini for
the relations or differentiations which those boundaries

58

Space and Time

would be were they possessed of terms. And the whole of
space taken without boundaries obviously is not even a
specious whole.

Thus we may see that "Space is not a mere relation".
It is more than a matter of relations for the reason that space
is extended as well as divided and therefore must consist
of extended parts, not of mere divisions or relations. " But
space is nothing but a relation." For the parts in question
consist of parts, and so on indefinitely. As infinitely divided,
space is an infinity of divisions or relations. Therefore, space
is no more than relations. And these relations would be
hopelessly self-discrepant; for they would be relations
without terms.

With regard to time, Bradley thinks that "the reader who
has followed the dilemma which was fatal to space, will
not require much explanation. If you take time as a relation
between units without duration, then the whole time has
no duration, and is not time at all. But, if you give duration
to the whole time, then at once the units themselves are
found to possess it; and they thus cease to be units. Time in
fact is before and after in one; and without this diversity
it is not time. But these differences cannot be asserted of
the unity; and, on the other hand and failing that, time is
helplessly dissolved. Hence they are asserted under a rela-
tion. Before m relation to after is_JLhe character of time;
and here the old difficulties about relation and quality
recommence. The relation is not a unity, and yet the terms
are nonentities, if left apart. Again, to import an independent
character into the terms is to make each somehow in itself
both before and after. But this brings on a process which
dissipates the terms into relations, which, in the end, end
in nothing." 33 ' 34 Time, like space, is at once "a relation
and, on the other side, it is not a relation; and it is, again,
incapable of being anything beyond a relation". 33 Time as
a relation has to be a relation between terms of some sort.

59

Bradley's Dialectic

If these terms have no duration, then the whole- time has
no duration, and hence it is timeless. If, on the contrary, the
terms have duration, then they cease to be distinct moments,
for they are all contemporaneous with their duration. In
order that they may remain distinct, they must be before
and after each other. Without that inner asymmetrical
diversity a duration is not temporal. Hence time is a relation
the relation of before and after. Yet these differences in
before and after cannot be asserted of a single duration
without thereby denying its unity. Therefore, time cannot
be a relation. Nevertheless, if those differences are not
asserted of that unity, the duration in question remains
timeless. Hence time can only be the relation of before and
after by virtue of which a duration is temporal. Thus we see
that, despite the dilemma whose horns were touched upon
just above, time "is, again, incapable of being anything
beyond a relation.' 533

The relation of before and after is the nature of time;
"and here the old difficulties about relation and quality
recommence." 34 Without their diverse relations the diverse
moments would not be differentiated, and so would not be
before and after each other. Yet, with their diverse relations,
each moment is infected with self-discrepancy. For each
moment, as it stands before and after other moments, is a
"now". Within each now either there is no process, or there
is process.

On the first alternative, a now would not be a moment
of duration in process. On the second alternative, the pre-
sumed integrity of the now is destroyed from within. "Before
and after are diverse, and their incompatibility compels us
to use a relation between them. Then at once the old
wearisome game is played again. The aspects become parts,
the 'now' consists of 'nows', and in the end these c nows'
prove undiscoverable. For as a solid part of time, the 'now 5
does not exist." Midday is after morning and before after-
Go

Space and Time

noon. But in this process there is no self-sustaining point at
which a final line may be drawn between before and after.
No matter how broadly that process be measured, or how
minutely, the unit of process measured will contain within
itself processes, and these will be before and after one another.
"Pieces of duration may to us appear not to be composite;
but a very little reflection lays bare their inherent fraudu-
lence. If they are not duration, they do not contain an after
and before, and they have, by themselves, no beginning or
end, and are by themselves outside of time." 35 Moments of
duration are themselves in process of becoming; theyjxmtain
within themselves not barren^ simplicity butjprocesses^ which
are likewise durations ^^tjiT^^^r^and_a^r_ojie another..
To deny this is tantamount to affirming that the moments
of duration you started with have no beginning and no end,
no before and no after. But without beginning or end, these
moments would fall outside of time. And, "if so, time becomes
merely the relation between them; and duration is a number
of relations of the timeless, themselves also, I suppose, related
so as to make one duration. But how a relation is to be a
unity, of which these differences are predicable, we have
seen is incomprehensible. And, if it fails to be a unity, time
is forthwith dissolved. But why should I weary the reader
by developing in detail the impossible consequences of
either alternative ?" 35 A relation cannot be a unity of any
sort, for, as we have seen, a relation is any aspect of differen-
tiation in a process. This aspect is no static, vacuous being;
a differentiation is itself in process, and so contains within
itself differentiations and the moments of relative immediacy,
or quality, which they differentiate or relate. Yet if the
relation which allegedly relates the timeless units fails to
effect somehow a unity among them, we are left with the
notion of a time that would consist of timeless terms with-
out relations. Any such "time" would consist of "timeless
moments" that were not even differentiated, but simply one.

61

CHAPTER III

The General Nature of Reality

IN the course of the several remaining chapters of Book I,
Bradley illustrates his conclusion that "a relational way of
thought ... is a makeshift, a device, a mere practical
compromise, most necessary, but in the end most indefens-
ible". 28 The contradiction or contrariety essential to the
relational situation is likewise found in space and time,
motion and change, causation, activity, and the self.
Nowhere in appearance do we find a relational situation,
whatever its character, that is altogether self-coherent.
"The result of our First Book has been mainly negative. We
have taken up a number of ways of regarding reality, and
we have found that they all are vitiated by self-discrep-
ancy." 119 Yet this very negation requires a positive ground;
and it is to a consideration of this that Bradley turns in the
opening chapters of Book II.

When you condemn the self-discrepant as appearance,
and hold that appearance is not absolute reality, either you
have some notion or other, however dim and vague, of
what you mean by the reality which you say the self-dis-
crepant is not, or else your condemnation of appearance
as being not reality is groundless. For, on the second alterna-
tive, you have not the least notion of what it is that you say
appearance is not.

Even so, it may be rejoined, surely the ultimacy of the
relational and self-discrepant must be suspect, whether or
not a man be cognizant of the grounds of that suspicion.

62

The General Nature of Reality

And no less surely, about reality in itself, the sensible man
can be only agnostic. "To know the truth, we shall be told,
is impossible, or is, at all events, wholly impracticable.
We cannot have positive knowledge about first principles;
and, if we could possess it, we should not know when actually
we had got it. What is denied is, in short, the existence of a
criterion." 119 Yet, with no criterion of reality, how can you
deny that anything, however incoherent, is real ?

Bradley insists there can be but one answer to this question:
in condemning the inconsistent as appearance, we are
contrasting all this with what is not inconsistent; and this,
the self-consistent, is ultimate reality. "Our standard denies
inconsistency, and therefore asserts consistency. If we can
be sure that the inconsistent is unreal, we must, logically,
be just as sure that the reality is consistent. The question
is solely as to the meaning to be given to consistency." 119
If, by definition, the inconsistent is unreal, then by the
contradictory (or contrary) of the same definition the self-
consistent is real. A denial of the conclusion of this immediate
inference would deny the definition from which it follows.

Still, it may be objected, this is a barren conclusion.
To know, if we do, that the ultimate nature of things excludes
contradiction or contrariety is only to know that much; and
this is not to know anything positive about absolute reality.
"The denial of inconsistency, therefore, does not predicate
any positive quality. But such an objection is untenable." 122
In his Logic (Vol. I, chapter III), to which Bradley here
refers us, the reasons why such an objection as this one is
not tenable are given at length.

In the negative judgment "A is not B" the negation may
not be a merely external relation. Neither may the negation
be said to affect only the copula: nor may it be said to
belong only to the predicate. For then the judgment "A is
not B" would read "A is not-B"; and we should be judging
A to be qualified by not-Bness. In short, * 'negation presup-

63

Bradley's Dialectic

poses a positive ground".* This ground will be that in virtue
of which A is not, or excludes B. "Every negation must have
a ground, and this ground is positive. It is that quality x
in the subject which is incompatible with the suggested
idea. A is not B because A is such that, if it were B, it
would cease to be itself. ... In other words, its quality x
and B are discrepant. And we can not deny B without
affirming in A the pre-existence of this disparate quality."
Thus the redness of a rose here and now is incompatible
with its being yellow. And that very determinate red
would be the positive ground of the judgement "this rose
is not yellow".

But more often than not, perhaps, this positive ground
is not made explicit in the mind of the subject who makes
the judgement. He might assume, and without thinking
about it, that as an extended thing, the rose recognized at
a distance has some colour or other, and so judge signifi-
cantly, "that rose is not yellow" whatever the perceived
colour of it might turn out to be. Yet were a negative judge-
ment without any such ground, however far to seek, or
difficult to discriminate, it would be groundless altogether,
and so not a judgement at all.

Thus, even though our objector should take it that
Bradley's criterion here is fairly stated, "reality is not
self-discrepant", still he must acknowledge that (on Bradley's
theory of negation) this negative judgement requires a
positive ground. And, at the very least, this ground can
only be the self-consistency of the real. Even so, it may still
be urged, "the criterion is a basis, which serves as the founda-
tion of denial; but, since this basis cannot be exposed, we are
but able to stand on it and unable to see it". 122 If, at this
stage, more than a criterion of the nature of reality were
in question, such an objection as this one might seem
plausible: "and there is a sense in which I am prepared to
* Lovic> Vol. I, p. 114. f Ibid., p. 117.

The General Nature of Reality

admit that it is valid". 122 The plausibility here would derive
from the assumption that the criterion of reality, as opposed
to appearance, is taken to be more than a criterion. As such
it does indeed afford no detailed knowledge of reality.
"But that is very far from being the point here at issue.
For the objection denies that we have a standard which
gives any positive knowledge, any information, complete
or incomplete, about the genuine reality. And this denial
assuredly is mistaken." 122 It is assuredly mistaken, because
appearance is not blank nothing and is, therefore, in some
sense real. The real, we have found, is the self-consistent.
And so we know "that everything, which appears, is some-
how real in such a way as to be self-consistent. The character
of the real is to possess everything phenomenal in a har-
monious form". 123 How this may be so, it is the burden of
Bradley's theory of degrees of truth and reality to explain
in some detail.

To find that no relational situation is wholly self-consistent
is not to find that any situation is entirely or absolutely
incoherent. For the absolutely incoherent would be not a
unitary situation at all, but rather a mere aggregation of
items. These items would be other than each other; and, by
hypothesis, each one would be incompatible with, and so
would exclude, every other one. Such a process would
be not a process at all, but a fixed set of incompatibles.

The internality of relations .rules out any plurality of
reals as merely other than each other. "Otherness" would
be the name of what (in his essay on Relations) Bradley calls
a "merely external relation". For it would in no wise
differentiate its intrinsically diverse terms. Such terms,
intrinsically other than each other, would be what they were
respectively in virtue of themselves alone. They would be
self-identical, and therefore changeless. For that reason alone,
such terms would be, at best, the merest abstractions from
process.

65 E

Bradley's Dialectic

That there may be different qualities in appearance,
there must be distinct differentiations by which those distinct
moments of immediacy in process are differentiated. The
character of any and every quality requires the differentia-
tions in virtue of which the quality is what it is. This is
to say that "identity implies difference". Since identity
everywhere and always implies difference, this oft-quoted
dictum may not be taken in a restricted sense. Were there
some qualities, X, Y, Z, whose identity were not relational
but intrinsic, their being different likewise would be intrinsic
to those qualities themselves. This difference would be
what Bradley calls a ' 'merely internal" relation. Such a
relation would be not a connection or differentiation at all;
for an intrinsic difference could be nothing distinct from the
intrinsically different terms themselves. And terms that were
intrinsically individuated, or differentiated, would be self-
identical and so changeless. A relation, to be a differentia-
tion at all, and not a merely internal relation, must be dis-
tinguishable from its terms.*

But more than this: were the identity of X, Y, and Z not
relational but intrinsic, those terms would be absolute,
each one in its own right. No one of them would be related
with those terms whose identity is not absolute, but rela-
tional. And so X, Y, and Z would stand in a merely external
relation to those terms whose identity implies their respective
difference from all else. But any such externality is ruled
out by Bradley's elucidation in the Chapter on Relation and
Quality of how it is that differentiations in process are

* What is the difference between a merely internal and merely external
relation ? Neither the one nor the other is a differentiation. Intrinsic
difference is internal because any alteration in the terms ipso facto
is an alteration in their intrinsic difference : and that difference is merely
internal because it is intrinsic. Such is not the case with a merely external
relation. If the otherness of a hue and a figure be taken to be relation,
then as a merely external relation it does not relate them. And so there
could be nothing in that merely external relation to be altered by an
alteration in the hue or the figure.

66

The General Nature of Reality

internal to the qualities thus differentiated. The internality
of relations holds not at all, or it holds everywhere and always.
To object that it holds not at all, is to invite oneself to expose
as fallacious the dialectic by which Bradley finds the intern-
ality of relations to be the outcome of his initial elucidation
of appearance in process.

The quality A is A, not in virtue of its difference from
some other qualities and relation, but in virtue of its difference
from all else. And A is any quality: where and whenever a
moment of immediacy is differentiated there will be a
quality whose character is what it is in virtue of its difference
from all else. Thus, by its differentiations from all that is
not itself, any quality is related with the indefinite variety
of appearances from which it differs. And any attempt to
maintain the reality of relations as merely external must
fail. "For if, wrongly and for argument's sake, we admit
processes and arrangements which do not qualify their terms,
yet such arrangements, if admitted, are at any rate not
ultimate." 125 For nothing short of the whole system of
internally related qualities, and internally qualified rela-
tions, can be ultimate. Anything less could be final in its
own right only if it were absolute, and so external to the
system of internal relations. But that this is not a real alter-
native, the dialectic of the internality of relations makes
plain.

Whatever is distinct is different; and to be different is
to be differentiated, and so we have relation at once. ' 'Rela-
tions are unmeaning except within and on the basis of a
substantial whole, and related terms, if made absolute,
are forthwith destroyed." 125 For absolute terms would be
non-relational, and therefore not differentiated, and so
nothing at all. Thus identity implies difference, everywhere
and always. Any case of appearance will be what it is in
virtue of its differences from all else. Thus, by the considera-
tion of any form of process, we are led at once to the conclu-

Bradley's Dialectic

sion that reality cannot be less than a systematic whole of
qualities and relations. That absolute Reality is more than
this, the essential (although relative) incoherence of the
relational situation indicates: Reality is the absolutely self-
within which the various ^degrees of in-

coherence in Appearance are resolved.

How, even in principle, this resolution of degrees of
coherence into an absolute reality is made out by Bradley,
is a topic to be dealt with in its proper place. "Our result
so far is this. Everything phenomenal is somehow real;
and the absolute must at least be as rich as the relative.
And, further, the absolute is not many; there are no indepen-
dent reals. The universe is one in this sense that its differences
exist harmoniously within one whole, beyond which there
is nothing. Hence the absolute is, so far, an individual and a
system, but, if we stop here, it remains but formal and
abstract." Can we then, the question is, say anything about
the concrete nature of the system? Bradley answers that
reality Jis jsentient experience^ For everywhere and always
what we have is the felt content of consciousness. Any attempt
to deny this, say by way of a distinction between the experi-
encing and the experience, which held the "ing" and the
"ed" to be not distinct merely, but independent, could only
ignore the internality of relations.

The Cogito surely illustrates a truth; viz, the truth that no
content of consciousness may deny its own reality. And
the identity, or character, of any content implies its difference
from all else. In virtue of the relations by which any content
is differentiated continuously within the whole, any content
will be continuous with the whole; and the whole thus will
be continuous with any content. Anything other than tht
systematic whole of internal relations would stand in an
external relation to that whole. Yet this something or other
is but a verbal fraud. For, as standing out of all relation
to the system of internal relations, there would be nothing

68

The General Nature of Reality

by which it could be differentiated and made different and
itself: such an "it" could be nothing at all.

The conclusion that the real is sentient experience follows
from an acknowledgement of the fact of sentient experience
in any form, and an understanding of Bradley's theory of
relations. In his conclusion here, Bradley fears he "may be
understood to endorse a common error. I may be taken first
to divide the percipient subject from the universe; and then,
resting on that subject, as on a thing actual by itself, I may
be supposed to urge that it cannot transcend its own states.
Such an argument would lead to impossible results, and
would stand on a foundation of faulty abstraction. To set
up the subject as real independently of the whole, and to
make the whole into experience in the sense of an adjective
of that subject, seems to me indefensible. And when I contend
that reality must be sentient, my conclusion almost consists
in the denial of this fundamental error. For if^/seeking for
reality, we go to experience, what we certainly do not find
is a subject or an object, or indeed any other thing whatever,
standing separate and on its own bottom. What we discover
rather is a whole in which distinctions can be made, but in
which divisions do not exist. And this is the point on which
I insist, and it is the very ground on which I stand, when I
urge that reality is sentient experience". 128 That there must
be distinctions, and that there may be no separations, follows
from the doctrine of relations that is elucidated in the course
of chapters II and III of Appearance and Reality. Appearance,
whatever its form, is systematic: there are to be found, and
may be, no isolated and self-contained appearances existing
out of relation with the systematic whole of internal relations.
The real is sentient experience, not because reality is the
adjective oi a sell-Identical finite subject Jbut ratherjjecause
tKereTT^entient experience, and because ^nything^jother
^Kan~6r~ merely external to it, the _injternality_^.jel.ations
rules out. It may^be that solipsism is implied by any subjective

Bradley's Dialectic

Idealism: certainly it is impossible within Bradley's meta-
physics. For no finite centre of experience may be absolute:
it is what it is in virtue of its relations with all else. Any
attempt to posit a self that is a finite individual alone and
in its own right can only ignore the theory of relations in
virtue of which there is but one individual that is Absolute.

The dialectic or elucidation of relation and quality leads
to a monism at once. A "some- what" that were not dif-
ferentiated from all else could be nothing at all. And once
differentiated from all else, any content whatever will be
related with all that is not that relational content. The
character of the content of reality may not be other than
sentient experience. For the reality of sentient experience
may not be denied. And anything other than sentient
experience would be the verbiage of a term standing in a
merely external relation.

Bradley is well aware that the concrete nature of reality
is thus hardly more than indicated, and not made out. 146
Yet a conclusion about the general nature of reality has
been reached. "Our conclusion, so far, will be this, that the
absolute is one system, and that its contents are nothing
but sentient experience. It will hence be a single and all-
inclusive experience, which embraces every partial diversity
in concord. For it cannot be less than appearance, and hence
no feeling or thought, of any kind, can fall outside its limits.
And if it is more than any feeling or thought which we know,
it must still remain more of the same nature. It cannot pass
into another region beyond what falls under the general
head of sentience. For to assert that possibility would be in
the end to use words without a meaning. We can entertain
no such suggestion except as self-contradictory and as
therefore impossible." 129 We may reject this conclusion,
I submit, only if we reject Bradley's dialectic of relation
and quality. And to that elucidation of the contraries which
constitute process, any alternative on which relations and

70

The General Nature of Reality

their terms respectively are intrinsically self-identical, and
so absolute, each one in its own right, is of no avail. For
such self-identical terms and relations may not become.
Hence, to adopt any such alternative would be to deny
not only Bradley's dialectic of relation and quality, but
also, and altogether, the very possibility of becoming
itself.

That, it may be urged, is what ought to have been done
in the first place. It is an old story that what is called
becoming is unintelligible, and an error to suppose that
any case of development exists. Hume, for example, was a
sensible man: he considered what we actually have in
appearance; viz. succession: not development, or becoming.
And if he did little or nothing toward elucidating the nature
of succession, still that is not difficult to do in terms of the
substitution of items in a compact series.

This hearty optimism would not be easy to uphold under
criticism. For in a case of succession such as A, and then b,
and then C, where is the successiveness to be found? By hypo-
thesis, A and B and C are respectively self-identical and
therefore changeless. Evidently, then, the succession of

A, B, G is not to be found either in A or in B or in C. The
change in virtue of which B is substituted for A, and C for

B, may not exist within those items themselves. Where then
does it exist ? And what is its nature ? To these questions
our optimism affords no answer. And if, consistently with the
self-identity of each one of the successive items, we try to
think of the change in virtue of which they are successive
as self-identical, we are faced with a flat impossibility.
A self-identical change could only be changeless; and a
changeless change is a contradiction in terms.

The change, in virtue of which successive items succeed
one another, can hardly be a self-identical nature or form
of any sort. And since no change may be self-identical, no
change may be the contradictory of, or merely other than,

Bradley's Dialectic

any different change. Thus but one logical alternative
remains: we must seek to elucidate the nature of process, not
by any method for which identity is absolute, but rather by
a dialectic of contraries. This allows us to notice that the
identity of any moment of process is not absolute but
relational. The quality A is not moment B in any sense in
which the "not" here would be the sign of pure negation.
Rather, the quality A is a contrary of moment B; and one
middle term which falls between A and B is C. By this middle
term they are differentiated; but not by it alone. And, since
there is nowhere pure negation, this differentiation will be
not external to, or other than its qualities: rather it will
contribute to constitute their character, and so be internal
to them. The respective identities of A and B imply that very
differentiation, as the identity of that differentiation implies
those qualities. And the conclusion that identity is relational
that identity implies difference holds without exception.
For the identity of any exception would not be relational
but absolute. And for that reason, any alleged exception
to the conclusion that identity implies difference could have
no existence in process. The identity of A implies its differ-
ence from all else; and so at once we have a systematic whole
of internal relations.

The use of the term implication in this context does not
introduce a new factor into the dialectic of relations. No case
of systematic implication (as distinguished from "linear
inference", and from the implication which "linear infer-
ence" requires that it may be a priori) is anything distinct
from the relational situation which is that implication. To
say that qualities imply relations, and that relations imply
qualities, means that qualities are internal to their dif-
ferentiations, which are internal to those qualities. And this
means that the quality contributes to constitute the differen-
tiation, and the differentiation the quality: where the
phrase "contributes to constitute" designates nothing in the

72

The General Nature of Reality

relational situation distinct from the moments of immediacy
and mediation which are the process itself.

The absolute to which this relational process leads we
may know not concretely and in detail, but rather in abstract
terms; and in a certain intimation of its character which our
experience affords us. And our position here is ineluctable.
So long as the relational way of thought is ours, then no
matter to what level of concreteness we may attain, still we
fall short of the Absolute. In order to know the Absolute,
we should have to be absolute, and so cease to be relational
and finite and ourselves. "What is impossible is to construct
absolute life in its detail, to have the specific experience in
which it consists. But to gain an idea of its main features
an idea true so far as it goes, though abstract and incomplete
is a different endeavour. And it is a task, so far as I
see, in which we may succeed. For these main features, to
some extent, are within our own experience; and again the
idea of their combination is, in the abstract, /quite intelligible.
And surely no more than this is wanted for a knowledge of
the Absolute. It is knowledge which of course differs enor-
mously from the fact. But it is true, for all that, while it
respects its own limits; and it seems fully attainable by the
finite intellect." 140 The source of this imperfect knowledge
is two-fold: on the one hand, there is "mere feeling or
immediate presentation", 140 > 141 on the other, there is the
relational character of process, which "implies a substantial
totality beyond relations and above them, a whole endeavour-
ing without success to realize itself in their detail". 140 How
it is that these two aspects of the matter may be taken
together as affording an intimation of the character of the
Absolute is a question to which we may now turn.

In immediate presentation we have what Bradley calls
"the This and the Mine". For in the moment of immediacy
that is quality, there is an immediacy of feeling in virtue of
which the quality is this moment, and mine. "The 'this 5 and

73

Bradley's Dialectic

the c mine' are names which stand for the immediacy of
feeling, andr each serves to call attention to one side of that
fact. There is no 'mine' which is not 'this', nor any 'this'
which fails, in a sense, to be 'mine 5 . " 197

Any experience, in being mine, will be also this experience.
And the "this" has a quite general referent: for, as so used,
the term refers to the immediacy of feeling with which any
moment of experience will 'be suffused. This immediacy
of feeling "brings a sense of superior reality, a sense which is
far from being wholly deceptive and untrue. For all our
knowledge, in the first place, arises from the 'this'. It is the
one source of our experience, and every element of the world
must submit to pass through it. And the 'this', secondly,
has a genuine feature of ultimate reality. With however great
imperfection and inconsistency it owns an individual charac-
ter. The 'this' is real for us in a sense in which nothing else
is real." 198 Any moment of process will be this moment, so
that any knowledge we may have will "arise from the this".
Again, any moment of process will be internally related with
my finite centre and thus, in that sense, any "this will also
be mine".

It is in Bradley's explanation of his second point here that
we have an account of the sense in which the "this" affords
us some intimation of the character of the Absolute. Ultimate
reality is not self- transcendent; for it is not relational. The
Absolute, that is to say, is self-identical: it is not mediated,
or in process in any sense of the term. Now "the 'this'
possesses to some extent the same wholeness of character.
Both the 'this' and reality, we may say, are immediate".
Yet whereas the Absolute is above relations and wholly
self-identical, the this is immediate "because it is at a level
below distinctions", 198 ' 199 and not absolutely immediate.
For the "this" is the aspect of undiscriminating feeling(which
is the felt immediacy of any moment of sentient process.
As undiscriminating, such feeling exhibits no differentiations

74

The General Nature of Reality

within itself; it is almost mere sentience; it is the freshness
or the apathy or the dullness or the vivacity, and so on, of
an experience. Yet the "this", though it be the undiscrim-
inating feeling tone of any experience, is not itself undis-
criminated. And in being discriminated, however slightly,
the this is differentiated, and so it is relational.

Again, if within the immediate feeling that is a this there
be a tendency toward discrimination, and so toward internal
differentiation, still such a "this" remains below the level
of fully explicit differentiation. As thus at a level in process
below that of explicit distinctions, the "this" is such that
"Its elements are but conjoined, and are not connected.
And its content, hence, is unstable, and essentially tends to
disruption, and by its own nature must pass beyond the
being of the e this s . But every 'this' still shows a passing aspect
of undivided singleness. In the mental background specially
such a fused unity remains a constant factor, and can never
be dissipated." 199 Although the "this" is never absolutely
stable, and tends to pass over into differentiation, still at the
same time no "this" fails to present a momentary aspect of
undivided unity. And it is in this momentary phase of
undifferentiated unity which any momentjbf experience,
no matter how elaborate, will bear with it, that we have our
intimation of the character of the absolute whole. That is
why "The 'this 5 is real for us in a sense in which nothing
else is real". 198 For the "this", though never absolutely
stable, and still relational in virtue of its differentiations
from all else, is more nearly absolute in its identity than is
anything else to be found in process.

Yet if we take the "this" less narrowly, and in something
of its context, we find that the phase of passing unity, which
is its content, may be but fugitive. For "this" excludes
"that"; and this exclusion may not be an external relation.
Hence we see that the "this", taken negatively or as exclud-
ing a "that", is relational and therefore not ultimate. But if

75

Bradley's Dialectic

we take the "this" quite narrowly and as no more than a bare
moment of immediacy, then the "this" excludes nothing and
is merely positive. "We have found that the 'this', taken as
exclusive, proclaims itself relative, and in that relation for-
feits independence. And we have seen that, as positive, the
'this' is not exclusive at all. The 'this 5 is inconsistent always,
but so far as it excludes, so far already has it begun internally
to suffer dissipation." 203 The "this" is always inconsistent
because, like any other quality in process, it is at once imme-
diate and self- transcendent: yet taken as it occurs in process,
and so with its differentiations, the "this" is relational.
Again, as sheer immediacy, the "this", Bradley holds, is
without content. For content implies distinction; 203 and,
within utter immediacy, there are no distinctions at all.

Even so, it may be urged, the "this" is surely something
more than the content of a moment of process. For any such
content may be this content and mine; so that there is no
specific content that is proper and peculiar to the "this"
and the "mine". "In the 'this 5 . . . there is something
more than content. For by combining qualities indefinitely
we seem unable to arrive at the 'this 5 . The same difficulty
may be stated perhaps in a way which points to its solution.
The 'this 5 on one hand, we may say, is nothing at all beside
content, and on the other hand the 'this 5 is not content
at all. For in the term 'content 5 there lies an ambiguity. It
may mean a what that is, or again is not, distinct from
its 'that'. And the 'this 5 , we have already seen, has incon-
sistent aspects. It offers, from one aspect, an immediate
undivided experience, a whole in which 'that 5 and 'what'
are felt as one. And here content, as implying distinction,
will be absent from the 'this 5 . " 203 Within a whole devoid
of distinctions, plainly there may be no distinct content that
might be proper and peculiar to the "this 55 . Thus, in an
immediate undivided experience there will be a passing
aspect of singleness, in virtue of the undifferentiated charac-

The General Nature of Reality

ter of the immediacy thus presented. Within this undivided
aspect there are no distinctions; for it is internally undifferen-
tiated,ythe bare immediacy in question exhibits no differen-
tiations within itself; and so it offers no distinct content which
might belong to the "this". Yet any such utter immediacy
may be this immediacy and mine; and so we notice that the
"this" designates no content proper and peculiar to itself.
"If, on the other hand, we use content generally, and if
we employ it in the sense of 'what' without distinction from
'that' if we take it to mean something which is experi-
enced then, most emphatically the 'this' is not anything
but content." 203 In this content, taken as in no wise trans-
cending itself, there will be no distinction between what
the content is and the fact that the content is this somewhat.
So taken, this content will be immediate. In the sense that
this content, so taken, is in no wise distinct from this same
content, the "this" is nothing other than content.

In short, there is no sense of the term "this" in which it
designates a content proper and peculiar to itself. If this
content be taken in point of its immediacy, then there are
to be found within that immediacy no distinctions whatever.
And this is to say that the content, so taken, exhibits nothing
that could be proper and peculiar to the "this", other than
the single content itself. Yet we have noticed that any content
may be this content, and that this content may be any con-
tent. There is, and could be, no content which would be the
proper and peculiar referent of the term "this". "If we are
asked what content is appropriated by the 'this', we may
reply that there is none. There is no inalienable/content
which belongs to the 'this' or the 'mine'." 206 The term
"this" is not the proper name of any quality or relation. It has
no proper and peculiar connotation of its own. And in every
case of its use, the significance of the term "this" derives
from the connotation of its context. The context will describe
a content, and will describe it as being partially, at least

77

Bradley's Dialectic

immediate. In so far, then, as the context conveys the felt
immediacy of the content, the "this" is significant of that
immediacy. Yet, since such a content in its felt immediacy
may be any content, there is no inalienable content which
belongs to the "this" or the "mine". No content is proper
and peculiar to the "this", simply because any content may
be this content. "We have found that, in a sense, the 'this'
is not, and does not own, content. But, in another sense,
we have seen that it contains, and is, nothing else." 206 If
/it be undeniable that the referent of the term "this" may be
no matter what, then the referent of the term^rhis" may not
be something distinct from any content; something which
as distinct, would belong only to the "this"; and so the
referent of the term "this" is nothing else than content, or
just any content in its felt immediacy.

Were the "this" and the "mine" not exclusive of the
"that" and the "yours", it might be urged that they are
real in isolation from all else. To be sure Bradley writes of the
"this" and the "mine" as being below the level of relations.
But to be below the level of distinctions is not to be by that
very fact unrelated altogether. Within the aspect of the
"this" that is its bare immediacy, there are no distinctions,
and, therefore, no relations. Yet the question, it may be
urged, is how this aspect which, within itself, is below the
level of relations, is related to the content of which it is
a passing aspect ? And the answer here could only be that it
is by virtue of the difference of this aspect from the rest
of the content that the aspect is related to or continuous
with the content. For within itself the aspect is devoid of
relations: so taken it is immediate, and thus different from
what of the content is relational. And by virtue of that
difference, the passing aspect of immediacy is continuous
with its content.

Moreover, to point out that the "this" is below the level
of relations is not to say that the "this" is relationless, "An

78

The General Nature of Reality

immediate experience, viewed as positive, is so far not
exclusive. It is, so far, what it is, and it does not repel any-
thing. But the 'this' certainly is used also with a negative
bearing. It may mean 'this one', in distinction from that
one and the other one. And here it shows obviously an ex-
clusive/aspect, and it implies an external and negative
relation." But every such relation, we have found, is incon-
sistent with itself (chapter III). For it exists within, and by
virtue of an embracing unity, and apart from that totality
both itself and its terms would be nothing. And the relation
also must penetrate the inner being of its terms. " 'This', in
other words, would not exclude 'that', unless in the exclusion
'this', so far, passed out of itself. Its repulsion of others is thus
incompatible with self-contained singleness, and involves
subordination to an including whole. But to the ultimate
whole nothing can be opposed, or even related." 201 The
contrariety in all process is presented in its most immediate
and crucial form in the "this" and the "mine". As positive,
the "this" is unmediated within itself. Yet, that it may be
"this" rather than "that", a "this" must be different from
a "that". By virtue of that difference a "this" will transcend
itself. And that self-transcendence, which is the very process
of becoming, is not consistent with the positive immediacy
of the "this". For as positive, "It is, so far, what it is, and it
does not repel anything." 201 Yet any "this", while yc is
immediate, is also relational; it is at once both itself[and
self-transcendent in virtue of the very differentiation by
which its immediacy is this immediacy and not that one.
Thus we may come to see that even immediate experience
is related within the systematic whole of sentience that is the
content of Reality, and of whose Absolute immediacy the
"this" affords us an intimation.

The other source of our knowledge of the absolute, which
Bradley mentions, is that of the relational character of
experience. "And, again, the relational form, as we saw,

79

Bradley's Dialectic

pointed everywhere to a unity. It implies a substantial
totality beyond relations and above them, a whole endeavour-
ing without success to realize itself in their detail." 141 The
sense, if any, in which the internality of relations may imply
a whole that would be different from, and over and above,
the system of internally related qualities itself, remains to be
seen. Yet, whether or not the internality of qualities and
relations leads to an Absolute whose identity is not relational,
it does supply the nexus in virtue of which we may know that
the relational way of thought, though it be never more than
an ineluctable compromise, is always thinking what is some
degree or other of reality.

80

CHAPTER IV

The Internality of Thought
and Reality

AT the outset of his elucidation of how thought and reality
are related, Bradley distinguished between the "what"
and the "that"; and this distinction is fundamental for his
theory of how it is that we know the real. "If we take up
anything considered real, no matter what it is, we find in
it two aspects. . . . There is a 'what' and a 'that', an
existence and a content, and the two are inseparable." 143 1
We can say of anything we choose to consider both that it
is and what it is; where the "that" marks the existence of the
what, and the "what" refers to the quality or character of
that existent. These two aspects of anything are distinct and
distinguishable, but they are not separate or separable.
A bare existent: one that were without a character or content
of any sort: would be nothing at all. Again, a content that
were without character of any sort would be nothing at all.
Again, a content that were not that content, or this content,
would be nothing distinct from anything else, and so it
would be nothing at all.

Yet, in the process of thinking, a "what" will be differen-
tiated from its "that". "For thought is clearly, to some extent
at least, ideal. Without an idea there is no thinking, and an
idea implies the separation of content from existence. It is a
what which, so far as it is a mere idea, clearly is not, and if
it also were, could, so far, not be called ideal. For ideality
lies in the disjoining of quality from being." 142 ' 143 This tells
us what Bradley means by the "ideality of the finite".

81 F

Bradley's Dialectic

Thinking is ideal in that the content of thought will be a
"what" partially prescinded from its "that". This abstrac-
tion may in no case be more than partial. For if a what were
wholly abstract and quite independent of any context of
relations whatever, it would be differentiated by nothing,
and so "it" would be nothing at all. "If we try to get the
'what' by itself, we find at once that it is not alJL/It points
to something beyond and cannot exist by itself and as a bare
adjective. Neither of these aspects, if you isolate it, can be
taken as real, or indeed in that case is itself any longer.
They are distinguishable only and are not divisible." 142
As we abstract, we progressively alter the relations in which
the content initially stood. A "what" which were quite
prescinded from its "that" would be a content cut off from
all relations. Any such what would be a "floating idea"; a
something that were external to the internality of relations.
We have seen above the reasons why nothing real may be
external to all relations. Without relations, "it" would not
be differentiated from anything and so "it" would be
nothing at all. The "what" and the "that" are distinguish-
able, not separable.

Bradley warns us against the error of taking it that the
ideal may be mere psychical fact, such as an image or a
sense-perception. Since the ideal consists of a "what" which
is partially transcending its "that," "the common view which
identifies image and idea is fundamentally in error. For an
image is a fact, just as real as any sensation; it is merely a
fact of another kind and it is not one whit more ideal. But
an idea is any part of the content of a fact, so far as that works
out of immediate unity with its existence. And an idea's
factual existence may consist in a sensation or perception,
just as in an image. The main point and the essence is that
some features in the 'what' of a given fact should be alienated
from its/ c that' so far as to work beyond it, or at all events
loose from it. Such a movement is ideality, and, where it is

82

The Internality of Thought and Reality

absent, there is nothing ideal. " 14 ^/The psychical fact that is
an image of imagination, is a fact which is no more ideal
than is the psychical fact that is a sense perception. It is in
the movement by which moments of sentience are in processes
of differentiation, or self-transcendence, that ideality con-
sists. The ideality of the finite, or the partial transcendence
of a "that" by its "what", is in no respect different from the
process of the internality of relations.

We have noticed that the "what", or the character of a
quality, inevitably transcends itself in and through the very
differentiation, or relation, by virtue of which it is that
quality, and not a different one. Likewise, any differentia-
tion, or relation, will transcend itself in so far as what falls
between the qualities thus differentiated is internal to, or
in and of the terms thus related by that "what". Thfs move-
ment of self-transcendence is everywhere the nature of
process. In the "this" and the "mine" self-transcendence is
but incipient, to be sure; yet, as we noticed above, it is there
present. And in explicit thinking the ideality of the content
is realized to a discriminable degree. This ideality of the
finite is the very same as the self-transcendence of moments
of process or Appearance. This self-transcendence of content
arises by virtue of the incessant internal differentiation, or
relation of moments of Appearance.

In the moment of differentiation that is thought, as dis-
tinguished from judgement, we have no more than the
partial transcendence of a "that" by its "what", without
the consummation of the reference of this "what" to a
"that" which lies beyond the context from which that
"what" is passing. With the consummation of that process,
we have judgement, which is "thought in its completed
form". And in considering the nature of judgement, we may
see more fully why it is that no psychicaHact is an idea.
"We can understand this most clearly if we consider the
nature of judgement, for there we find thought in its com-

83

Bradley's Dialectic

pleted form. In judgement an idea is predicated of a reality.
Now, in the first place, what is predicated is not a mental
image. It is not a fact inside my head which the judgement
wishes to attach to another fact outside. The predicate is a
mere 'what', a mere feature of content, which is used to
qualify further the 'that' of the subject. And this predicate
is divorced from its psychical existence in my head, and is
used without any regard to its being there. When I say 'this
horse is a mammal', it is surely absurd to suppose that I am
harnessing my mental state to the beast between the shafts.
Judgement adds an adjective to reality, and this adjective
is an idea, because it is a quality made loose from its own
existence, and is working free from its implication with that.
And, even when a fact is merely analysed when the
predicate i appears not to go beyond its own subject, or to
have been imported divorced from another fact outside
our account still holds good. For here obviously our syn-
thesis is a re-union of the distinguished, and it implies a
separation, which, though it is over-ridden, is never un-
made. The predicate is a content which has been made loose
from its own immediate existence and is used in divorce from
that first unity." 144 In a judgement, the "what" which is its
content, exists not in a stage of mere self-transcendence
referring to a subject beyond itself. Rather, the "what" of
a judgement exists as referred to a subject. This is to say
that the moment of mediation or self-transcendence or
relation that is thinking, is, in judgement, fully referred
to the subject which, in and through that judgement, this
thinking comes to qualify. Thus the initial partial estrange-
ment of the "what" from its "that" is healed in the union
of that "what" with the further psychical fact which is the
proximate subject of the thought thus completed in judge-
ment. "Judgement is essentially the re-union of two sides,
'what' and 'that', provisionally estranged. But it is the
alienation of these aspects in which thought's ideality con-

The Internality of Thought and Reality

sists." 176 In the re-union of "what" and "that," however,
the what need not be re-united with the very same psychical
fact from which initially it was estranged. For the whati
may be referred to, and finally joined with, new psychical
fact which would be quite different from the old.

Indeed, once it be considered that any case of the ideality
of thought is a case of the self-transcendence that is of the
essence of the internality of relations, it is then difficult to
see how the reunion that is judgement could be a mere
repetition of the psychical fact in which the content of the
judgement had its origins. For the "what", in passing beyond
its "that", is altering its context of relations. This is to say
that the "relations", by which the what is differentiated,
are being altered. And in any alteration of its relations the
"what" itself is being altered. Therefore, the "what" that is
re-united in judgement may not be qualitatively identical
with what it was at the inception of the moment of ideality;
nor, for that matter, at any stage in the course of the passing
of that moment of thinking over into the fulfilment of it in
judgement. It would seem, then, that both the "what" which
is re-united in judgement and the relational context of this
consummation could only be different from the initial "what"
and its "that". Indeed, since identity implies qualitative as
distinguished from numerical difference, a mere repetition
of content would be impossible.

The sharpness of the contrast drawn by Bradley between
the ideality of self-transcendence, and the comparative
inanition of psychical fact, might suggest that most psychical
fact is congealed sentience, barely alive. Worse still, the
suggestion might be conveyed that the ideal is a light that
never was on sea or land, except as a sort of aura diffused
here and there on earth by precious minds. This sort of
misunderstanding of the matter Bradley takes up as "a
most important point". "There exists a notion that ideality
is something outside of facts, something imported into them,

85

Bradley's Dialectic

or imposed as a sort of layer above them; and we talk as if
facts, when let alone, were in no sense ideal. But any such
notion is illusory. For facts which are not ideal, and which
show no looseness of content from existence, seem hardly
actual. They would be found, if anywhere, in feelings
without internal lapse, and with a content wholly single." 146
Within a content without internal lapse there would be no
differentiations. Hence there would be within it no relations,
no process of fission through which it would transcend itself.
Any such content would be hardly actual; it would be this
content and mine, but at a level of sentience almost below
that of discrimination.

Nevertheless, and no matter how lethargic any such
moment of sentience might be within itself, it would be
that lethargic moment and different from all else. Although
barely actual or definite, or hardly explicit, that content
would differ from its relational context. To that extent
it would be ideal. And any moment of sentience, however
lethargic and smug, will be a different moment. Hence any
content, even though it be almost devoid of differentiation
within itself as immediacy or quality, still will be incipient
with change. For it will be the quality it is by virtue of its
relations. These will be neither lethargic nor smug, but
actively alive. As those relations change, the content they
differentiate will change also.

This much is true of any content or fact. That is why
any fact, however subliminal and undifferentiated within the
immediacy of it, nevertheless is in change and so in a refer-
ence beyond itself that is incipient at least.

Hence Bradley continues with the text quoted above as
follows: "But if we keep to fact which is given, this changes
in our hands, and it compels us to perceive inconsistency of
content. And then this content cannot be referred merely
to its given 'that', but is forced beyond it, and is made to*
qualify something outside. But, if so, in the simplest change

86

The Internality of Thought and Reality

we have at once ideality the use of content in separation
from its actual existence." 146 There is ideality in the simplest
change because any change in any content ipso factcTis a
changeTn the relations which differentiate it, and by which
it transcends itself. The change that is this self-transcen-
dence is ideal, for the ideality of the finite consists in the
self-transcendence of psychical fact. And this transcendence
consists in the process of relational becoming that is Appear-
ance. Through that self-transcendence the content of a
psychical fact comes to be referred beyond its "that". "For
the content of the given is for ever relative to something
not given, and the nature of its 'what' is hence essentially
to transcend its 'that'. This we may call the ideality of the
given finite. It is not manufactured by thought, but thought
itself is its development and product. The essential nature of
the finite is that everywhere, as it presents itself, its character
should slide beyond the limits of its existence." 146 Such is
the character of the finite because everywhere and always
the finite is relational and in process: it is Appearance. The
ideality of the finite is anything but intellectual and factitious;
rather it is the very nature of the incessant fission in sentience
which issues in the self-transcendence of psychical fact that
is thought. Hence the sense in which thought is thought, the
sense in which thought is ideal, and the sense in which the
internality of relations is in process, are all one and the same.
Having thus explained briefly the sense in which thought
is properly said to be ideal, and how it is that thinking is
completed in judgement, Bradley now proceeds to indicate
what is for him the nature of truth. "Truth is the predication
of such content as, when predicated, is harmonious, and
removes inconsistency and with it unrest. And because the
given reality is never consistent, thought is compelled to
take the road of immediate expansion. If thought were
successful, it would have a predicate consistent in itself and
agreeing entirely with the subject. But, on the other hand,

87

Bradley's Dialectic

the predicate must be always ideal. It must, that is, be a
'what' not in unity with its own 'that', and therefore, in and
by itself, devoid of existence. Hence, so far as in thought
this alienation is not made good, thought can never be more
than merely ideal." The truth attainable by thought, and
attained in judgement, may not be absolute. Thought and
judgement are ineluctably relational. A thought, that is,
to be non-relational or^absolute, would require "a pre-
dicate consistent in itself and agreeing entirely with the
subject". Such a thought would be a tautology; for the
predicate A would "agree entirely with," i.e. would repeat,
the subject A. And such a thought would be intrinsically
tautologous. For A p , in being itself absolutely, would be
intrinsically individuated; and thus would be true also of A s .
Therefore A p and A s would be no more than numerically,
or enumerably, different from each other. Numerical
difference would be a merely external relation, and any such
notion is excluded by the internality of relations.

Thus thought may not choose to be tautological: it is
ineluctably relational; which is to say that in thought
"the predicate must be always ideal". What is predicated
in a process of thinking may be only a "what" which is in the
process of self-transcendence. This process of self-transcend-
ence is the very process of predication which is completed
in the re-union of the "what" with a new "that". Without
this fact of self-transcendence, there would be no process in
sentience at most there would be only changeless states of
feeling in succession. Indeed, within any such changeless
sentience, there would hardly be even a plurality of states.
For as changeless, sentience could not become differentiated
at all, and so it would remain everywhere and always
homogeneous and undifferentiated. No such sentience could
give rise to so much as a vague question about itself, much
less a thought, or a judgement. For in being altogether
unmediated, such sentience could only be undifFerentiated,

88

The Internality of Thought and Reality

or relationless, and hence mere psychical fact devoid of
all thought.

In the absence of all becoming or self-transcendence, there
could be no thinking, but only mere psychical fact. But
change is an experience whose actuality can hardly be
questioned. Where there is change at all there is differentia-
tion and thus "inconsistency of content," or self-transceh-
dence. And the "what" which thus is transcending its
"that" "cannot be referred merely to its given 'that' but is
forced beyond it, and is made to qualify something outside."
And that is why it is that "in the simplest change we have
ideality "; for within no matter what sort of change there will
be differentiation, and we have noticed that in any quality
or relation there is self-transcendence. Indeed, self-tran-
scendence is but another name for the relational character
of appearance. Qualities transcend themselves as qualities
in so far as they are related: relations likewise transcend
themselves as relations in so far as they contribute to con-
stitute their qualities.

Yet, no matter how elaborate the origins of a thought,
and no matter how comprehensive and self-consistent the
resulting judgement may be, this judgement will fall short of
absolute truth. "For the content of the given is for ever
relative to something not given, and the nature of its 'what'
is hence essentially to transcend its 'that'. This we may call
the ideality of the finite. It is not manufactured by thought,
but thought itself is its development and product. The
essential nature of the finite is that everywhere, as it
presents itself, its character should slide beyond the limits of
its existence." 146 Even so, it may be asked, why is thought
not able through the internality of relations to arrive at the
positive nature of the Absolute? For the internality of rela-
tions is the very being in process of thought and of judgement.
And to this the answer is again the same in principle.

Absolute knowledge of the Absolute would be either (i) a

Bradley 's Dialectic

knowledge that were related to, and by virtue of that very
relation, differentiated from the object known, or (2) this
knowledge would be not relational at all, but rather identical
with what thus would be known. On the first alternative
the knowledge in question still would be relational and
therefore not absolute knowledge. Moreover, were anything
whatever held to be related to or with the Absolute, that
Being itself would forthwith become relational, and thus it
would cease to be Absolute.

On the second alternative, the knowledge in question
would be one with its object. Such alleged knowledge
would consist of a judgement in which the subject and the
predicate were qualitatively identical. And this tautology
would be the suicide of thought and judgement. ' ' If_ there
is no judgement, there is no thought; and if there is no
difference, there is no judgement, nor any self-consciousness.
But if, on the other hand, there is a difference, then the
subject is beyond the predicated content." Were there no
differences anywhere in sentience, there could be no self-
consciousness and no thinking; for all would be one and
homogeneous. And this differentiation which is the
self-transcendence of process must be a difference between
qualities: it may not be the mere numerical difference of
two qualitatively identical items. For such items would be
self-identical, and therefore changeless. The differentiation
or self-transcendence which is the active nerve of thinking
and judging must be qualitative. Consequently and inevit-
ably in thought and in judgement there will be a difference
between the subject of thinking and what is predicated of
that subject in the completed judgement. An absolute, or
non-relational, judgement is a tautology; and in the bare
repetition of tautology we have the suicide of judgement.

In whatever form we may have sentience, it will be in
becoming; differentiations will be breaking out within it,
and the qualities thus emerging will be related by those very

90

The Internality of Thought and Reality

differentiations. Qualities as qualities, we have seen, are the
moments of immediacy in sentient process. And relations
are the moments of differentiation in that process. The
self-transcendence of quality by relation is the ideality of
thought and the content of judgement. It is because thought
and judgement are thus essentially relational that, no matter
how comprehensive and internally coherent a judgement
may become, it can only fall short of being absolutely
self-coherent. "Thought is relational and discursive, and,
if it ceases to be this, it commits suicide; and yet, if it remains
thus, how does it contain^immediate presentation?" 150 The
conclusion to which we are driven is that if, by "immediate
presentation", we mean something absolute in its own right,
then thought neither contains nor attains immediate presen-
tation. The immediacy of the "this" and the "mine" is, we
have noticed, at once immediate and relational, thus
illustrating in an acute form the essential incoherence of
any partial whole of quality and relation. Any whole in
judgement, however elaborate and comparatively self-
coherent, will still fall short of the absolutely self-coherent
Individual. Yet, again it may be urged, "A harmonious
system of content predicating itself, a subject self-conscious
in that system of content, this is what thought should
mean." 150 But this system would be simply the systematic
whole of internal relations. For, since that whole would be
sentience, it would be a systematic whole conscious of itself.
Yet, here again, no advance has been made. What is
conscious "of" the content of the systematic whole either
is somehow different from that content, or it is not different
from it. The first alternative gives us self-consciousness,
but it gives it to us as the feding of a systematic whole from
which that feeling remains distinct; and, so far, excluded
from that whole. The second alternative gives us no self-
consciousness at all no consciousness, that is to say, distinct
from that of the systematic whole of sentience itself.

Bradley's Dialectic

Though any attempt to exhibit the positive character of
the Absolute in and through thought be fore-doomed to
failure by the essentially relational nature of judgement,
still the theory of predication which Bradley presents in
Note A affords another clue to the function of the Absolute
in judgement, and so in Appearance itself. If, in thinking,
we would have more than a mere association of ideas, there
must be in some sense the assertion of "unity in diversity."
Differences in no sense united are merely different; and, on
the other hand, it is an old story that the formula of thought
may not be "A is A". Yet the requirement that thought
must unite differences is not an easy one to fulfil. If the
formula for significant thinking may not be A is A, neither
may it be A is Y. For here the predicate Y, being different
from A, is not A. Since this would be to assert and to deny
in the same thought, the process of uniting differences in
judgement may not be expressed in the form "A is Y".

The alternative that, in thinking A is Y, we are really
meaning "A has Y", is of no avail. For either "A has Y"
means no more than "A is A and has Y", where the "has"
expresses bare conjunction; or "A has Y" means that "A
is-such-as-to-have Y". On the first alternative the connec-
tion of Y with A remains unexplained; on the second, the
copula is again introduced, and the original dilemma
remains. The one possible method of resolving this dilemma,
Bradley concludes, is afforded by the notion of identity in
difference. Judgements uniting differences are themselves
adjectives of Reality, which is a systematic Whole. The
absolute identity of this Whole is the ultimate identity in all
differences. Hence, it is ultimately in virtue of this identity
that A and Y are united in thought.

In the statement of a judgement the subject will be
grammatically distinct from the predicate. But nothing in
judgement itself corresponds to this grammatical distinction.
For both of the terms which contribute to constitute a

92

The Internality of Thought and Reality

judgement are adjectival. This means that the subject and
predicate terms are alike in qualifying a reality which lies
beyond those constituents of the judgement itself. Nor may
it be otherwise within the internality of relations. For the
two terms in question will be related in virtue of their
qualitative difference. Again, in virtue of their difference
from their proximate subject that to which they initially
refer they are related with that subject of their reference.
And their reference, or their being referred to that proximate
subject, will consist in the very differentiations by which those
terms are related to that subject. Moreover, the respective
identities of those terms imply their differences from, and
so their relations to, all else. This is true also of the identity
of the relational situation which is their proximate subject.
Thus the ultimate subject of reality is the absolute identity
of the systematic whole. It is to this, ultimately, that every
judgement refers. Nor can there be any mystery as to the
nature of that reference. For it is in virtue of the internality
of the relations of its contents to all else, that the final
subject of any judgement is ultimate Reality.

Thus we may see that on any view of the nature of
relations that is cognate with that of Bradley, the question,
what is the nature of the relation between thought and
reality? would be a meaningless question. It would be
meaningless because there is and could be no single or
unique relation that were "the relation" between thought and
reality. The object of thought is not in any sense independent
of what is thought. With Bradley's dialectic of relations in
mind, the internality of thought and reality will be plain
enough. For the content of any judgement, in virtue of its
difference from its proximate subject, is related with that
subject. And the proximate subject, with that related content,
likewise are related within and to the whole of the systematic
whole that is Appearance.

93

CHAPTER V

The Coherence Theory of
Truth and Reality

BRADLEY deals with the problem of error on his way to his
theory of truth, and we shall examine his theory of error as a
preface to the main topic of this, chapter. "Error is without
any question a dangerous subject, and the chief difficulty is i
as follows: We cannot, on the one hand, accept anything
between non-existence and reality, while, on the other hand,
error obstinately refuses to be either. It persistently attempts;
to maintain a third position, which appears nowhere to)
exist, and yet somehow is occupied. In false appearance
there is something attributed to the real which does not
belong to it. But if the appearance is not real, then it is
not false appearance, because it is nothing. On the other
hand, if it is false, it must therefore be true reality, for it is
something which is. And this dilemma at first sight seems
insoluble." 164 ' 165

Error, it would seem, must be real in some sense: yet an
error neither may belong to the Absolute, nor may it
qualify, nor contribute to constitute any erroneous judge-
ment; for all judgements^are m and of the Real. Yet it is
being erroneous That is the defining characteristic of the
erroneous judgement. Thus it would seem that error both
must be, and yet may not be real.

Accordingly, the two main questions with which Bradley
is here concerned are (i) why error may not be real, and
(2) the sense in which error is (as it must be) real. 166 Error "is
at any rate one kind of false appearance. Now appearance is

94

The Coherence Theory of Truth and Reality

content not at one with its existence, a 'what 5 loosened from
its 'that 5 ." 168 And, "appearance will be truth when a
content, made alien to its own being, is related to some fact
which accepts its qualification. The true idea is appearance
in respect of its own being as fact and event, but is reality
in connection with other being which it qualifies. 55166 As
ideal, a true idea is an appearance in that the ideality of the
finite is the process of self- transcendence that is Appearance.
But a true idea, as a content of completed thinking or
judgement, is real to the degree to which it coheres with the
proximate subject that it qualifies. "Error, on the other hand,
is content made loose from its own reality, and related to a
reality with which it is discrepant. It is the rejection of an
idea by existence which is not the existence of the idea as
made loose. It is the repulse by a substantive of a liberated
adjective.' 5166 And in a footnote to v this passage, Bradley
adds that "whether the adjective has been liberated from
this substantive or from another makes no difference 55 .
An erroneous judgement will be one whose content is, to a
degree, incoherent with the proximate subject to which that
content is referred.

But this is not to be taken to mean that the content qua
content is erroneous. To be sure, the identity of the content
is relational, and so the content itself will illustrate the
comparative incoherence of any relational reality. Yet this
comparative incoherence is not error. Nor will error result
before judgement is attempted and completed. For it is in
the predication of a content which is not compatible with
its proximate subject that error exists. Thus it is not the
relational incoherence or self-discrepancy of the content
itself that is error; rather it is the failure of the content
predicated of the proximate subject to cohere (to any but a
slight degree) with the subject it is judged to qualify.

The proximate subject, S, will be a certain relational
situation. As such it will be a relational concretion having a

95

Bradley's Dialectic

certain character, which will be different from that of any
other relational situation. Now predicate A, let us say, will
be more fully coherent with the character of S than will
predicate Z. This is to say that all the differentiations by
which A is related with S are indefinitely more elaborate
than are those by which Z is thus related. That being the
case, the predication of A of S would result in a judgement
whose content were elaborately relevant to, and therefore
coherent with, the subject thus qualified. Such a judgement
would be true to a high degree. If, on the contrary, predicate
Z were predicated of S, then there would be but little
coherence of that predicate with the subject of the judgement.
To be sure Z could not be out of all relation to S; for nothing
may be quite external to anything else. But Z has so little
bearing on (i.e. so few, or such tenuous relations with) S,
that the character of S repels Z when it is judged to qualify
S with a relevance of which Z is not capable in that
connection.

Lest the meaning of the term "relevance" here seem to be
in need of a definition which, it may seem, could only be
circular, it may be well to take up that point before going
on. The term "relevant", it may be urged, is not definable
without circularity; for any definition offered would have
to be relevant. The force of this objection is specious, on
Bradley's theory of appearance and reality.* For the relevant
is the internally related. All appearances are internally
related: hence Bosanquet's dictum, "all is relevant to all".
Nor does this bi-verbal definition beg the question. For by
this definition the meaning of "relevance" is simply identified
with that of the internality of relations; and (as is even
obvious) the meaning of that doctrine is not thus called in
question.

In an erroneous judgement, some content or other is

* It is also specious on other grounds that need not be entered into
in this connection.

96

The Coherence Theory of Truth and Reality

alleged to be relevant to a subject in respects in which that
content is not thus relevant. " Suppose that when William
has been hung, I assert that it was John. My assertion will
then be false, because William is certain. And if so, then after
all my error surely will consist in giving to the real a self-
discrepant content. For otherwise, when John is suggested,
I could not reject the idea." 163 Here the error lies in the
discrepancy between the relevance of the living character
of John and the nature of William's being hung. And
taking the judgement as being completed in its alleged
reference* to its subject, this discrepancy between the
adjective and substantive terms of the judgemental situation
will be the self-discrepancy of the judgement, taken as the
total situation. It is this discrepancy between asserted
relevance, and the actual substantive situation, that is at
once error in judgement and, seemingly, a flat inconsistency
in monistic theory.

Turning to the "second main problem of the chapter", 169
that "about the relation of error to the Absolute", 169
Bradley gives the following statement of the matter to be
explained: "There is no way but in accepting the whole
mass of fact, and in then attempting to correct it and make
it good. Error is truth, it is partial truth that is false only
because partial and left incomplete. The Absolute has
without subtraction all those qualities, and it has every
arrangement which we seem to confer upon it by our mere
mistake. The only mistake lies in our failure to give also the
complement. The reality owns the discordance and the
discrepancy of false appearance; but it possesses also much
else in which this jarring character is swallowed up and is
dissolved in fuller harmony. I do not mean that by a mere
re-arrangement of the matter which is given to us, we could
remove its contradictions. For, being limited, we cannot
apprehend all the details of the whole. And we must remem-

* Where the term "reference" means what is meant by "relevance".

97 o

Bradley's Dialectic

her that every old arrangement, condemned as erroneous,
itself forms part of that detail. To know all the elements of
the universe, with all the conjunctions of those elements,
good and bad, is impossible for finite minds. And hence
obviously we are unable throughout to reconstruct our
discrepancies. But we can comprehend in general what we
cannot see exhibited in detail." 169 ' 17 The statement that
"error is truth" may seem merely paradoxical; yet it
expresses literally a part of Bradley's theory of error.

We have noticed how error arises. When a content, Z, is
almost irrelevant to S, and yet is asserted to qualify S, then
we have error. This error is not a nature or form that is
distinct from the internally related content which is the
adjective, and the relational situation which is the sub-
stantive, of the judgement. For the error will consist of the
almost complete failure of the content Z to qualify S. And
this failure is no nature or form of any sort. Rather it is
almost complete irrelevance.

Now, any case of relevance will be some case or other of
the internality of relations realized in a judgement. This
judgement will be an appearance; and it will be real to the
degree to which it is self-coherent. But irrelevance, taken as
utter irrelevance, has no referent anywhere in appearance
or reality. The utterly irrelevant would be the non-relevant.
This verbiage would have no bearing (beyond that of its
existence as a verbal construction) within the internality of
relations, and so it would have no logical bearing (as
distinguished from a merely verbal ontological standing)
at all. Thus, in so far as the content of an erroneous judge-
ment is logical or coherent with the subject of that judgement,
to that extent the judgement is true. The rest of it is verbiage.
For whatever of a judgement is not true is irrelevant, and
what is irrelevant may have no referent within the realm
of the internality of relations. Short of the very best that
Rimbaud, Peguy, and Stein have produced, there will be

98

The Coherence Theory of Truth and Reality

no utter irrelevance, and so no absolute error. Failure in
relevant qualification will usually be but partial: when
utter meaninglessness is achieved, this will be utterly non-
logical. Such achievements in the use of language, however,
are very rare. Almost always the erroneous judgement will
be more than irrelevant and merely verbal. However
slightly, it will still qualify the real; and, to that degree, it
will be, not the irrelevance that is mere error; rather, it will
be true to a certain degree.

By the emendation of its content, an erroneous judgement
may be made less erroneous, or more coherent with its sub-
ject, and so more true. Thus "error is truth, it is partial truth,
that is false only because partial and left incomplete." 169
As the truth of a judgement becomes less partial, less in-
complete, the content of the judgement will be more
fully related to the subject. The judgement will be the
more relevant, or the more true. It is in this failure to be
relevant that error lies, and that failure is nothing positive:
it is not in any sense a quality or a relation. "The only
mistake lies in our failure to give also the complement." 170
For by any emendation of the original content of an
erroneous judgement, the initial error is transcended. And
by an emendation that were almost all-comprehensive, we
would arrive at a judgement that were but slightly erroneous.
But to accomplish so much as this would be practically
impossible. "I do not mean that by a mere re-arrangement
of the matter which is given to us, we could remove its
contradictions." 170 Since our outlook is limited, we may not
apprehend anything like all of the details of the systematic
whole of internal relations. Moreover, as Bradley reminds us,
we must not forget "that every old arrangement, condemned
as erroneous, itself forms part of that detail." This brings
us to the third and final one of the main topics of Bradley' s
chapter on error.

Up to this point, we have noticed that error lies in the

99

Bradley's Dialectic

conflict between the relevance (or the internal relations)
of a content predicated of a subject, and the character of
the subject which repels the predication of that content.
We have noticed also that, in so far as error is a failure in
relevance and so is irrelevance, error is nothing real. An
erroneous judgement is significant, and a judgement at all,
only in so far as it is true. And by a progressive emendation
of an erroneous judgement, the initial coherence of it with
the real may be deepened and widened. Thus, so far, it
would seem that error is merely negative, or unreal. "But
our account, it will fairly be objected, is untenable because
incomplete. For error is not merely negative." 172 Because the
content, predicated as thoroughly relevant, fails to fulfil the
intended predication. This failure is nothing real, to be sure.
But the difference between Z (the jhredicate which is mainly
irrelevant to S) and A (the predicate which is almost fully
relevant to the character of S) is a relation. And this relation
(this difference between Z and A), like any other relation,
is not unreal. It constitutes the reason why it must be
admitted that error is not merely negative. For though
irrelevance is verbiage, still the difference between the
predicate of a judgement that is largely irrelevant, and any
other content that is relevant to the subject of that judge-
ment, is real.

Error is not a special sort of appearance; taken as a
characteristic of some (and only some) judgements, error is
unreal. Any judgement, in being less than wholly self-
coherent, will be less than wholly true. Hence any judgement
will be erroneous to a degree.

But if error is no peculiar disease, amenable to a logical
cordon sanitaire, none the less any error is real. The reality of
it consists of the relations that are the differences between the
predicate predicated by a judgement that is erroneous and
the quality of the subject that repels that predicate. Those
relations are as real as any others. And they may be any

100

The Coherence Theory of Truth and Reality

relations whatever; not relations or differences of a peculiar
sort.

The relations that constitute the reality of an error are
of no special brand; the difference between those relations
as they contribute to constitute that judgement, rather than
any other, is a difference of degree. For the difference
between a judgement that is erroneous and one that is true
is a difference not of kind. Any judgement is at once true
and erroneous. A judgement that is more comprehensively
self-coherent than not, is properly called true. A judgement
that is more fully self-discrepant than coherent, is properly
called an error.

Far from being a peculiarity of some judgements only,
error is of the nature of all judgements, to some degree or
other. For a judgement is a relational situation, and the
relational is everywhere and always self-discrepant: not
absolutely, of course; that would be the blank of a pure
negation; but self-discrepant to a degree.

Error, then, is all-pervasive in appearance; for the self-
discrepant reality of error consists of the degrees of self-
discrepancy that are the differentiations in the process of
becoming. Thus, in reality error is not different in kind from
truth. An error is a judgement that is more comprehensively
self-discrepant a judgement in which more contrarieties
break out between the proximate subject and the predicate
than our aspiration to self-coherence can tolerate.

And the reality of any error itself is self-discrepant
because, taken as degrees of contrariety, errors are not self-
sufficient. Our aspiration to the attainment of complete
self-coherence makes us discriminate between degrees of
self-discrepancy and self-coherence. " Error is truth, it
is partial truth, that is false only because partial and left
incomplete. The Absolute has without subtraction all those
qualities, and it has every arrangement which we seem to
confer upon it by our mere mistake. The only mistake lies

101

Bradley's Dialectic

in our failure to give also the complement. ... I do not
mean that by a mere re-arrangement of the matter which is
given to us, we could remove its contradictions. For, being
limited, we cannot apprehend all the details of the whole.
And we must remember that every old arrangement,
condemned as erroneous, itself forms part of that detail.
To know all the elements of the universe, with all the
conjunctions of those elements, good and bad, is impossible
for finite minds. And hence obviously we are unable through-
out to reconstruct our discrepancies. But we can comprehend
in general what we cannot see exhibited in detail." 169 ' 17
We can understand in principle that, through progressive
emendation, we can render our judgements less self-
discrepant and thus more self-coherent. "Error is truth when
it is supplemented." 170 An erroneous judgement is true to the
degree to which it is self-coherent, and the truth of it is
increased as the scope of that coherence is enlarged.

Bradley' s insistence on the continuity of error with truth
might well lead an ill-disposed reader of Appearance and
Reality to the conclusion that there is no difference between
error and ignorance. Our judgements are always as coherent
as we can make them; where we fail to make them more so
than they are, our failure is in no sense a mistake, but
merely a matter of ignorance.

This would be to overlook die point that in an erroneous
judgement there is present an insistence that the judgement
is true, not efroneous. "For the point of error, when all is
said, lies in this very insistence on the partial and discrepant,
and this discordant emphasis will fall outside of every
possible rearrangement. I admit this objection, and I
endorse it. The problem of error cannot be solved by an
enlarged scheme of relations." 172 That could not be the
whole of the solution, because it is just the insistence on
what is self-discrepant that sustains the conviction that an
erroneous judgement is not in error, or is, at least, not as

102

The Coherence Theory of Truth and Reality

erroneous as it is. Any emendation of the judgement in the
direction of more self-coherence will dissipate that insistence,
to be sure; but the fact of the past existence of it will remain.

Thus Bradley recognizes that it is not enough to explain the
ontological nature of error. Take any view of that question
you prefer, and a further question still remains; namely,
why a man who is in error feels or insists that he is in the
right ? Bradley answers that "Error is, but is not barely
what it takes itself to be. And its mere onesidedness again is
but a partial emphasis, a note of insistence which contributes,
we know not how, to the greater energy of life. And, if so,
the whole problem has, so far, been disposed of." 173 This
follows because the insistence itself, in being but a one-sided
emphasis, ipso facto is differentiated from, and thereby
related to, all else. Thus while the fatuous insistence charac-
teristic of any erroneous judgement is real and forcible, still,
at the same time, the relational status of it within Appear-
ance is in no wise peculiar, but simply that of any other
appearance.

However, the question of the status of error in Bradley's
metaphysics is thus dealt with only "so far", and not com-
pletely. So far we have seen why Bradley holds that "error"
is not the name of any special sort of appearance. -The
question remains as to how the self-discrepant may be taken
up into the Absolute. For, as Bradley reminds us in this
connection, "the Absolute is not, and cannot be thought as,
any scheme of relations. If we keep to these, there is no har-
monious unity in the whole. The Absolute is beyond a mere
arrangement, however well compensated, though an arrange-
ment is assuredly one aspect of its being. Reality consists,
as we saw, in a higher experience, superior to the distinc-
tions which it includes and overrides." 172 Yet, since relations,
however badly mistaken and mixed up in judgement, still
are relations, there remains the question as to how such judge-
ments, along with all appearances, stand to the x Absolute.

103

Bradley 5 s Dialectic

We shall follow Bradley in deferring consideration of this
question to a subsequent chapter. At the close of his discus-
sion of Error (ch. XVI), Bradley foreshadows the answer
he gives to the question thus deferred. "We have pointed out
that it is at least possible for errors to correct themselves,
and, as such, to disappear in a higher experience. But, if^
so, we -must affirm that they are thus absorbed and made
good. For what is possible, and what a general principle
compels us to say must be, that certainly is." 178

With the nature and status of error and irrelevance thus
considered and partially, at least, accounted for in Bradley's
view of the matter, we may now turn to the dialectic of
degrees of truth and reality. That truth for us may be in
no case absolute, we have noticed in several connections;
and here, at the beginning of his theory of truth, Bradley
reminds us that there may be no degrees in the Absolute.
"The Absolute, considered as such, has of course no degrees;
for it is perfect, and there can be no more or less in perfection
(chapter XX). Such predicates belong to, and have a
meaning only in the world of appearance." 318 What is itself
absolutely, may not be itself "more or less". On the contrary,
no relational fact ever is less than absolutely self-coherent.
Hence, in no judgement, properly so called, may the con-
notation of the predicate term coincide wholly with that of
the subject term. No matter how fully self-coherent the
terms of a judgement may be, "there is still a difference, un-
removed, between the subject and the predicate, a difference
which, while it persists, shows a failure in thought, but
which, if removed, would wholly destroy the special think-
ing". 319 For were there no self-discrepancy in sentient
process, there would be no self-transcendence. And without
the transcendence of the "that" by the "what", there would
be no thinking. "We have already perceived the main
nature of the process of thinking. Thought essentially
consists in the separation of the 'what' from the 'that'.

104

The Coherence Theory of Truth and Reality

It may be said to accept this dissolution as its effective
principle." 311

Consequently, no judgement may but fall short of full
self-coherence. "We may put this otherwise by laying down
that any categorical judgement must be false. The subject
and the predicate, in the end, cannot either be the other.
If however we stop short of this goal, our judgement has
failed to reach truth; while, if we attained it, the terms
and their relations would have ceased. And hence all our
judgements, to be true, must become conditional. The
predicate, that is, does not hold unless by the help of some-
thing else. And this 'something else' cannot be stated, so as
to fall inside even a new and conditional predicate." 320 *
No matter how elaborately a judgement might be expanded
by the progressive emendation of it, still there would remain
"something else" which fell beyond that judgement; for
no judgement may be more than relational, or less than
absolute. The conditional nature of judgement is the rela-
tional nature of process. And the scope of the relational re-
gress in virtue of which any relational situation is itself, we
have seen to be endless for the way of thought that is in-
eluctably ours. Although a given judgement, J 1 , may be so
expanded as to exhibit, in J 2 , many qualities and relations
relevant to (and, therefore, conditions of) J 1 , still the truth of
J 2 can only fall short of being absolute. \/

To be sure, on Bradley's view, it is not inconceivable that
this process of expansion, by the progressive emendation of
increasingly comprehensive judgements, might lead to a
concrete universal that fell but little short of being absolutely
self-coherent. Even so, there would remain a "something
else" not comprehended by that ail-but omniscient judge-
ment. This something else would be, of course, reality not
as relational and self-coherent more or less, but reality as

* "Even metaphysical statements about the Absolute, I would add,
are not strictly categorical." P. 320 n.

105

Bradley's Dialectic

unconditioned and absolute. "But with this we have arrived
at the meeting ground of error and truth. There will be no
truth which is entirely true, just as there will be no error
which is totally false. With all alike, if taken strictly, it will
be a question of amount, and will be a matter of more or
less. Our thoughts certainly, for some purposes, may be taken
as wholly false, or again as quite accurate; but truth and
error, measured by the Absolute, must each be subject
always to degree. Our judgements, in a word, can never
reach as far as perfect truth, and must be content to enjoy
more or less of Validity"* 321 The criterion of absolute
truth being the Absolute, truth unqualified would exist in a
judgement whose subject and predicate terms were identical.
Absolute truth would be the knowing and the being in one
of an omniscient tautology.

If, out of carelessness, haste in speaking, or in virtue of
habitual references to a set of partial standards, we advance
judgements as though they were quite true or wholly false,
this is permissible only in so far as it may prove useful
and convenient. But no such usage is admissible in meta-
physics; nor is it in any way a comment on metaphysical
principles. Judgements are true as appearances are real.
To the degree, or to the extent, to which a judgement
is comprehensively self-coherent, it is true; and, mutatis
mutandis ', this is so of any appearance whatever.

There is but one absolute individual, the Absolute.
Ultimate reality alone is wholly individual; for it alone
is absolute in its identity with itself. "Perfection of truth
and reality has in the end the same character. It consists
in positive, self-subsisting individuality. . . . Truth must
exhibit the mark of internal harmony, or again, the mark
of expansion and all-inclusiveness. And these two charac-
teristics are diverse aspects of a single principle. That which
contradicts itself, in the first place, jars, because the whole,
immanent within it, drives its parts into collision. And the

106

The Coherence Theory of Truth and Reality

way to find harmony, as we have seen, is to redistribute
these discrepancies in a wider arrangement. But, in the
second place, harmony is incompatible with restriction and
finitude. For that which is not all-inclusive must by virtue
of its essence internally disagree; and if we reflect the reason
of this becomes plain. That which exists in a whole has
external relations. Whatever it fails to include in its own
nature, must be related to it by the whole, and related
externally. Now these extrinsic relations, on the one hand,
fall outside of itself, but, upon the other hand, cannot do so.
For a relation must at both ends affect, and pass into, the
being of its terms. And hence the inner essence of what is
finite itself both is, and is not, the relations which limit it.
Its nature is hence incurably relative, passing, that is, beyond
itself, and importing, again, into its own core a mass of
foreign connections. Thus to be defined from without is, in
principle, to be distracted within." 321 ' 322

The length of this quotation may be justified in so far as
in it we find the internality of relations to be the content
of judgement and so of truth. Anything less than the sys-
tematic whole of qualities in relation will have relations
which fall beyond that relational situation; relations which
are " external " to (but not separate from) the contents of that
situation. Since these extrinsic relations contribute to the
identity of that situation no less than do those which lie
within it, the situation, in being limited from without, is
distracted within itself. And so, here again we have the
contrariety that is of the being or essence of relational
experience. Any quality, however comparatively simple or
however elaborate it may be, will be at once itself as felt
immediacy, while, at the same time, it transcends itself in
being the relations by which it is differentiated from all
else. It is not surprising, then, that the internality of relations
should constitute the content of judgement and so of truth.
For the identity of any appearance, however coherent it be,

107

Bradley's Dialectic

still is relational. But for all that, the internality of relations
does not supply us with a criterion of truth: rather, it carries
that criterion with it, therein to be found and understood.
And, since the criterion of absolute reality is an Individual
that is absolutely self-sustaining, the criterion of truth and
reality, for Bradley, is not far to seek. The more compre-
hensive in intension and extension a relational situation
becomes, the more fully real it is. For to the degree to which
it is inclusive, a relational situation approaches absolute,
self-sustaining reality. Likewise, as a judgement is expanded
in relevance, it loses in irrelevance in becoming thus more
fully coherent. This "internal harmony", or self-coherence,
is the truth of that judgement. "Comprehensiveness" and
"internal self-coherence", or "harmony", are different names
not of as many difference criteria, but rather of a single
principle, that of degrees of coherence within and among the
concrete universals that constitute the self-fulfilling processes
that are Appearance at every level short of their absolute
self-fulfilment in the Absolute harmony that is Reality.

1 08

CHAPTER VI

Identity in Difference

THE title of this chapter designates the logical principle
of the internality of quality and relation that is the coherence
of degrees of truth and reality. Some understanding of that
principle will carry us further into Bradley's theory of truth,
and it will also prepare us the better to ask about how, in
Bradley's own metaphysics, the relational stands to the
Absolute.

Bradley holds that philosophy is an attempt to gain a
view of reality that will satisfy the intellect. He finds, we
have seen, one mark, albeit a negative one, of what is
intellectually satisfactory. The intellect rejects the self-
contradictory, and accepts the self-consistent. But just
what is the contradictory? We have seen something of the
answer Bradley gave to that question in the Logic. Now we
turn to his most extensive discussion of the matter; namely,
Note A appended to Appearance and Reality.

Contradiction is not explicable in terms of opposites.
"If we are asked 'What is contrary or contradictory ?' (I
do not find it necessary here to distinguish between these),
the more we consider the more difficult we find it to answer.
'A thing cannot be or do two opposites at once and in the
same respect' this reply at first sight may seem clear, but on
reflection may threaten us with an unmeaning circle. For
what are 'opposites' except the adjectives which the thing
cannot so combine ? Hence we have said no more than that
we in fact find predicates which in fact will not go together,
and our further introduction of their 'opposite' nature

109

Bradley's Dialectic

seems to add nothing." 500 To say of predicates that they
are "opposites" is to explain nothing about those predicates
or about anything else.

The view that opposites are predicates that cannot be
united is shortsighted. For "if one arrangement has made
them opposite, a wider arrangement may perhaps unmake
their opposition, and may include them all at once and
harmoniously." 500 The mistake that takes opposites to be
opposed eternally is the mistake that takes diverse beings
to be merely other than each other, unmediated by any
common ground. But no beings are thus merely external
to each other; no beings are wholly themselves and simply
not each other. Rather, beings are different, and different
beings may and do exist in a unity.

If there is to be a unity of any sort, then "otherness"
must be a word without a referent; and the fact of unity is
a fact whose actuality is beyond dispute. But a whole can
hardly be the blank of vacuous homogeneity. Without
internal distinction, there would be nothing within the whole
that it could unite. " 'A thing cannot without an internal
distinction be (or do)* two different things, and differences
cannot belong to the same thing in the same point unless
in that point there is diversity. The appearance of such a
union may be fact, but is for thought a contradiction.'
This is the thesis which to me seems to contain the truth
about the contrary, and I will now try to recommend this
thesis to the reader." 601 Were a whole without internal
diversity, there would be nothing in it to be united; and
that there may be a union of differences, there must be
diversity within that unity.

This thesis is not a statement of the Law of (abstract)

Identity. That law states no more than the emptiest of

tautologies, A is A. "Thought most certainly does not

demand mere sameness, which to it would be nothing. A

* "This addition is superfluous," Footnote, p. 501.

110

Identity in Difference

bare tautology (Hegel has taught us this, and I wish we could
all learn it) is not even so much as a poor truth or a thin
truth. It is not a truth in any way, in any sense, or at all.
Thought involves analysis and synthesis, and if the Law of
Contradiction forbade diversity, it would forbid thinking
altogether. And with this too necessary warning I will turn
to the other side of the difficulty. Thought cannot do without
differences, but on the other hand it cannot make them.
And, as it cannot make them, so it cannot receive them
merely from the outside and ready-made. Thought demands
to go proprio motu, or, what is the same thing, with a ground
and reason. Now to pass from A to B, if the ground remains
external, is for thought to pass with no ground at all. But if,
again, the external fact of A's and B's conjunction is offered
as a reason, then that conjunction itself creates the same
difficulty. For thought's analysis can respect nothing, nor is
there any principle by which at a certain point it should arrest
itself or be arrested. Every distinguishable aspect becomes
therefore for thought a diverse element to be brought to unity.
Hence thought can no more pass without a reason from A or
from B to its conjunction, than before it could pass ground-
lessly from A to B. The transition, being offered as a mere
datum, or effected as a mere fact, is not thought's own self-
movement. Or in other words, because for thought no ground
can be merely external, the passage is groundless. Thus A and
B and their conjunction are, like atoms, pushed in from the
outside by chance or fate; and what is thought to do with
them, but either make or accept an arrangement which to
it is wanton and without reason or, having no reason for
anything else, attempt against reason to identify them
simply?" 501 This lengthy passage contains the gist of
Bradley's theory of identity and predication.

In the course of all thinking whatever there is the assertion
of unity in diversity. Judgement cannot dispense with unity;
for, without it, there would be at best a bare association of

in

Bradley's Dialectic

ideas. And the barren unity of repetition in tautology is
not thought at all, but mental inanity that is "filled in with
the verbiage C A is A'." Hence judgement also cannot
dispense with differences, for without differences there is
bare tautology, and thus no movement of thought at all.
Hegel has taught us this, and Bradley wishes that all of us
would learn it.

The formula of judgement cannot be A is A, for thought
must unite differences if there is to be any movement or
transition in thought at all. But if A is A cannot be the ,
formula of judgement, can that formula be A is B? Evidently
not. For B is different from A. Therefore B is not-A. Thus
the formula in question appears to say, A is not-A; it
yields abstract contradiction, utterly unsatisfactory to the
intellect.

But, it may be objected, this is a pseudo-problem which
arises out of your failure to distinguish between the "is"
of identity and the "is" of predication. When we say "A is
B", surely no one takes us to mean that A is identical with
B. Rather, we are understood to mean what in fact we do
mean; namely, that B is a predicate of A, or in a word, that
A has B.

Yet this comment on the matter is fatuous. For to mean
that A has B is to mean that A is such-as-to-have-B. The
alleged distinction between the "is" of predication and the
"is" of identity enables us to do no more than change our
verbal symbol for the copula, and thus merely evade the
central question as to how the copula may in logic as in
existence unite differences.

That question, it may be objected, so far from being
central, is quite unreal. But " 'This is not so 5 , I shall be told,
'and the whole case is otherwise. There are certain ultimate
complexes given to us as facts, and these uJtimates, as they
are given, thought simply takes up as principles and employs
them to explain the detail of the world. And with this process

112

Identity in Difference

thought is satisfied.' To me such a doctrine is quite erroneous.
For these ultimates (a) cannot make the world intelligible,
and again (b) they are not given, and (c) in themselves they
are self-contradictory, and not truth but appearance." 501 502
These ultimates cannot be used to explain the world at all,
because no one of them bears on anything other than itself.
In being each one final and, as it were, all there at once:
ultimate and given: each ultimate is self-contained, and
thus isolated in experience and logic from every other one.
Such ultimates, and the complexes they constitute without
residuum, are external to each other. No one of them
contributes anything at all to any other one; hence they are
respectively without bearing on each other.

Then again, these alleged complexes are not given. "The
transition from A to B, the inherence of b and c as adjectives
in A, the union of discretion and continuity in time and
space 'such things are facts,' it is said. 'They are given to an
intellect which is satisfied to accept and to employ them.'
They may be facts, I reply, in some sense of that word, but
to say that, as such and in and by themselves, they are given
is erroneous. What is given is a presented whole, a sensuous
total in^hjgl^j^hese characteris are found; and beyond
and beside these characters there is always given something
else. And to urge 'but at any rate these characters are there,'
is surely futile. For certainly they are not, when there, as
they are when you by an abstraction have taken them out.
Your contention is that certain ultimate conjunctions of
elements are given. And I reply that no such bare conjunction
is or possibly can be given. For the background is present,
and the background and the conjunction are, I submit,
alike integral aspects of the fact. The background therefore
must be taken as a condition of the conjunction's existence,
and the intellect must assert the conjunction subject in this
way to a condition. The conjunction is hence not bare but
dependent, and it is really a connection mediated by

113 H

Bradley's Dialectic

something falling outside it." 502 ' 503 Facts given u in and by
themselves" would be facts without background; without
relations beyond themselves. We have looked into the
dialectic of quality and relation which constrains Bradley
to deny the reality of any such unrelated collocation of
(what are for him) fictions. And now we find him reminding
us that any set of "facts" we may hit upon or select stands in
a context of relations, not stark and alone. With his theory
of relations in mind, we can understand the force of the
"must" in his statement that this context must be taken as
a condition of the existence of any set of facts. They could
not be what they are without their context. For that context
of relation and quality, which, ultimately, is the systematic
whole that is appearance, is the relational situation in
virtue of which those "facts" are differentiated from all
else and thereby made the qualities and relations that
they are.

That is why the mere conjunction of characteristics is a
self-contradictory notion. "And any mere conjunction, I
go on to urge, is for thought self-contradictory. Thought, I
may perhaps assume, implies analysis and synthesis and
distinction in unity. Further, the mere conjunction offered
to thought cannot be set apart itself as something sacred,
but may itself properly, and indeed must, become thought's
object. There will be a passage therefore from one element
in this conjunction to its other element or elements. And
on the other hand, by its own nature, thought must hold
these in unity. But, in a bare conjunction, starting with A,
thought will externally be driven to B, and seeking to unite
these it will find no ground of union. Thought can of itself
supply no internal bond by which to hold them together,
nor has it any internal diversity by which to maintain them
apart. It must therefore seek barely to identify them, though
they are different, or somehow to unite both diversities
where it has no ground of distinction and union. And this

114

Identity in Difference

does not mean that the connection is merely unknown and
may be affirmed as unknown, and also, supposing it were
known, as rational. For, if so, the conjunction would at
once not be bare, and it is as bare that it is offered and not
as conditional. But, if on the other hand it remains bare,
then thought to affirm it must unite diversities without any
internal distinction, and the attempt to do this is precisely
what contradiction means." 504

So long as the differences in question are taken as self-
contained units, thought can only recoil from the affirmation
of their identity. For as respectively different and self-
contained, they have no common ground; there is nothing
internal to them in virtue of which they could be at one.
Such self-contained units are self-identical; the formula
for them is the empty truth A is A. And on this, to Bradley
a wholly perverse view of the matter, the formula for the
Law of Non-Contradiction is no less familiar and no less
inane. But if, instead of trying to take experience as a
complex of self-contained units, we see it as diverse expres-
sions of a system of mutually related qualities, the whole
matter is altered in principle. For now we see that to
predicate B of A is not to affirm that A per se is B per se. A is
not A intrinsically and independently of all else, and no
more is B. Rather A is what it is by virtue of its differentiations
from B and all else. These differentiations are relations by
which A is related throughout the systematic whole of
appearances, as B also is thus related by its differences from
all else.

Thus when we judge that A is B, there now is and could
be no question of our affirming that A is identical with B.
For neither A nor B is such that they could be merely
identical with each other. The identity of A 3 we have seen,
implies its difference from all else, and that of course
includes and also applies to B. By virtue of the differentiations
by which A is related throughout within the Whole, A is a

Bradley's Dialectic

relational adjective of the Whole. And, mutatis mutandis, so
also is B.

Within the verbal terms of a judgement, it is grammatically
correct to distinguish a predicate term and a subject term.
But this is a distinction in point of terms verbal; a distinction
to which nothing in judgement proper corresponds. For in
truth both of the constituents of the judgement proper, as
distinguished from the statement of it in words, are
adjectives of the Whole.

This means that the constituents of a judgement proper*
are related at once to their proximate subject a perceived
hammer, say and also to the Whole itself. For the con-
stituent terms of the judgement that is expressed by the
sentence, "that is a hammer," are not the same as, but
rather are different from, that instrument. In virtue of their
differences from it, those terms are related with and so
refer to that hammer. And we must not forget that those
terms and that hammer are different from and ipso facto
related to all else. This "all" will be the systematic Whole
of Appearances.

So much follows from the conclusion that relations and
qualities are mutually internal. Every aspect of experience
is related throughout the Whole in virtue of its differences
from everything else within that Whole. For this reason,
among others, the subject term in every judgement no less
than the predicate term is an adjective of the real. The
subject and the predicate, S and P, in being different
content are related to each other by the very fact of their
differences; and since S and P are different also from all
other moments of process, ipso facto S and P are related to
every single moment of the Whole.

Thus the judgement S is P asserts a diversity of connota-
tion in the adjectives S and P; while, at the same time, that
judgement asserts an identity in the denotation of those
adjectives through which the ultimate referent of that

116

Identity in Difference

judgement is organic Reality. In Bradley's view, this must
be so, for S and P are internally related throughout that
Whole. Thus in every judgement there is an identity in the
differences that are the content of that judgement. This
identity is the Absolute Reality that is the supra-relational
fulfilment of the development of process in appearance.

Thus we may see that ' 'Things are not contrary because
they are opposite, for things by themselves are not opposite.
And things are not contrary because they are diverse, for
the world as a fact holds diversity in unity. Things are self-
contrary when, and just so far as, they appear as bare
conjunctions, when in order to think them you would have
to predicate differences without an internal ground of
connection and distinction, when, in other words, you would
have to unite diversities simply, and that means in the same
point. This is what contradiction means, or I at least have
been able to find no other meaning." 605

Things qua things are not contrary; contrariety breaks out
as unity among difference develops. This contrariety is
diversity, not the otherness of blank negation between two
unrelated conjuncts. That bare negation yields self-con-
trariety: "Things are self-contrary when, and just so far as,
they appear as bare conjunctions." 505 For such self-identical
ultimates are "without an internal ground of connection and
distinction"; and we have seen something of the logical
results of attempting to unite any such fictions in judgement.
u Thought cannot accept tautology and yet demands unity
in diversity. But your offered conjunctions on the other side
are for it no connections or ways of union. They are them-
selves merely other external things to be connected. . . .
How can thought unite except so far as in itself it has a
mode of union? To unite without an internal ground of
connection and distinction is to strive to bring together
barely in the same point, and that is self-contradiction." 505
That is self-contradiction for the reason that to assert the

117

Bradley's Dialectic

identity of S and P is to contradict their difference. And to
affirm that difference is to contradict the attempted assertion
of identity.

Thus we see that the true formula of judgement is not
A is A, nor A is B, nor A has B. Rather the formula is Xa is
Xb; where X stands for the systematic Whole throughout
which a and b are related, and to which they refer by virtue
of those relations. To take it that A per se is B per se yields,
we have seen, the flat contradiction of a judgement which
affirms unity of contents which are posited as merely different.
Diverse contents may be united only as they are adjectives
of the systematic Whole that is the unity or principle of
identity in all differences.

118

CHAPTER VII

The Relational and the
Absolute

*!T is a central conclusion of Bradley's dialectic, and one
which he frequently emphasizes, that appearances are
sublated or taken up into the Absolute. How this sublation
of the relational by the Absolute is to be explained is the
topic of the present chapter.

Bradley writes, not once but again and again, that the
Absolute is in some sense beyond and superior to the
relational. For example, in writing of the correction of
errors and the relation of erroneous judgements to the
Absolute, he says, "But on the other side the Absolute is not,
and cannot be thought as, any scheme of relations. If we
keep to these, there is no harmonious unity in the whole.
The Absolute is beyond a mere arrangement, however well
compensated, though an arrangement is assuredly one aspect
of its being. Reality consists, as we saw, in a higher experience,
superior to the distinctions which it includes and over-
rides." 172 The Absolute, then, is "beyond" relations; it is a
"higher experience", which "overrides" the distinctions to
which it is "superior".

Before going on to ask about what it means to say even
this much of the Absolute, it may be well to remind our-
selves of what Bradley elucidates as being the relational
nature of Appearance. As a propaedeutic to the dialectic of
Relation and Quality in chapter III, Bradley urges in
chapter II that because relations can be neither attributes of,
nor independent of, their terms, they must be internal to

Bradley's Dialectic

them. Relations can not be attributes because predication
is either tautological or flatly self-contradictory. "If you
predicate what is different, you ascribe to the subject what
it is not\ and if you predicate what is not different, you say
nothing at all." A distinction between the "is" of identity
and the "is" of predication is not mentioned as an alternative
on which this dilemma would be resolved, because that
distinction would beg the question of the reality of external
relations. Rather, it is suggested that the relation in question
be regarded as "more or less" independent of its terms/
"But such a makeshift leads at once to the infinite process."
For on it the question, "what relates the independent
relation to its terms?", remains and breaks out afresh at
each attempted answer. It is thus concluded that only one
alternative remains: "A is in relation to B" means neither
that "A" is identical with nor independent of "in relation
with B"; rather it means that A and B stand in a relation
that differentiates the qualitative characters of A and B
respectively.

The assumption that every difference is a difference in
quality, and that numerical difference is not an alternative
at all, underlies the argument of chapter III, "Relation and
Quality." We have seen that in chapter III Bradley goes
on to urge that relations and qualities are not found apart
in fact, and that they can not be separated by any process of
abstraction. Mere separateness or otherness is not possible:
where there is separation there is difference; and difference,
to be at all, must be difference in quality. "For consider, the
qualities A and B are to be different from each other; and,
if so, that difference must fall somewhere. If it falls, in any
degree, or to any extent, outside A or B, we have relation
at once. But, on the other hand, how can difference and
otherness fall inside? If we have in A any such otherness,
then inside A we must distinguish its own quality and its
otherness." 24 Either the difference between A and B, in

120

The Relational and the Absolute

virtue of which they are distinct, differentiates A from B and
so "falls outside" A or B, or that difference "falls within''
A and B respectively thus to differentiate them within
themselves indefinitely. Hence the difference that differ-
entiates A and B must fall outside or between them. This
difference can not be numerical merely, for that would be
a difference that in no way contributes to qualify what it
differentiates. And on Bradley 5 s assumption that "difference"
is the name of a relation, a difference that in no way con-
tributed to the character of its qualities would be a relation
"separate" from its terms, and so not a genuine differ-
entiation at all.

It could hardly be denied that distinct qualities cannot
be distinct unless they are different in some sense. This
difference, if we be constrained to say so, "must fall some-
where"; and this "somewhere" must be "outside" A or B.
Thus "we have relation at once". 24 For in falling outside
A or B, their difference falls between them, and thus that
differentiation relates A and B.

In a synoptic way, the main conclusions of chapters II
and III of Appearance and Reality may be seen to derive from
a pair of alleged disjunctions In the first of these, that of
chapter II, it is urged that relations cannot be both separate
entities on the one hand, and capable of relating terms on
the other. And since any view of relations as separate
entities entails an infinite regress in separate entities that
can only fail to relate terms, this alternative is a delusion,
or worse. Hence we are forced to conclude, on what is
assumed to be the sole alternative remaining to us, that
relations are internal to their terms; where "being internal
to" means "affecting" or "making a difference in" the
qualitative character of the terms thus related. What is
meant in this connection by "affecting" or "making a
difference in" is explained in the course of the statement of
the second disjunction.

121

Bradley's Dialectic

Just as (in chapter II) there may be no relations without
qualities, so also (in chapter III) there may be no distinct
qualities without relations. Qualities without relations
would be not diverse qualities at all, but undifferentiated
and at one. In chapter III ("Relation and Quality") the
alleged disjunction is between an infinite regress of differ-
ences within the character of every single quality, on the
one hand, and relations internal to the character of the
qualities they differentiate, on the other; where "being
internal to" means being the very differences by which*
qualities are differentiated. Therefore relations (i.e.
differentiations) must "fall somewhere". They cannot be
and yet have no status. Now the differentiations that rela-
tions are can not fall wholly within qualities. For that results
in the indefinite differentiation of qualities within them-
selves. Hence relations must fall at once within and between
the qualities they differentiate. An example of the meaning
of relation as thus making a difference in the terms that
conversely make a difference in their relations is afforded
by cell-fission. In this sense of the term relation, "difference
implies relation" for the reason that to relate is to differen-
tiate. As the differentiation of the moments of fission alters,
the moments themselves alter; and, conversely, as the
moments of fission alter, their differentiation also alters.

The question which concerns us here is how the relational
is taken up into the Absolute. In some quarters it is some-
times suggested that the theory of the "this" and the
"mine" affords an intimation, which is all we can have, of
how this question is answered in the fact of feeling. It may
be well, then, to consider first the nature of this alleged
intimation.

We are told that "The 'this' and the 'mine' are names
which stand for the immediacy of feeling, and each serves
to call attention to one side of that fact. There is no 'mine'
which is not 'this', nor any 'this' which fails, in a sense,

122

The Relational and the Absolute

to be 'mine'. The immediate fact must always come as
something felt in an experience, and an experience always
must be particular, and, in a sense, must be unique." 197
Moreover, "all our knowledge, in the first place, arises from
the 'this'. It is the one source of our experience, and every
element of the world must submit to pass through iy And
the c this', secondly, has a genuine feature of ultimate
reality. With however great imperfection and inconsistency
it owns an individual character." 198 The "this" is also "mine"
for the same reason that any experience of yours is at once
that experience and yours. Such feelings are immediate in
that no feelings are less differentiated internally, or more
self- transcendent; this feeling of mine is compelling largely
in and through its sentient focus, with little felt fringe of
prehensions into the sentient context of it. Nevertheless, the
"this" and the "mine", the "that" and the "yours" are self-
transcendent. "And the self- transcendent character of the
'this' is, on all sides, open and plain. Appearing as immediate,
it, on the other side, has contents which are not consistent
with themselves, and which refer themselves beyond. Hence
the inner nature of the 'this' leads it to pass outside itself
toward a higher totality. And its negative aspect is but one
appearance of this general tendency. Its very exclusiveness
involves the reference of itself beyond itself, and is but a
proof of its necessary absorption in the Absolute." 201 ' 202 The
"this" and the "mine", though immediate, are less than the
Whole, and are in process. Therefore the "this" and the
"mine" exhibit, albeit in its incipience, that self-transcend-
ence which is the principle of differentiation and thus of
relation. Nevertheless, the "this" and the "mine" afford "the
sense of immediate reality" wherein, it is sometimes urged,
there is intimated the answer to our question about the rela-
tional and the Absolute.

Now there are at least two difficulties in the way of our
accepting this suggestion. In the first place, it would seem

123

Bradley's Dialectic

that the theory of the this and the mine denies what it also
affirms. Consider that the this and the mine "is at a level
below distinctions". "Both the 'this' and reality, we may say,
are immediate. But reality is immediate because it includes
and is superior to mediation. It develops, and it brings to
unity, the distinctions it contains. The 'this' is immediate,
on the other side, because it is at a level below distinc-
tions." 198 ' 199 Now below the level of distinctions there are
no distinctions. Where there are no distinctions there are no
differences. At a level below distinctions, the "this" could
not be distinct from the "that".

This is not to forget that the immediate experience in
question is also self-transcendent. The "this" and the
"mine" has two sides, "positive" and "negative".* On its
negative side the "this" and the "mine" is self-transcendent,
as is any other relational situation. But on its positive side
it is below the level of distinctions. And where there are no
distinctions there are no differences. Therefore it is impos-
sible that there should be a plurality of positive sides.
Without that plurality of positive sides there cannot be a
plurality of immediate experiences. In so far as the "this"
and the "mine" is below the level of distinctions it is in-
determinate because it is not distinct from anything else.
And it is only in so far as the "this" and the "mine" is
below the level of relations that it is immediate.

* "An immediate experience, viewed as positive, is so far not exclusive.
It is, so far, what it is, and it does not repel anything. But the 'this'
certainly is used also with a negative bearing. It may mean 'this one*,
in distinction from that one and the other one. And here it shows ob-
viously an exclusive aspect, and it implies an external and negative
relation. But every such relation, we have found, is inconsistent with
itself (ch. III). For it exists within, and by virtue of an embracing unity,
and apart from that totality both itself and its terms would be nothing.
And the relation also must penetrate the inner being of its terms. 'This',
in other words, would not exclude 'that', unless in the exclusion 'this',
so far, passed out of itself. Its repulsion of others is thus incompatible
with self-contained singleness, and involves subordination to an includ-
ing whole. But to the ultimate whole nothing can be opposed, or even
related." 201

124

The Relational and the Absolute

In the second place, why does the " this" and the "mine",
and no other mode of the relational, give us an intimation
of the nature of Absolute Reality? Because "the 'this' and
the 'mine' express the immediate character of feeling, and
the appearance of this character in a finite centre." 198 As
immediate experience, "the 'this' is real for us in a sense in
which nothing else is real". 198 It has " a genuine feature of
ultimate reality", for in its immediacy the "this" "owns
an individual character". 198 The ownership is transitory,
5^et the individuality thus possessed is alleged to be genuine
and felt.

But it is only on the positive, unmediated side of the "this"
that it is immediate experience; on the negative side of it,
the "this" is self-transcendent. The "this" is immediate
experience, and possessed of a feature of ultimate reality,
only in so far as it is unmediated, or "below the level of
distinctions". And at that level of experience there would be
sentience that were wholly homogeneous. The "this" could
not be distinct from the "that"; there could not be even
two contents, for below the level of distinctions all would
be one.

This homogeneity of sentience could hardly be the
Absolute into which all differences are sublated. "Both the
'this' and reality, we may say, are immediate. But reality
is immediate because it includes and is superior to media-
tion. It develops, and it brings to unity, the distinctions it
contains." 198) 199 Thus Reality is said at once to comprehend
and be superior to the distinctions it contains. But the "this"
contains no distinctions. "The 'this' is immediate, on the
other side, because it is at a level below distinctions." 199
The contrast is fairly broad. Reality includes in its superiority
the distinctions it contains. The "this" contains no distinc-
tions; it is below them all.

Nevertheless, it may be urged, the "this" does afford
an intimation by analogy of the nature of the real, In so

125

Bradley's Dialectic

far as the "this" is unmediated by distinctions it is immediate
and homogeneous and therefore one. The Absolute is one.
Thus we may see that Appearance is to Reality as the nega-
tive or mediated side of the "this" is to the positive or
unmediated side of it. But this suggestion is anything but
helpful. How does the relational, the mediated, the self-
transcendent aspect of the "this" stand to the aspect of
it that is undifFerentiated, homogeneous, and wholly one?
How can any such immediacy be at once devoid of internal
differences and possessed of internal relations by which it
could transcend itself and exist at a relational level? Without
differentiation or mediation there are and can *be no rela-
tions. At a level below distinctions, sentience would be
below relations. It would be out of all relation; it could not
stand to anything at all. On the side of the "this" and the
"mine" the proposed analogy is vitiated by the nature of
the case.

On the side of Appearance and the Absolute the analogy
is likewise unavailing. For "to the ultimate whole nothing
can be opposed, or even related". 201 The Absolute, being
absolute, is not relational. But what is the analogy proposed
if not a relation? Appearance is to the Absolute as the
mediated is to the immediate in the "this" and the "mine".
And even if this analogy were instructive, it still would be
fatal to the character of the Absolute as absolute. For if the
Absolute stood in any relation at all, ipso facto it would be
relational and appearance, and thus fail to be absolute.
The proposed analogy can only fail to answer the question
as to how Appearance can be taken up into, resolved,
made one with, or comprehended in the Absolute. For the
analogy in question would establish a relation between
Appearance and the Absolute. That would destroy the
absolute nature of ultimate reality.

There are those who urge that the main question at issue
in this connection is unreal; that it arises out of a failure to

126

The Relational and the Absolute

understand. The Absolute is no supra-relational being.
Rather it is nothing more than the systematic whole that is
exhaustively constituted by the internality of qualities and
relations. There is no relevant question about how Appear-
ance stands to the Absolute. For Appearance, not as self-
fulfilling, but as self-fulfilled, is the Absolute. The Absolute
is nothing distinct from the systematic totality of relation
and quality in the mutual harmony which their self-fulfil-
ment would realize and internally sustain.
This interpretation, and others cognate with it, were
anticipated by Bradley, and he flatly rejected it. The antici-
pation runs as follows: Writing of the "infinite process" in re-
lation and quality, Bradley says: u The remedy might lie here.
If the diversities were complementary aspects of a process
of connection and distinction, the process not being external
to the elements or again a foreign compulsion of the intellect,
but itself the intellect's own proprius motus, the case would be
altered. Each aspect would of itself be a transition to the
other aspect, a transition intrinsic and natural at once to
itself and to the intellect. And the Whole would be a self-evident
analysis and synthesis of the intellect itself by itself. Synthesis here
has ceased to be mere synthesis and has become self-completion,
and analysis, no longer mere analysis, is self -explication. And the
question how or why the many are one and the one is many
here loses its. meaning. There is no why or how beside the self-
evident process, and towards its own differences this whole is at once
their how and their why, their being, substance and system, their
reason, ground, and principle of diversity and unity."* This passage
would seem to make it clear that Bradley foresaw the inter-
pretation that is in question. The Absolute would be one
with the self-completing and self-explicating whole of rela-
tional quality. That systematic whole, in its self-fulfilment,
would be its own rationale. Thus there could be no question
about the relation of the relational to the real. For the real
* P. 507. My italics.

127

Bradley's Dialectic

is the relational, as the relational is self-fulfilling and self-
fulfilled.

But Bradley not only anticipated this interpretation of
the matter in one paragraph, he also rejected it in the
next. "Has the Law of Contradiction anything here to
condemn? It seems to me it has nothing. The identity of which
diversities are predicated is in no case simple. There is no
point which is not itself internally the transition to its
complement, and there is no unity which fails in internal
diversity and ground of distinction. In short 'the identity
of opposites,' far from conflicting with the Law of Contra-
diction, may claim to be the one view which satisfies its
demands, the only theory which everywhere refuses to
accept a standing contradiction. And if all that we find were
in the end such a self-evident and self-complete whole,
containing in itself as constituent processes the detail of
the Universe, so far as I can see the intellect would receive
satisfaction in full. But for myself, unable to verify a solution
of this kind, connections in the end must remain in part mere
syntheses, the putting together of differences external to
one ^another and to that which couples them. And against
my intellectual world the Law of Contradiction has there-
fore claims nowhere satisfied in full. And since, on the other
hand, the intellect insists that these demands must be and
are met, I am led to hold that they are met, in and by a whole
beyond the mere intellect. And in the intellect itself I seem to find an
inner want and defect and a demand thus to pass itself beyond itself.
And against this conclusion I have not yet seen any tenable objection"

"Self-existence and self-identity are to be found, I would
urge, in a whole beyond thought, a whole to which thought
points and in which it is included, but which is known
only in abstract character and could not be verified in its
detail." 607 ' 508 The Absolute, then, is a whole not merely of,
but also beyond thought; it is supra-relational.*
* Cf . Appearance and Reality, pp . 159-161.
128

The Relational and the Absolute

Yet the interpretation of his metaphysics that Bradley
discards is not rejected by him because it is self-stultifying.
He does not find it absurd that the real be viewed as a
systematic whole of mutually implicative qualities and rela-
tions; an organic whole that is self-fulfilling and therefore
self-fulfilled. Nor does he think it absurd that within that
systematic whole analysis and synthesis in judgement would
be respectively self-explicative and self-completed. Rather
he concludes that the intellect finds no such logically
stable systematic whole, or concrete universal that is fully
concrete.

Everywhere and always judgement falls short of ultimate
self-coherence. Any judgement will transcend itself by virtue
of the relational way of thought that is ineluctably ours.
Bradley finds in the intellect "an inner want and defect and
a demand to pass beyond itself" 508 because, as he has
explained over and over again, self-transcendence is of the
essence of thought. That is why for the intellect to go beyond
the relational to the Absolute would be for the intellect to
commit suicide.

The Absolute is not to be discerned in any view of relation
and quality as a systematic whole. No more is it to be found
in any one aspect of Appearance. "We have seen that the
various aspects of experience imply one another, that all
point to a unity which comprehends and perfects them.
And I would urge next that the unity of these aspects is
unknown. By this I certainly do not mean to deny that it
essentially is experience, but it is an experience of which,
as such, we have no direct knowledge. We never have, or
are, a state which is the perfect unity of all aspects: and we
must admit that in their special natures they remain in-
explicable. An explanation would be the reduction of their
plurality to unity, in such a way that the relation between
the unity and the variety was understood. And everywhere
an explanation of this kind in the end is beyond us. If we

129 i

Bradley's Dialectic

abstract one or more of the aspects of experience, and use
this known element as a ground to which the others are
referred, our failure is evident." 414 ' 415 For the aspects
abstracted would remain abstractions, no matter how we
went about insisting that they were the ground of all Appear-
ance. Each aspect would remain an aspect; thus it would
be incomplete, unstable, inherently self-transcendent. No
aspect of experience could be absolute, if only because it
is an aspect.

Thus we may see that, in Bradley's own judgement, the
Absolute is to be discerned neither in any phase of experi-
ence, nor in Appearance taken as a systematic whole.
Nevertheless he urges repeatedly that the relational is taken
up into a higher unity of logical harmony in which the
inherent self-discrepancy of appearance is fully resolved and
thus healed. And yet, it may be asked, what grounds has
Bradley for this insistence ? They are not afforded by any
one aspect, or any range of aspects, of Appearance. And the
systematic totality of appearances is less than self-grounded,
or self-contained. But those grounds are supplied by the
theory of negation upon which Bradley stands.

The result of Book I of Appearance and Reality is "mainly
negative", 119 we are assured. The aspects of Appearance
therein examined reveal the self-contrariety that is the
fission of relational quality through which this quality
that is also mine is differentiated from all else. The "this"
and the u mine" is any quality. Thus we are constrained to
conclude that Appearance throughout is self-discrepant
or self-contrary. By this very same conclusion, Bradley
argues, we are constrained to pronounce the contrary
of Appearance self-consistent. Thus, in finding that every
appearance is self-contrary to some degree or other, ipso
facto we posit the contrary of our negation; namely, a self-
consistent, and therefore absolute Reality.

This conclusion is truly inescapable on the theory of

130

The Relational and the Absolute

negation that is characteristic of systematic Idealism. On
that theory, we have seen, every negation presupposes a
positive ground. "Pure negation", "mere otherness",
"external relations", these phrases are alike without meaning,
for there is no mere negation. "The contradictory idea, if
we take it in a merely negative form, must be banished from
logic. If Not- A were solely the negation of A, it would be an
assertion without a quality, and would be a denial without
anything positive to serve as its ground. ... It is impossible
for anything to be only Not-A. It is impossible to realize
Not-A in thought."* The contradictory of A cannot be a
mere not-A. Any such conception must be abandoned, and
with it the empty, wholly verbal distinction between the
contradictory and the contrary. In truth they are one;
and in that fact we see that every opposition in psychical
fact, in thought, and in judgement is mediated by a third
term. This mediation of the differentiation that is a negation
is the positive ground of that negation.

Now every relation is a differentiation, and every dif-
ferentiation is a negation, and every negation requires
a positive ground. Consequently, the relational requires a
ground that is positive. Thus, both Bradley's criterion of
Appearance: namely, self-contrariety: and his theory of
negation, entail a ground that is self-coherent or wholly
positive.

Yet it is difficult to see how the Absolute that is the
identity in all differences may be in any sense related to
those appearances. For were the Absolute in any relation
whatever, ipso facto it would cease to be absolute; by the
very fact of that relation, the Absolute would be the term
of a relation, and therefore it would be relational, not
absolute. The notion of the Absolute as the identity in
differences: as the positive ground of the negations which
those differentiations are, to put the same thing in another

* Logic, p. 123.

Bradley's Dialectic

way: is as basic to Bradley' s metaphysics as is his theory of
negation. And without that conception of negation, the
Hegelian identification of the contradictory with the
contrary would lose its basis in theory. Nevertheless, it is
more than difficult to see how the Absolute can be in any
relation at all and remain absolute.

132

CHAPTER VIII

Some Basic Difficulties

BRADLEY'S monism is derived from his theory of relation and
Duality. The validity of that theory is a necessary condition
of the validity of his monism. It is the conception of relations
as internal to the terms they differentiate that excludes the
possibility of a plurality of externally related realities. It is
that same conception of relation that constrains us to con-
clude that every single reality is internally related to all else.
And it is the theory of relation as the aspect of self-transcend-
ence or self-differentiation that compels us to see thought, and
the consummation of thought in judgement, as the ideality
of the finite, ineluctably relational and hence for ever falling
short of the Absolute that is, nevertheless, implied by the
dialectic of quality and relation.

It may be well then to ask first about some of the more
obtrusive difficulties in Bradley' s theory of relations. These
initial questions will lead us in a natural order to some other
basic difficulties in Bradley's dialectic.

We have seen that, for Bradley, "quality" is the name of
any moment of experience wherein mediation or differen-
tiation is recessive, while immediacy is dominant. Con-
versely, "relation" is the name of any moment of experience
in which differentiation is dominant, while immediacy is
recessive. In this matter, the principal point is illustrated
by the process of cell-fission. If, while regarding this, we
attend mainly to the new cells that are emerging, then what
is most immediate, or qualitative, in that situation will
be dominant within the focus of our attention. If, to

133

Bradley's Dialectic

the contrary, our attention emphasizes the mediation
or differentiation there in process, then relation will be
dominant, while quality is recessive.

Thus conceived of, "qualities are nothing without

relations". 21 Should we attempt to arrive at a relationless

quality by abstraction, we could only fail. The process of

abstraction is a process of differentiation; therefore it is

relational. Again, we should be balked were we to "appeal

to a lower and undistinguished state of mind, where in one

feeling are many aspects. ... I admit the existence of such

states without any relation, but I wholly deny there the

presence of qualities". 22 For these felt aspects, if undifferen-

tiated, are not qualities; and if they are differentiated, then,

by that very fact, they are related. "In short, if you go

back to mere unbroken feelings, you have no relations and

no qualities. But if you come to what is distinct, you get

relations at once." 22 Where there are no distinctions, there

are no qualities: where there are qualities, there are

distinctions or differences and, by that very fact, relations.

Thus, on the assumption that a difference is a relation

and a relation is a difference, or differentiation, Bradley

submits that, where we find different qualities, there we

find qualities which are related by their differences. "For

consider, the qualities A and B are to be different from each

other; and, if so, the difference must fall somewhere. If it

falls, in any degree or to any extent, outside A or B, we have

relation at once. But on the other hand, how can difference

and otherness fall inside? If we have in A any such otherness,

then inside A we must distinguish its own quality and its

otherness. And if so, then the unsolved problem breaks out

inside each quality, and separates each into two qualities

in relation. In brief, diversity without relation seems a word

without meaning." 24 Either the difference between A and B,

in virtue of which they are distinct, "falls outside A or B",

thus to relate them; or else that difference falls within A

134

Some Basic Difficulties

and B respectively, thus to differentiate each one of them
within itself; and so the moment of relation would break
out within A, and within B. Hence the difference that
differentiates A and B can only fall "outside", or between
them, and so "we have relation at once". 24

Thus, to return again to our example, in the incipience
of a process of cell-fission, the incipient differentiation
"must fall somewhere". And it falls "outside" or "between"
the incipient cells in the observable sense that it is what
differentiates them. As soon as we see this, then (on the
assumption that differentiation, or difference in process,
is relation), we are aware of relation at once. For we are
then aware of the difference in virtue of which the incipient
qualities are differentiated, or related. In this sense of the
term "relation", were a quality without relations, it would
be in no wise different from anything else, and so would
fail to be anything at all.

To be sure, no fixed line can be drawn between a differ-
entiation and what is thus differentiated. For any relation
and any quality will be in process. As the incipient qualities
become more and more determinate, their differentiation
alters; and, conversely, as their differentiation becomes
more and more marked, the qualities likewise are altered.
"Hence the qualities must be, and must also be related. But
there is hence diversity which falls inside each quality.
Each has a double character, as both supporting and as
being made by the relation." 26

In order that a quality may be distinct, it must be differ-
entiated from other qualities. This differentiation is no
separate relation: rather, it contributes to constitute what it
differentiates. Thus, in so far as A is quality, A is not
relation; and yet, that it may be distinct, A must be both
itself and its differentiation. This is the reason why no
quality, however concrete and comprehensive,, may be
wholly self-coherent. That is why, as we have seen, "A is

135

Bradley's Dialectic

both made, and is not made, what it is by relation; and
these different aspects are not each the other, nor again is
either A. If we call its diverse aspects V and a, then A is
partly each of these. As V it is the difference on which
distinction is based, while as a it is the distinctness that
results from connection. A is really both somehow together
as A (a a). But (as we saw in chapter II) without the use of a
relation it is impossible to predicate this variety of A. And,
on the other hand, with an internal relation ^4's unity
disappears, and its contents are dissipated in an endless 1
process of distinction. . . . We, in brief are led by a prin-
ciple of fission which conducts us to no end". 26 Without a
relation, A would be undifferentiated, and so it would be
nothing at all. With a relation, A is at once the a that is
differentiated, and the ' V that is the differentiation. Thus
'M is partly each of these". Neither "a", nor "<z", is the other;
"nor again is either ^4"; for a is what is differentiated, while
"#" is the differentiation. And this differentiation, 'V, is
essential to the a which it differentiates.

That is why "A is both made, and is not made, what it
is by relation. ... It may be taken as at once condition
and result, and the question is as to how it can combine this
variety. For it must combine the diversity, and yet it fails
to do so". 26 A must be at once "a", the aspect differentiated,
and V, the phase of differentiation. Without V nothing is
differentiated, and so there is no quality: without V there is
no differentiation, and so there is nothing at all. Thus we may
see that it is divided within itself. It is at once itself as "a"
and not itself as the "<z" in virtue of which a is differentiated.
The same difficulty appears when experience is "taken
from the side of relations. They are nothing intelligible,
either with or without their qualities". Just as relations
apart from qualities are a delusion, so together with their
qualities relations are in no finite context completely intel-
ligible. For that a relation may differentiate its terms, it

136

Some Basic Difficulties

must "penetrate and alter" them, and so be implicated in
their respective natures. Yet, that this differentiation may
not disappear altogether, it must "fall" to some extent
"between" the qualities which it differentiates, and ipso
facto relates. That is why a relation involves within itself
a contrariety: a relation must be at once implicated in and
yet transcendent of its terms. And so "again we are hurried
off into the eddy of a hopeless process, since we are forced
to go on finding relations without end". 27 ' 28 For, in so far as
a relation or differentiation is implicated in its qualities,
it does not fall between them; and, in this respect, it fails to
be a relation at all: in so far as a differentiation falls between
qualities, it is outside them both, and so again fails to relate
them. Thus we may notice that relations, taken as moments
of differentiation which are at once implicated in and
transcendent of their qualities, are in no case self-consistent
realities; for they involve within themselves that "infinite
process" in relational identity which is the moving principle
and content of degrees of truth and reality.

In an unfinished, posthumously published essay on
Relations that was written some thirty years after chapter III
of Appearance and Reality was composed, Bradley reverts to
the "infinite process" of self-discrepant quality and relation.
"A relation (we find) holds between its terms, and no term
(we find) can itself simply be or become a 'between'. On
the contrary, in order to be related, a term must keep still
within itself enough character to make it, in short, itself and
not anything diverse. And again, while the relations are not
the terms and the terms are not the relations, neither the
terms nor the relations can make that whole, in which
nevertheless we find them. For the terms and the relations
. . . cease as such to exist, unless each maintains itself
against whatever is not itself but is outside. And the attempt
to find the required unity and totality in the terms and the
relations taken somehow together must end obviously in

137

Bradley's Dialectic

failure."* The attempt in question can end only in failure
because neither the terms of a relational situation, nor their
relation, may be self-consistent, or self-identical. For each
one of the two terms will be a moment of immediacy, while,
at the same time, it will transcend itself in being continuous
with the differentiation by which the two terms are differ-
entiated. Likewise, that differentiation will be internal to,
or confluent with, the terms which it relates; while, at the
same time, it will fall between them, and thus transcend
them.

This is why "A relation both is and is not what may be
called the entire relational situation, and hence in this
respect contradicts itself". t A relation is the whole relational
situation in the sense that it is what differentiates the
qualities of that situation. These qualities, were they differ-
entiated otherwise, would be different qualities. Conversely,
were the qualities of the situation different, their differ-
entiation would be different. As thus determining the
qualities of the situation, and so its own character as
a differentiation, "A relation to be actual cannot itself
be less than all and everything that makes the entire
relational fact."* For it is in virtue of the relation that the
qualities which it differentiates are the distinct moments
which they are. In a case of cell-fission, the qualities
which are being differentiated are confluent with their
differentiation; as, likewise, that relation is continuous with
those qualities. At any two points in the process of fission,
this relation may be marked off from its qualities, and its
qualities thus will be marked off from their relation. To do
this may well serve a purpose in practice, but it can establish
no final distinction; for, on this theory, no distinction short
of that between Appearance and the Absolute may be final.
Any distinctions, which were marked off, would themselves

* Collected Essays, p. 634 /. f Ibid., p. 635.

J Ibid., p. 636.

138

Some Basic Difficulties

be differentiations. These relations would break out, on the
one hand, between the moments of fission which we had
marked off as qualities, and, on the other, between what of
the process would then be marked off as the differentiation
of those new qualities. This "what" thus would become a
quality in its own right: for it would itself be differentiated
by the distinctions in question. Plainly, such distinctions
might be multiplied indefinitely within process. Since to
make distinctions is ipso facto to relate, no limit to the
relational constituents of a relational situation may be
established in judgment.

Yet we have seen that "This on the other hand must be
denied. For a relation is not its terms, but, on the contrary,
it is between them. And though the terms may 'enter into
the relation, 3 yet, if they were nothing beyond it, they
obviously would no longer be terms".* Within a relational
situation, differentiations may be found wherever distinctions
break out. It is in this sense that relations pervade and
determine the character of that partial whole; and it is in
this sense that they may be said to be not less than "all ...
that makes the entire relational fact". Still, in no case is the
differentiation identical with what is differentiated. The
qualities differentiated are, to be sure, continuous with their
relation. Yet, that either the quality or the relation may be
at all, each one must be distinct.

That there may be qualities at all, distinctions must
occur, or be made, and no distinction in Appearance may be
ultimate. Once the differentiation is marked off it becomes
distinct in thus being singled out; and hence, between it
and the initial qualities, fresh relations break out with
their qualities. No distinction, no differentiation in process,
may be self-sustaining or absolute. For a differentiation
will be at once continuous with its qualities and distinct
from them. In virtue of that distinction, and no matter
* Ibid., p. 636.

139

Bradley's Dialectic

where in the process it may break out, or where it be dis-
criminated, a fresh relation, with its own qualities, then
appears. This new relation, though continuous with its
qualities, yet is different from them. And thus, again,
there appear fresh relations with their new qualities. The
notion of ' 'independent relations" yields an "infinite process"
in relations that do not relate: the dialectic of relations
which differentiate their terms exhibits a process wherein
neither the differentiation, nor the quality differentiated,
is absolute in its own right, or self-identical. Hence the con-
clusion that no relational situation is wholly self-consistent,
or intelligible.

And so we may come to see that "Every relation (unless
our previous inquiries have led to error) has a connexion
with its terms which, not simply internal or external, must
in principle be both at once."* A relation must be internal
to its terms in the sense that, as their differentiation, it
contributes to constitute their qualitative character: yet,
if the relation is not to disappear altogether, it must, to some
extent, "fall between" its terms. And, in that sense, a rela-
tion is external to the qualities which it differentiates.
The central point for Bradley' s theory of the matter, that
relations are at once internal and external, is made out at
length in his unfinished essay. This is done not only in an
extended discussion of the point itself, but also in an explana-
tion of why Bradley thinks it "ludicrous" to suppose that,
on his view, relations could be merely internal. We shall
revert to this latter point in the next chapter.

With this body of doctrine freshly in mind, it will be the
easier to see why it is for Bradley that "identity implies
difference". We have noticed more than once that the
identity of a quality is what it is by virtue of the relations
that differentiate it that make it the quality it is. And since
a quality is that quality and no other one by virtue of

* Ibid., p. 641.
140

Some Basic Difficulties

its relations, any alteration in those differentiations ipso
facto is an alteration in the quality they differentiate. For
those relations constitute the context which determines
the identity of that quality.

We have also noticed that the nature or character of a
relation is what it is by virtue of the qualities that are
differentiated by that relation. The infinite process is recip-
rocal. Just as qualities are determined by their differ-
entiations to be what respectively they are, so those rela-
tions are determined to be the differentiations they are by
the qualities they differentiate. That is why any altera-
tion in those qualities ipso facto alters those relations.

Now this is held to be true of all qualities, and of all rela-
tions, not of only some of either, or of both. Nor is Bradley's
position here at all arbitrary. That all relations are internal
everywhere and always follows from the exclusion of external
relations. Were it suggested that some relations only are
internal, whereas some are external, the suggestion would
fly in the face of the conclusion of chapter II of Appearance
and Reality.

It follows from the universality of internal relations that
no limits to the differences that differentiate a quality may
be prescribed. Any quality is differentiated from all else,
not merely from some other realities. Identity implies dif-
ference because where there were no difference there would
be no distinction and therefore nothing distinct from any-
thing else.* This difference cannot be the bare numerical

* "I am not urging that quality without difference is in every sense
impossible. For all I know, creatures may exist whose life consists, for
themselves, in one unbroken simple feeling. . . . And if you want to
call this feeling a quality, by all means gratify your desire. But then
remember that the whole point is quite irrelevant. For no one is con-
tending whether the universe is or is not a quality in this sense; but the
question is entirely as to qualities. And a universe confined to one feeling
would not only not be qualities, but it would fail even to be one quality,
as different from others and as distinct from relation." 24

"I rest my argument upon this, that if there are no differences, there
are no qualities, since all must fall into one. But, if there is any difference,

141

Bradley's Dialectic

or existential difference of two qualities that would be in-
discernibly the same. Identity implies qualitative difference.
"All identity then is qualitative in the sense that it all must
consist in content and character. There is no sameness of
mere existence, for mere existence is a vicious abstraction." 527
Again, "numerical distinction is not distinction without
difference, that once more is senseless . . . " 531 "Without
difference in character there can be no distinction, and the
opposite would seem to be nonsense." 532 Two indiscernible
qualities would be not two but one. A quality that were not*
different from all other qualities would be indiscernibly
the same as some other quality, or qualities. The only
alternative to this (if there be qualities at all) is that every
quality should be differentiated from all else.

The same considerations, mutatis mutandis, apply to
relations. It follows that every quality and every relation
are unique. No two qualities, no two relations, can be the
same or indiscernible. Were it true that all difference is
difference in quality, it would be the case that every quality
and every relation is unmatched. The contradictory of this
consequence, viz., that some qualities and relations are
not unique but numerically different merely, is incompatible
with the conclusion that identity implies qualitative dif-
ference. For were there two simple qualities A x and A 2
that differed solo numero, there then would be one quality,
A 23 that did not imply its qualitative difference from all
else.

The suggestion that two qualities may be merely numer-
ically different (or qualitatively the same) in one respect
(e.g., hue) and qualitatively different in another respect

then that implies a relation. Without a relation it has no meaning;
it is a mere word, and not a thought, and no one would take it for
a thought if he did not, in spite of his protests, import relation into it.
And this is the point on which all seems to turn. Is it possible to think
of qualities without thinking of distinct characters?" Bradley answers
(p. 25) that this is not possible.

142

Some Basic Difficulties

(e.g., saturation), plainly fails to meet the requirement that
all difference be difference in quality. On that requirement
every difference would be unique; no quality or relation
could be matched. In his Logic, Bradley says that the first
of the principles of reasoning is that what is true in one
context is true in another context.* His relational view of
identity on which identity implies qualitative difference is,
then, absurd; for on it what is true in one context may not
be so in any other. The attempt to maintain that a quality
or a relation in this context may be partially the same in that
context, can hardly succeed. The "part" or "respect" or
"aspect" that is said to be the same in this context and in
that one, is enumerably different in the two contexts; and
yet, by hypothesis, it is qualitatively identical in them both.
Hence the "part" or "respect" or "aspect" that is in question
in the second context does imply its qualitative difference
from that given in the first context. Thus the notion of
partial sameness affords no escape from the conclusion that
identity implies qualitative difference, and the converse of
it that qualitative difference implies uniqueness.

To those who find in their experience qualities and rela-
tions that are strictly speaking the same or exactly matched,
this consequence of the conclusion that identity implies
qualitative difference will be a difficulty in Bradley's
dialectic. That difficulty, or the exclusion by Bradley's
doctrine of qualities and relations that are the same in the
sense that they differ solo numero, raises a question about the
validity of his alleged demonstration that all relations are
internal.

At the outset of the preceding chapter, we noticed that
the main arguments of that demonstration stem from a pair
of proposed disjunctions. The first of these is that presented

* "What is true in one context is true in another, and what holds of
a subject within an experiment is valid also beyond that experiment.**
p. 470.

143

Bradley's Dialectic

in chapter II of Appearance and Reality. On the one hand,
it is there argued that relations cannot be both separate
and capable of relating terms. And this alternative is rejected
because any view of relations as separate entities entails an
infinite regress in entities that can only fail to relate terms.
On the other hand, it is argued, on what is assumed to be
the sole remaining alternative, that relations are internal
to their terms. Bradley's explanation of what this means
is given in the course of the statement of the second one of
the two disjunctions that are in question.

After having shown there can be no relations that are
without qualities, Bradley proceeds (in chapter III) to urge
that distinct qualities without relations are impossible.
In this connection, the alleged disjunction is between the
alternative of an infinite process of differences "within"
every single quality on the one hand, and relations that
are internal to qualities, on the other hand; where being
"internal to" means what is meant by being the difference
that differentiates the qualities thus related. Differentiations
must "fall somewhere". They may not fall wholly "within"
the qualities they relate, for that would be to differentiate
qualities within themselves in an indefinite "process of
fission". The sole remaining alternative, it is alleged, is that
on which relations partially would "fall between" the
qualities they relate and thus be a difference between them.
In this sense of the term "relation", relations are qualitative
differences, and qualitative differences are relations.

Even though the exclusion in chapter II of the alternative
of separate relations be regarded as final, still the alternative
on which relations would be internal to their terms, in the
sense that any alteration in a relation must make a qualitative
difference in its terms, is not the sole alternative that remains.
For there is the further alternative on which an altera-
tion in a relation might entail a merely numerical difference
in its terms. On this alternative, a relation might be altered

144

Some Basic Difficulties

in character while its new terms would remain no more than
numerically different from the old ones. The disjunction
of chapter II between relations as separate, termless entities,
on the one hand, and relations as making a qualitative
difference in their terms, on the other, is thus vitiated by the
fact of an alternative which it does not exclude; namely,
that on which relations make a difference in their terms that
is not qualitative, but numerical merely. ^

The disjunction of chapter II likewise fails to exclude the
alternative of numerical difference. The truism that distinct
qualities must be different does not imply that distinct
qualities can only be different in nature. Thus, although
relations without terms, and qualities in no sense different
are indeed delusions, the exclusion of them by the arguments
of chapters II and III does not suffice to establish as the
final alternative that on which terms and their relations
mutually differentiate themselves in point of quality. For
there remains the further alternative on which a relation
requires terms that may be no more than numerically
different from other qualities, while at the same time terms
are related by relations that may be no more than numeric-
ally different from other relations. This would seem to satisfy
the logic of Bradley's requirement that relations relate terms,
and that distinct qualities should be different.

Some Idealists will reply to this that any notion of mere
numerical difference is nonsensical on Bradley's theory of
identity as relational. And that is even obvious. But it ought
to be clear enough to students of Bradley that his doctrine
of relational identity is no absolute fulguration of spirit.
Relational identity is the moment of immediacy in the process
of mediation that is the internality of relations. The doctrine
that identity is relational is one with the internality of
relations: it designates and lays emphasis on the essentially
mediated character of the identity of any appearance.

If an "external relation" be considered not as a "separate"

145 K

Bradley's Dialectic

entity, but as a universal that requires at least some pair
of terms or other for its illustration, then Bradley begs the
question of external relations by his doctrinaire assumption
that all differences are differences in quality. Now to say
this is not even to question, much less to deny, the validity
of the arguments of chapter II against the reality of relations
as separate, termless entities. Any such notion about relations
would be a delusion indeed. The point is simply that a
denial of reality to such entities and their exclusion from
theory constrains us to accept Bradley's theory of relations
if and only if it is the sole remaining alternative. And that is
not the case.

Let us acknowledge that relations require terms. External
relations as separate, termless entities are the merest of
chimeras. How much follows from this? At least that any
relation requires some terms or other. A relation with no
terms of any sort would be verbiage. But this much falls far
short of what Bradley is concerned to demonstrate by the
method of exclusion; namely, that relations and qualities
mutually differentiate and thus determine the qualitative
character of each other with the consequence that identity
implies qualitative difference.

Bradley's arguments do indeed exclude the alternative
of separate relations, either as found in experience or as
produced there by abstraction. But that much leaves
something more to the matter than the alternative that is
elucidated by Bradley. Before the view that qualities and
relations are such that the identity of any quality (or any
relation) implies its qualitative difference from all else,
our position is not ineluctable For all that the exclusion
of separate relations proves to the contrary, a quality and, a
relation may be no more than numerically different from
other qualities and relations.

We are assured by Bradley that * 'numerical difference"
is nonsense. That it surely is, within the system of Bradley's

146

Some Basic Difficulties

dialectic. Two qualities differing solo numero would violate
the requirement that identity imply qualitative difference.
And "numerical difference" would be misguided nonsense
in any case if the alternative view of relations that Bradley
elucidates showed the alternative of qualities and relations
that are enumerably, not qualitatively, different to be self-
stultifying. That much is shown in the matter of relations
as separate entities. But Bradley's dialectic of relations
excludes the alternative of numerical difference, not by
'stultifying it, but by the assumption that all difference is
difference in quality.

So far, the requirement that identity imply qualitative
difference has been referred to as a consequence of Bradley's
theory of quality and relation. That requirement can be
brought out as a consequence of the dialectic of relations
because the dialectic must assume it to be the dialectic it is.
The reason why there could not be two processes of cell-
fission that were in every qualitative respect the same is that
Bradley takes it for granted that to differentiate is to make
a difference that is a qualitative difference. This presup-
position excludes the very possibility of two processes of
development that would differ solo numero. For the quali-
ties and relations constituting those processes would be
enumerably different merely; they would not differ in
character.

Consider that in two processes which were the same in
character, the qualities could be differentiated by differences
that fell at once within and between those qualities. Thus
we would "have relation at once". 24 And in that unity of
immanence within quality and transcendence of quality
which is the very being of relation we would have the self-
discrepancy of all relations. On the other hand, qualities
would be both within and transcendent of their relations.
And this would yield the self-discrepancy in the aspect of
immediacy that is the fate also of quality. "We have found

147

Bradley's Dialectic

that qualities, taken without relations, have no intelligible
meaning. Unfortunately, taken together with them, they are
equally unintelligible. They cannot, in the first place,
be wholly resolved into the relations." 25 "Hence the qualities
must be, and must also be related. But there is hence a
diversity which falls inside each quality. Each has a double
character, as both supporting and as being made by the
relation. It may be taken as at once condition and result,
and the question is as to how it can combine this variety.
For it must combine the diversity, and yet it fails to do so." 26 '
Bradley's elucidation of the principle of this failure has been
quoted above. Because self-transcendence is self-discrepancy,
and because quality transcends itself in relation, quality is
self-discrepant. Quality must be "both made, and not made,
what it is by relation". 26 It must be made by relation in that
without relations a quality would be undifFerentiated. It
must be not made by relation in that the relations must fall
beyond the quality and be something in their own right.
That is why quality is self-transcendent. And whatever is
self-transcendent ipso facto is self-discrepant. ,/

Bradley brings out the same consequences on the side of
relation. The burden of chapter III is the mutual self-
transcendence of quality and relation. Quality without
relation would be nothing, for it would be undifFerentiated.
Were the relation contained wholly within the quality, the
relation would be merely internal to the quality. In. that
case, the relation would be wholly quality, and so it would
fail to be relation. Therefore, relations must fall between
the qualities they relate. In so far as relations are the self-
transcendent aspect of quality they are relations and
not sdf-discrepant. But in so far as they are in and of the
qualities they differentiate, relations transcend themselves
and thus they are self-discrepant. Now in the course of the
entire elucidation of the mutual self-discrepancy of quality
and relation, no demonstration that all difference is differ-

148

Some Basic Difficulties

ence in quality is advanced. The thesis of the argument is
the two-edged dialectic of quality and relation.

It would be futile to attempt to demonstrate that all
difference is difference in character or quality. Consider that,
for familiar reasons, you could not demonstrate it by
induction. That leaves deduction, and intuition. Your
premises have to contain anything you demonstrate by
deduction, so that method would beg this question. And
your intuition or mine demonstrates nothing to the other
fellow.

If Bradley assumes that all difference is difference in
quality, this is not to say that the assumption is made
arbitrarily. It was forced upon Bradley by his view of
appearance as a process of becoming or development,
wherein qualities or moments of immediacy are being related
by their differences. That differences may be no more than
enumerably different is not an alternative for Bradley
because, on that alternative, becoming is perforce left out
of account. Qualities and relations no more than numerically
different would be the same in character. Hence the self-
same quality and the self-same relation might be repeated
in different contexts without thereby being altered. This is
to say, that a set of qualities and a set of relations which
differed respectively solo numero would exhibit (say) ten
cases of a single quality, and (say) ten cases of a single
relation. These qualities and these relations would be the
same, regardless of their contexts. Therefore the qualities
would be respectively self-identical absolutely, as would be
also the relations; their identity would be absolute, not
relative. Hence those qualities and relations could only be
changeless; they could be in change or in succession, but they
themselves could not be changing. Therefore they could not
be moments of any process of becoming.

It is because of his assumption that experience is becoming
that Bradley implicitly excludes the alternative of numerical

Bradley's Dialectic

difference from the relational way of thought, and that he
explicitly denounces it as the nonsense it is on the pre-
supposition 'of his dialectic of experience as becoming. It is
also because of that same assumption that Bradley is
constrained (e.g., in the Logic and in Note A of Appearance
and Reality) to deny the validity of the Laws of Thought and
to identify the contradictory with the contrary. For so long
as the validity of the Law of Non-Contradiction is not
denied, the validity of the Law of Identity of course is not
denied, and the identity of A is absolute, not relational. If
the identity of A be absolute, then A may not be a moment
in any process of becoming.

Since, for Bradley, Appearance is in becoming, and no
process of becoming may be elucidated^onsistently_b)^ a
logic of contradictories, the contradictory must be identified
with the contrary, if Appearance is not to remain, in his
view, a succession of atomic mysteries. This way of repudiat-
ing logic yields a middle term between any two opposites.
Once this middle term is construed as being the moment of
mediation that differentiates the qualitative characters of
its terms, the necessary condition of the internality of
relations is laid down. The middle term has become the
moment of mediation or differentiation that Bradley calls
relation; and the terms thus differentiated are the moments
of immediacy or quality in the process that is Appearance.
If only because this repudiation of logic, made by identifying
the contradictory with the contrary, excludes the Law of
Excluded Middle, ipso facto it denies the very possibility of a
difference that were merely numerical. Some questions as
to how this procedure in philosophy may be justified will be
considered in the last chapter of this essay.

The relational way of thought, we have seen, carries us
on to the notion of the absolute Whole. This is something
that transcends any moment of mediation in an Identity that
is absolute. Since this Identity is absolute, not relational,

150

Some Basic Difficulties

we cannot attain it by the relational way of thought that is
ours without choice. No aspect of this Whole, no coherence
of aspects, however comprehensive, may be the ultimate and
absolute subject of any judgement. "We never have, or are, a
state which is the perfect unity of all aspects; and we must
admit that in their special natures they remain inexplicable.
An explanation would be the reduction of their plurality to
unity, in such a way that the relation between the unity and
the variety was understood. And everywhere an explanation
bf this kind in the end is beyond us." 415 It is beyond us
because we never arrive at the end of the relational way of
thought. To do so would be to commit intellectual suicide.
For we are finite centres,, "The internal being of everything
finite~3epends on that which is beyond it. Hence everywhere,
insisting on a so-called fact, we have found ourselves led by
its inner character into something outside itself. And this
self-contradiction, this unrest and ideality of all things
existing is a clear proof that, though such things are, their
being is but appearance." 404 The being of the finite depends
on its context; it is essentially self-transcendent, or self-
discrepant, and that is the mark of appearance. Yet the self-
discrepant posits its contrary; namely, self-coherence. Thus
the degree to which an appearance is self-coherent the
extent to which its self-discrepancies are resolved is the
degree to which it is real. The internality or coherence of
quality and relation carries judgement ineluctably to the
notion of a Unity that is absolute, not relational.

This Unity can only be single and unmatched. Two
Absolutes would stand to each other in external, empty
relation; and any such "relation" is verbiage. "Reality is
one. It must be single, because plurality, taken as real,
contradicts itself. Plurality implies relations, and, through its
relations, it unwillingly asserts always a superior unity. To
suppose the universe plural is therefore to contradict oneself
and, after all, to suppose that it is one. Add one world to

Bradley's Dialectic

another, and forthwith both worlds have become relative,
each the finite appearance of a higher and single Reality.
And plurality as appearance (we have seen) must fall
within, must belong to, and must qualify the unity." 460
Any attempt to assert the reality of a plurality of ultimates
would be in effect an attempt to resuscitate the delusion of
separate relations.

That the Absolute is one and single, that it is sentient,
that it is a trans-relational harmony wherein all self-
discrepancy is resolved, these are conclusions to which w6
are carried by Bradley's dialectic. Matters of final detail are
questions about which Bradley expresses ultimate doubts.
But he denies that the finality of the notion of the Absolute
is open to doubt. "With regard to the main character of that
Absolute our position is briefly this. We hold that our
conclusion is certain, and that to doubt it logically is
impossible. There is no other view, there is no other idea
beyond the view here put forward. It is impossible rationally
even to entertain the question of another possibility." 459
These are very strong claims indeed. But in the next sentences
Bradley goes on to give reasons why they must be made.
As we should expect, he goes right on to say that "Outside
our main result there is nothing except the wholly un-
meaning, or else something which on scrutiny is seen really
not to fall outside. Thus the supposed Other will, in short,
turn out to be actually the same; or it will contain elements
included within our view of the Absolute, but elements
dislocated and so distorted into erroneous appearance.
And the dislocation itself will find a place within the limits
of our system." 459 ' 460

"Our result, in brief, cannot be doubted, since it contains
all possibilities. Show us an idea, we can proclaim, which is
hostile to our scheme, and we will show you an element
which really is contained within it. And we will demonstrate
your idea to be a self-contradictory piece of our system, an

152

Some Basic Difficulties

internal fragment which only through sheer blindness can
fancy itself outside. We will prove that its independence and
isolation are nothing in the world but a failure to perceive
more than one aspect of its own nature." 460 The proposed
alien would be foreign to and therefore different from the
Whole. On that account alone the alleged alien could not
be a separate entity, out of all relation to all else. For by
virtue of its differences it would be in relation with all that
from which it is different. So long as Bradley's dialectic of
(Duality and relation be regarded as irrefragable, his monism
is beyond successful denial.

It may be well to add that Bradley expresses himself ex-
plicitly in the matter of the strength and scope of his claims.
"And the shocked appeal to our modesty and our weakness
will not trouble us. It is on this very weakness that, in a
sense, we have taken our stand. We are impotent to divide
the universe into the universe and something outside. We
are incapable of finding another field in which to place our
inability and give play to our modesty. This other area for
us is mere pretentious nonsense; and on the ground of our
weakness we do not feel strong enough to assume that non-
sense is fact. We, in other words, protest against the sense-
less attempt to transcend experience. We urge that a mere
doubt entertained may involve that attempt, and that in
the case of our main conclusion it certainly does so. Hence
in its outline that conclusion for us is certain; and let us
endeavour to see how far the certainty goes." 460 Bradley's
claim for the finality of his main conclusion is forced upon
him by the nature of his case. The internality of relations
prevents us from dividing the universe into itself and some-
thing beyond it. The alleged Other could only be different
from its foil, and that difference would be the relation of
the asserted Other with and so within the Whole. That is
why anything whatever we may mention will "be included
in our view of the Absolute". For that it may be mentionable

153

Bradley's Dialectic

at all it must be distinguishable, and therefore somehow
distinct, or related. The assertion of an ultimate, intrin-
sically individuated plurality of reals contradicts the assump-
tion that relation means what is meant by difference, where
difference is difference in quality.

Bradley's main conclusion about the Absolute is certain
in that before it Bradley has no redress. But if that conclusion
"cannot be doubted, since it contains all possibilities", at the
same time it does not exhibit many actualities. Each chapter
of Appearance and Reality elucidates a phase of appearance ill
as much detail, presumably, as Bradley discerns and can
muster in discourse. That this falls far short of a realization
in any judgement of the self-fulfilling and self-fulfilled is
insisted upon as inevitable. Our relational way of thought
cannot even pretend to omniscience and not thereby deny
its own nature. The claim to certainty about the main charac-
ter of the Absolute affirms no more than a notion of
absolute identity in differences; and this is affirmed under a
constraint that is without choice within Bradley's doctrine
of relations.

This constraint does not carry us very far into the self-
coherent details of the matter. For Bradley's " endeavour
to see how far the certainty goes" is unrelenting but not
self-stultifying. He takes up aspects of experience in some
detail, and presses them hard enough to make them disclose
the self-discrepancy of their content that is at once the
development of all thought and the ultimate defeat of any
judgement, no matter how self-coherent. The immediate
referent of any judgement is a finite centre of qualities and
relations. This focus of the judgement is real to the degree
to which it is coherent with the Absolute. We have repeatedly
noticed that every distinct content is related within the
Whole by its very differences from everything else within
the Absolute. For this reason, among others, the subject
term of every judgement is an adjective of the Real no less

154

Some Basic Difficulties

than is the predicate. The subject term and the predicate
term, in being different content, are thus related to each
other; and, in being different from all other qualities and
relations, S and P ipso facto are related to every other content
of the organic Whole. For that reason the Absolute is the
ultimate,* but unattainable, referent of every judgement.
The Absolute is the ultimate referent because it is the prin-
ciple of identity in the differences, or adjectives of the Whole;
and they are the content of judgements. Yet it is unattain-
eible by the relational way of thought. For to attain to the
Absolute would be to transcend the relational, and that
would be the suicide of the finite mind.

Nevertheless, it is urged that we are not lost in a relativism
that is without an attainable principle of survey and com-
parison. For the criterion of comparative degrees of reality in
Appearance, and of comparative degrees of truth in judge-
ment, is that of being comprehensive. "Hence to be more
or less true, and to be more or less real, is to be separated
by an interval, smaller or greater, from all-inclusiveness or
self-consistency. Of two given appearances the one more
wide, or more harmonious, is more real. It approaches
nearer to a single, all-containing individuality. To remedy its
imperfections, in other words, we should have to make a
smaller alteration. The truth and the fact, which, to be
converted into the Absolute, would require less re-arrange-
ment and addition, is more real and truer. And this is what
we mean by degrees of reality and truth. To possess more the
character of reality, and to contain within oneself a greater
amount of the real, are two expressions for the same thing.'*
322, 323 The earmark of degrees of individuality and so of
reality and truth is comparative self-coherence. Since the
identity of any finite experience implies the qualitative differ-
ence of it from all else, the identity or individuality of that
single experience is determined by its relations within the
* See above, pp. 126-129.
155

Bradley's Dialectic

systematic Whole of which it is an appearance. As the focus
of a judgement is enlarged, the scope of the coherence of its
constituent qualities and relations within the Whole is thus
rendered the more comprehensive. For with that enlarge-
ment in focus, the judgement has become a less partial aspect.
And with that increase in comprehension, the judgement
has become less erroneous, or truer. "Truth must exhibit
the mark of internal harmony, or, again, the mark of
expansion and all-inclusiveness. And these two characteris-
tics are diverse aspects of a single principle. That which
contradicts itself, in the first place, jars, because the whole,
immanent within it, drives its parts into collision. And the
way to find harmony, as we have seen, is to re-distribute these
discrepancies in a wider arrangement. But, in the second
place, harmony is incompatible with restriction and finitude.
For that which is not all-inclusive must by virtue of its
essence internally disagree; and, if we reflect, the reason of
this becomes plain. That which exists in a whole has external
relations. Whatever it fails to include within its own nature,
must be related to it by the whole, and related externally.
Now these extrinsic relations, on the one hand, fall outside
of itself, but, upon the other hand, cannot do so. For a
relation must at both ends affect, and pass into, the being
of its terms. And hence the inner essence of what is finite
itself both is, and is not, the relations which limit it. Its
nature is hence incurably relative, passing, that is, beyond
itself, and importing, again, into its own core a mass of
foreign connections. Thus to be defined from without is,
in principle, to be distracted within." 321 ' 322 For to be
defined externally or extrinsically is to be differentiated by
the relations that thus define their qualities from without.
Since these relations fall at once without and within their
qualities, those qualities are distracted within by the
internality in process of their relations, or aspects of self-
transcendence. This element of inner distraction is the

156

Some Basic Difficulties

principle of growth in degrees of individuality. "By growth
the element becomes, more and more, a consistent individual,
containing in itself its own nature; and it forms, more and
more, a whole inclusive of discrepancies and reducing them
to system. The two aspects, of extension and harmony, are
thus in principle one. . . ," 322 That is why to be more and
more true, or more and more real, is to be deprived less and
less of all-inclusiveness or self-coherence.

The ultimate resolution of discrepancies is fulfilled in an
Fdentity that is not relational, but Absolute. We have
noticed that the Absolute may stand in no relations. For
were the Absolute a term in any relation whatever, ipso
facto the Absolute would be relational, not absolute. It is
an emphasized, not incidental, conclusion of Bradley's
dialectic that the Absolute is not available to finite minds.
There are, then, at least two reasons why we may not
avail ourselves of the Absolute as a criterion of truth and
reality. The Absolute may not stand in a relation, cognitive
or otherwise, without thereby falling from ultimacy. And
the finite mind could attain to the Absolute only by trans-
cending the relational, thereby committing suicide.

These consequences of Bradley's dialectic carry with
them further basic difficulties. For they leave us with no
criterion of degrees of truth and reality. To be sure, Bradley
writes of inclusiveness and harmony, or coherence, as being
the standard in question. Yet this does (and could) not
mean that this standard is anything distinguishable from the
very degrees of truth and reality themselves of which that
standard is the criterion. The degree to which any finite
whole is coherent is in no sense distinct from that appearance
itself. For the coherence of that finite whole is that whole
itself that very appearance, which is self-coherent to the
degree that it is individual, and is the individual it is in
virtue of that self-coherence.

It might well be urged that the Absolute is the criterion

157

Bradley's Dialectic

of degrees of coherence if that Being were, or could be,
available to finite minds. Yet the Absolute is something
distinct from any appearance, no matter how highly self-
coherent it be. Therefore, the Absolute could not have its
being at any level of degrees of truth and reality. And the
relational way of thought could not attain to the Absolute
and remain relational, just as the Absolute could stand in
no relation whatever and remain absolute. Willy nilly, with-
out choice, we are confined to the degrees of reality that
are appearance. This means that we have no criterion
of degrees of truth and reality that would be distinct from
those very appearances themselves. And that is tantamount
to having no criterion at all.

This is not intended to lead up to practical difficulties
that might be met with in the course of any attempt to apply
a criterion of comparison in point of coherence. Often
enough that may be very difficult to do. The second draft
of Hyperion, as read by Amy Lowell, is more coherent in her
interpretation of it than the first draft. Yet that comparison
was realized and made out in discourse by one who was a
poet in felt imagery and ambition. The difficulty in question,
however, lies far deeper down than any difficulty in the
application of a criterion in practice. Since there is no avail-
able standard of coherence, how are we to compare this
appearance as being more coherent than that one ? More-
over, in the absence of any available criterion, we cannot
determine just how coherent any single appearance may be.
If we cannot decide that much in some case or other, how
can we compare this Appearance as being more coherent
than that one ?

Surely, though, Wuthering Heights is more coherent than
An American Tragedy, we may be told. And on assignable
grounds that comparison might be easily made out.* But

* For example, on the grounds laid down by E. M. Forster in Aspects
of the Novel.

158

Some Basic Difficulties

in that case we have posited our criteria of comparison.
On Bradley's dialectic it is a repeatedly emphasized con-
clusion that knowledge of the Absolute is not available to us.
Now we may agree that we do see how, on the coherence
theory of truth and reality, it must be that an appearance
becomes the more true and real as it becomes the more
coherent. But still we are obliged to ask, how true, how
real is it at any stage of this expansion ?

Short of the Absolute, we have only appearances. The
degree to which an appearance is coherent is in no respect
or sense distinct from that appearance or concrete universal
itself. Therefore, either the coherence of an appearance is
its own index, or we have no index of degrees of coherence.
The suggestion that comprehensiveness is the test, and that
it can be applied by setting up a certain very comprehensive
appearance as a norm under which the comparative degrees
of other appearances might be decided upon, is really of no
avail. For it begs the question. Either the superior coherence
of the norm is its own index, or, again, we have no index.
And if, in the case of the norm, we assume that coherence
is its own index, then by the same token we should have to
make the same assumption about the coherence of any other
appearance; and so the suggested norm would be useless.
On the other hand, if we have no index of degrees of coher-
ence, then the selection of the norm could only be arbitrary.

It would seem that, in Bradley's view, coherence is its
own index. The coherence of a relational situation involves
satisfaction for the intellect. This satisfaction diminishes or
grows as there is a decrease or increase in coherence. And
Bradley holds that any growth in intellectual satisfaction
is the index of fuller coherence in judgement. Yet the
validity of this converse is at least doubtful. Even though it
be established by Bradley's arguments in that regard that
the higher the degree of coherence the fuller the intellectual
satisfaction, it would hardly follow (or appear to follow,

159

Bradley's Dialectic

except by illicit conversion) that the fuller the satisfaction
the higher the coherence. Degrees of coherence may be an
index of degrees of satisfaction. But if the coherence theory
about the nature of truth is to afford a criterion of degrees
of truth, what we need is an index of degrees of coherence.
And that we do not and cannot have, short of the unavailable
Absolute.

1 60

CHAPTER IX

Relational and Absolute
Identity

i

WE have seen why it is, for Bradley, that identity implies
qualitative difference. Since qualitative differences, or
differentiations, are relations, to say that identity implies
difference is to say that identity implies relation. Indeed,
the dialectic of quality and relation elucidates the mutually
relative character or identity of the aspect of quality and
the aspect of differentiation which together constitute "the
relational situation."

In being thus relational, the identity or character of any
experience or appearance is in process. If an experience
were self-identical absolutely and in its own ontological
right, it would be difficult indeed to see how it could be in
process. If A were A absolutely, rather than relatively,
it could not change into, or become, Y. For in the course
of the process A, Ay, Ayy, . . . Y, there would finally be a
point at which A was no longer A and was not yet Y. On a
view of identity as absolute, there could be no middle term
by which this gap in becoming might be mediated. Thus
one apparent virtue of a dialectic of contraries is that between
contraries there is a middle term in and through which the
contrarieties in becoming are incessantly sublated.

A failure to see that for Bradley appearance is process
wherein the identity or character of an experience is relational
and, short of the Absolute itself, never absolute, would

161 L

Bradley's Dialectic

blind us to the nature of his dialectic. Should we attempt
to illustrate Bradley's conception of relation as differentiation
by thinking of a relation as a universal that requires at
least two particulars, our attempt could only be irrelevant.
Defined as an abstract universal, a relation would be self-
identical and therefore changeless. Evidently such beings
could not be moments of differentiation in process. Then
again, were a sense-quality, conceived of as self-identical,
taken to illustrate Bradley's view of quality as the moment
of immediacy in the process that is appearance, this too
would be a mistake. For no such being as a self-identical and
therefore changeless sense-quality could be in process at all.
When largely irrelevant matters are taken to illustrate
Bradley's theory of relations, rather unfriendly criticism
sometimes results. Let us consider the following passage
from Professor C. D. Broad's Examination of McTaggarfs
Philosophy. In reply to his own question, "Is there any valid
objection to there being relations?"* Professor Broad
considers two lines of argument. The first of these is drawn
from Leibnitz, the second is ascribed to Bradley. It runs as
follows: "The second argument against relations is that of
Bradley. The argument is that, if A is to be related by R to B y
A must be related by a relation R l to /?, and R must be
related by a relation R l to B. On the same grounds A must
be related by a relation R n to R l9 R l must be related by a
relation R 12 to R, R must be related by a relation R 2 i to R 2>
and R 2 must be related by a relation R 22 to B. Similar
remarks will apply to all these four relational facts, and so
at the next stage there will be eight relational facts, at the
next to this sixteen, and so on without end. Bradley's
contention is that this series could not have a first term unless
it had a last term, which it plainly does not. McTaggart
admits that there is this endless series in connection with
any relational fact, but he denies that it is vicious. His

* P. 84.
162

Relational and Absolute Identity

answer amounts to saying that the first term, i.e. that A
has R to 5, is a fact in its own right, and that the rest of the
series consists merely of further consequences of this fact.
I think it might fairly be said that, whilst Leibniz's
argument depends on insisting that relations shall behave
as if they were qualities, Bradley's argument depends on
insisting that they shall behave as if they were particulars
like the terms which they relate. It is plain that Bradley
thinks of A and B as being like two objects fastened together with a
Bit of string, and he thinks of R as being like the bit of string. He
then remembers that the objects must be glued or sealed
to both ends of the bit of string if the latter is to fasten them
together. And then, I suppose, another kind of glue is needed
to fasten the first drop of glue to the object A on the one side
and to the bit of string on the other; and another kind of
glue is needed to fasten the second drop of glue to the object
B on the one side and to the string on the other. And so on
without end. Charity bids us avert our eyes from the
pitiable spectacle of a great philosopher using an argument
which would disgrace a child or a savage. 55 *

Professor Broad makes no reference to the text that he
calls in disgrace. But he would seem to have in mind, more
or less, the argument in the chapter on Substantive and
Adjective by which Bradley is brought to his denial of the
reality of relations as independent entities. In several ways,
Professor Broad misconstrues that argument. // is not an
argument against the reality of relations ; it is an argument against
the reality of relations as independent entities. Yet Professor
Broad introduces it as one reply to his own question, "Is
there any valid objection to there being relations? 55 not,
is there any valid objection to there being relations taken as
separate entities? In this latter form, the question has a
restricted bearing: it is a question about the reality of
relations tout court. And that one sense of the term relation
* Pp. 84, 85. My italics.
163

Bradley's Dialectic

is the concern of Bradley's argument to the conclusion that
any notion of relations as independent entities leads to an
"infinite process" 1 * (my italics) in relations that do not
relate.

Yet Professor Broad writes of this argument as though it
were Bradley's aim to show by it that there are no relations;
that there is a "valid objection to there being relations".
That this is so no one who has followed and remembered
the dialectic of chapter III, Relation and Quality, is likely to
agree. Bradley's argument against the possible reality df
independent entities as relations is just that, and not itself
an argument against anything more than that.

Professor Broad asserts: "It is plain that Bradley thinks of
A and B as being like two objects fastened together With a bit
of string, and he thinks of R as being like the bit of string."
If one wanted to one could adduce "the bit of string" as an
illustration of relations as independent entities. In that
connection also one could write with Professor Broad of the
glue, and that Bradley "remembers" that glue would have
to be introduced into an intellectual conclusion; perhaps in
order to make it at home in certain quarters. That way of
taking Bradley's argument against relations as independent
entities might be less objectionable were it restricted to that
argument, not presented as reasoning by him against the
reality of relations.

But pieces of string and bits of glue would hardly afford
an illustration of moments of differentiation in process. So
long as Bradley's critics take it that his argument to the
"infinite process" that results from the notion of relations as
independent entities is his constructive theory of relations,
they will criticize that argument of the chapter Substantive
and Adjective, and neglect the content of the chapter Relation
and Quality. That is what has happened in the past. There
are more than a few criticisms of the argument against
separate relations; but discussion and criticism of the

164

Relational and Absolute Identity

positive doctrine of chapter III is comparatively scarce and
far to seek.

And so long as Bradley's unfriendly critics concentrate
on the chosen argument in chapter II, it is likely they will
take a relation for Bradley as being something that is real
independently of its context in process or Appearance; as
something static, or as a piece of string and bits of glue.
For in that argument Bradley is concerned with something
that would be independently real, static, and so almost
anything different from a moment of differentiation. The
notion of that something is what involves the "infinite
process" in relations that do not relate simply because they
are ex hypothesi independent, disconnected entities. And
it is to that notion that Bradley denies any validity whatever,
not to the doctrine that is elucidated in the next chapter,
Relation and Quality.

To be sure, the validity of that theory of relations is not
regarded as absolute. That theory itself is a judgement;
albeit one to which we are brought by the elucidation of
the matter which Bradley has worked out. And short of the
unattainable Absolute Idea, any judgement will be con-
tingent upon factors lying beyond its scope, even though
that mental synthesis be carried out to a rather preternatural
extreme. As judgement is contingent and true to a degree,
so any appearance is contingent and real to some degree
or other. Short of the Absolute nothing is absolutely real.
But everything is real to a degree.

Yet we have seen that Dr. A C. Ewing, writing in answer
to a question as to which, if any, of his senses of "internal
relations" were appropriate to Bradley's theory of the
matter, asserts that Bradley denies the reality of relations.
"I did not say of which of my senses of 'internal relations'
Bradley is a supporter because Bradley denied the reality of
relations and therefore can not have held that relations
really were internal in any of my senses or in any other

165

Bradley's Dialectic

sense."* Apparently Dr. Ewing is at one with Professor
Broad in taking it that Bradley denies the reality of relations.
The consequence of this, that "Bradley cannot have held
that relations really were internal in ... any sense",!
Dr. Ewing does not boggle at pointing out himself.

It is the case that Bradley denies the reality of relations
as independent entities. And it is the case that he denies the
absolute reality of relations as differences. He denies also the
absolute reality of qualities. As relations independent of
qualities are a delusion, so together with their qualities
they are not completely intelligible in any finite context.
For that a relation may be internal, it must contribute to
constitute its qualities: yet, that it may not disappear
altogether, it must "fall" to some extent "between" them. A
relation thus involves within itself a contrariety. A relation
must be at once implicated in and transcendent of its
qualities. For in so far as a relation contributes to constitute
its qualities it does not fall between them, and so far it
fails to be a relation at all; on the other hand, in so far as a
relation falls between its qualities it is outside them both and
so again fails to relate them. 26 " 28

Hence no moment of differentiation could be absolute.
In a process of fission no absolute distinction is there to be
found between the moments of quality and the moments of
differentiation. Rather there is "a diversity which falls
inside each quality. Each has a double character, as both
supporting and as being made by the relation". 26 Likewise,
each relation has a double character, as at once making and
being made by its terms. Qualities taken without relations
(i.e., as absolute), and relations without qualities (i.e.,
independent or absolute), are both, Bradley finds, quite
unintelligible by the relational way of thought. These
extremes of abstraction are wholly verbal. Their reality is

* The Journal of Philosophy) Vol. XXXII, No. 10, May 9, 1935, p. 273.
My italics. f Ibid.

166

Relational and Absolute Identity

entirely a matter of words. Otherwise they are unreal.
For their "designations" refer to nothing. But a quality
that is a moment of immediacy which is at once making
and made by its differentiations or relations is not an
absolute. The self-identity of the character of it is relational,
not absolute. And a relation or differentiation that is at
once in and of its qualities and transcendent of them is
no absolute, no isolated entity, but rather a copula in
process.

'Appearance, as distinguished from Absolute reality, is
relational; any appearance is as unreal as it is inconsistent,
and as real as it is self-coherent. We have seen this much in
Bradley's view, but in view of the opposition, it may be well
to cause him to repeat it. At the outset of chapter XIII,
The General Nature of 'Reality, he writes, "Whatever is rejected
as appearance is, for that reason, no mere nonentity. It
cannot bodily be shelved and merely got rid of, and, there-
fore, since it must fall somewhere, it must belong to reality.
To take it as existing somehow and somewhere in the unreal,
would surely be quite meaningless. For reality must own and
cannot be less than appearance, and that is the one positive
result which, so far, we have reached." 119 "Is there an abso-
lute criterion?" Bradley goes on to ask. "This question, to
my mind, is answered by a second question: How otherwise
should we be able to say anything at all about appearance?
For through the last Book, the reader will remember, we
were for the most part criticizing. We were judging pheno-
mena and were condemning them, and throughout we
proceeded as if the self-contradictory could not be real.
But this was surely to have and to apply an absolute criterion.
. . . Ultimate reality is such that it does not contradict
itself; here is an absolute criterion. And it is proved absolute
by the fact that, either in endeavouring to deny it, or even
in attempting to doubt it, we tacitly assume its validity." 120
"We may say that everything, which appears, is somehow

Bradley's Dialectic

real in such a way as to be self-consistent. The character of the
real is to possess everything phenomenal in a harmonious
form." 123

Relations are the differences that differentiate the pheno-
mena of appearance. A relational situation is unreal to the
degree that it is self-contradictory or self-discrepant; it is
real to the degree to which it is self-consistent or self-
coherent. Nothing short of Absolute reality is real without
qualification. But to argue that the unreality of relations
(and qualities) means for Bradley that they are nothing at
all would be to argue that for him Appearance is nothing
at all. And that would be silly, at best. Moreover, any such
misunderstanding would ignore the doctrines of The General
Nature of Reality (chapters XIII and XIV), Thought and
Reality (chapter XV), Error (chapter XVI), The This and
the Mine (chapter XIX), Degrees of Truth and Reality (chapter
XXIV) and The Absolute and its Appearances; in short, the
notion that Bradley denies reality to relations leaves out of
account the basic passages of his dialectic.

The phases of Appearance that are examined in Book I
of Appearance and Reality are shown to be infected with the
self-discrepancy of their constituent qualities and relations
which is elucidated in detail in chapter III. Thus space
and time are less than absolutely real. But it would be mis-
leading to say that Bradley denies the reality of time.
He denies, over and over again, the self-consistency, the
absolute reality of any and every form of appearance.
How could he do otherwise? Appearances are processes,
whether they be motions,* causes, f activities,^ things,
or selves; || every process is self-transcendent, and, we have
seen, the self- transcendent is not to be nothing. "Whatever is
rejected as appearance is, for that very reason, no mere

* Chapter V. t Chapter VI.

J Chapter VII. Chapter VIII.

|| Chapters IX and X.

1 68

Relational and Absolute Identity

nonentity. ... To take it as existing somehow and some-
where in the unreal, would surely be quite meaning-
less." 119 Any phase of appearance a spatio-temporal
system, for example is real to the degree to which it is
self-coherent. And the degree to which a system is self-
coherent is held to be the extent to which it is com-
prehensive.

Those who would make Bradley a synonym for their own
mistakes appear to take it that by an appearance he means
something static and self-contained. Thus they find it con-
venient to concentrate on one of his arguments against
relations as separate entities (which would be static and
self-contained, if there were any such entities in Bradley 5 s
viev). They carry out that concentration with such single-
minded emphasis that this argument alone is made to seem
Bradley's positive theory of relations. Standing on that
misunderstanding, they make the monstrous assertion that
he defied reality to relations.

To take it that for Bradley relations are static, self-con-
tained entities, like pieces of string, and to leave out of
account tl\e dialectic of relational fact, or quality and
relation, that is set forth in chapter III, would be perforce
to ignore the relational identity of every being short of the
Absolute. It would be to take it that relations are self-
identical. Ten of the possibly many senses which "relation",
used to designate a self-identical entity, might have are
made out br Dr. Ewing. He acknowledges that none of
them is gernane to Bradley's theory of the matter. Two
general sense* of the term in question, used to designate
self-contained entities, are considered in detail and rejected
roundly by Bradley himself. "I will now deal briefly with
relations, takm as what may be called 'external 5 or 'internal'
merely. And, jthough at the cost of some repetition, I will
show how sucii a distinction, if we insist on it as ultimately
valid, involve* us again in contradiction. It exhibits once

169

Bradley's Dialectic

more the discrepancy inseparable from all relational
thought."*

"What should we mean (I will ask first) by a relation
asserted as simply and barely external? We have here, I
presume, to abstract so as to take terms and relations, all
and each, as something which in and by itself is real inde-
pendently. And we must, if so, assume that their coming 01
being together in fact, and as somehow actually in one, fe
due in no way to the particular characters of either thfe
relations or the terms. From neither side will there te
anything like a contribution to, or an entrance into, the
other side or again to, or into, that union of both which
we experience as a relational fact. Undeniably the fact is
somehow there, but in itself it remains irrational as admitting
no question as to its 'how' or 'why'. Or, if you insist !>n a
reason, that would have to be neither in the terms nor the
relation, but in a third element once more independently
real and neither affecting, nor again affected by, either the
relation or the terms. This, I suppose, is the way in which
relations have to be misunderstood, if you take them as
external merely and also as ultimately and absolutely real.

"What (I ask next) should, on the other hand, be meant
by a relation viewed as absolutely and merely internal?
You, I presume, still in this case would continue to take the
terms each one as, so far, in and by itself real, ind as inde-
pendent absolutely of any whole that could be said to contain
them. And you would go on to attribute to tlje particular
characters of the terms, as so taken, some acfual relation
or relations which you find, as you say, to fall between them.
Something like this, I suppose, is or ought to be meant by
a relation which is asserted to be real ultimately; and internal
merely. j

"The idea, I would add, that I myself accept aty such doctrine
as the above seems to myself even ludicrous. And tf whom, if to
* Collected Essays, p. 641.
170

Relational and Absolute Identity

any one, it should be attributed in fact, I will not offer to
discuss. In any case, to assume it as the necessary alternative,
when the mere externality of relations is denied, is (I
submit) an obvious, if perhaps a natural, mistake."* Having
given this detailed statement of the matter, Bradley proceeds
to deny reality to merely external relations on the familiar
grounds that, as quite external to their terms, such alleged
relations fail to relate anything. Therefore mediating rela-
tions would be required and so on indefinitely. He then turns
to his statement of the nature of a merely internal relation.
This alleged relation, like a piece of string in need of glue,
simply falls between its terms. The assumption that any such
view is the sole alternative to his denial of external relations
he terms an obvious mistake, and the notion that he accepts
it Bradley finds ludicrous.

Relations for Bradley are at once internal and external.
As the dialectic of chapter III makes out in detail, a rela-
tion is internal to its qualities in so far as it contributes to
constitute them; it is external to its qualities in so far as it
transcends them; "every relation (unless our previous
inquiries have led to error) has a connection with its terms
which, not simply internal or external, must in principle
be both at once."t The connection must be both at once
because if it fell wholly without the terms (and so were
wholly external) it would not connect them; and if it fell
wholly within the terms (and so were wholly internal) it
would be exhausted in them, and thus fail to be a nexus.

We have seen that the self-transcendence of quality through
relation is the relational identity of the character of any
appearance. Relational identity is other than the absolute
identity of A is A. In judgement the one yields degrees of
truth and error, in propositions the other gives barren
tautologies. The disjunction here is that between an identity

* Ibid., pp. 642, 643. My italics.
f Ibid., p. 641.

171

Bradley's Dialectic

differentiated by its context, on the one hand, and an
identity that is intrinsic, or differentiated by nothing extrinsic
to itself.

We have seen that on Bradley's doctrine of relations
identity implies qualitative difference. Any relational situa-
tion is different in character from any other one. This
excludes the possibility of two beings that differ solo numero.
The two symbols A, and A, could not be strictly the same.
On the other side of the disjunction between relational and
absolute identity, numerical difference is affirmed, and the
merest possibility that all beings are different in character
is excluded. This disjunction is the principle that divides the
relational dialectic from any form of procedure in thought
for which " A is A " is valid by virtue of the Law of Non-
Contradiction.

For Bradley, we have noticed, "A is A" is a barren verbal-
ism, as is "A is not both A and not-A". In Appearance
there are no absolute identities and there are no absolute
contradictories. Any identity is the result of differentiation,
and the differentiation mediates between that identity and
its opposite. Thus, in the process that Appearance is, there is
a middle term between any two moments of becoming.
Any phase of process will be intelligible to the extent that
this mediation of moments is realized in judgement. For
that mediation or differentiation of quality is the relational
identity of any appearance, any judgement, any degree
of reality however abstract, or however concrete.

II

In conclusion, let us ask whether or not Bradley can
answer Hume. A discussion of this question will afford an
illustration of the disjunction between relational and abso-
lute identity. We shall see that Bradley can no more answer

172

Relational and Absolute Identity

Hume than Hume can answer Bradley. Reduced to its
simplest terms, the main reason why this is so is that not-p
means something for Bradley that is radically different
from what it means in Hume.

We have seen that Bradley denies the validity of the Law
of Non-Contradiction and identifies the contradictory with
the contrary. The issue of this union we have seen to be
a theory of negation and identity as relational. In this matter
of dialectical principle Bradley is following Hegel. "Hegel
has taught us this and I wish we could all learn it."

There is no evidence that it ever occurred to Hume to
deny the validity of the Law of Non-Contradiction as it is
explained, for example, in the Port Royal Logic. For Hume
identity and negation are absolute, not relational. Any
impression, any idea, is what it is by virtue of its own intrinsic
nature, not by virtue of any context of relations whatever.
Thus for Bradley identity is relational; for Hume identity
is absolute.

The difference here is that of a disjunction: no qualita-
tive identity may be both relational and absolute. This is to
say that the dialectic of Bradley is utterly different in prin-
ciple from the logic in Hume. Hence neither one could be
expected to answer (as distinguished from condemn) the
other. They do not speak the same language. Bradley and
Hume can be and are opposed to each other, but neither
one can properly be taken to be a refutation of the
other.

This is not to forget that those who derive their philosophy
from Hegel, either directly or by way of Bradley and
Bosanquet, take it that they can and do refute Hume and
his kind. Following the lead of T. H. Green, they first
demolish the theory of impressions and ideas, and then
assume that Hume's negative analysis of causal inference
depends, both in point of the text and of the logic of it, on
that indefensible psychological theory. On two counts the

173

Bradley's Dialectic

assumed dependence in question may be seen to be ground-
less.

One of these counts is historical in nature. Hume's
failure to find a necessary connection between anything
designated cause and anything designated effect did not
begin with him. He drew it from Malebranche, as his
reference in the Treatise to the relevant portions of Recherche
would suffice to make plain, if there were no other evidence
in the matter.* And Malebranche derived the conclusion
from Cordemoy, to whom he refers in that connection. 4
Malebranche and Cordemoy were Rationalists and their
views in psychology do not comprise a theory of impressions
and ideas as copies of impressions. The line of argument
for which Hume is perhaps best known developed in-
dependently of Hume's psychology, if only because Hume
did not discover or develop it in the first place.

The other count in question is of a textual nature. The
relevant text of the Treatise does not bear out the assumption
that Hume's negative analysis of causal inference depends
on his dogmatic psychology.! Rather it constrains us to
conclude that (to put the matter in too few words) Hume
applied the tests of apagogic reasoning and sense-perception
to two well-known questions about experience: why is a
cause always necessary? and why must the same cause
produce the same effect?

Nevertheless, it may be urged in some quarters, Bradley's
dialectic does constitute an answer to Hume. For it shows
his faith in absolute identity and apagogic reasoning to be
groundless. Since apagogic reasoning proceeds to demon-
stration by showing that the contradictory of a given state-
ment is itself self-contradictory, to show a faith in that to
be misguided would be to explain that the Law of Non-

* See my Malebranche and Hume. Revue Internationale de philosophic,
Vol. I, No. i.

f See my Hume's Theory of the Understanding) London, 1935, Chapters
II and III.

174

Relational and Absolute Identity

Contradiction is a snare and a delusion. Yet that is the basic
issue as between the Hegelian dialectic and logic. Is the
contradictory one with the contrary or are they radically
distinct? These alternatives constitute a disjunction: both
of them cannot be true.

Bradley denounces A is A and A is not both A and not-A
as being tautologies and therefore empty. They do not and
cannot represent any advance in thought; therefore they
are not judgements. Now any tautology in thought will be
as empty or as full as it is. That depends wholly on what a
mind puts into it. Whenever A is A is construed in no sense
whatever, but is merely parroted, it is quite empty indeed.
Yet ordinarily A is A is used to designate the logical structure
of whatsoever is self-identical. This may be an analogy;
and it is fair to presume that no one denies that an analogy
may give information to anyone who thinks it out.

To denounce any case of A is A as a tautology is to object
to it for being what it is. And the categorical principle A
is A is invalid if and only if it be true that identity is
relational. The alleged truth of this latter statement is in
no wise brought to light by the denial of its opposite.
No more is it established by the identification of the con-
tradictory and the contrary. For to identify them is by
that very fact to assume that negation and identity are
relational.

If and only if the contradictory be identified with the
contrary does it follow that every opposition, every fission,
in experience is mediated by a third moment, or is relational.
That conclusion is a necessary condition of the dialectic of
relation and quality. Without the mediation of any process
of differentiation by a third moment, the mutual internality
of relation and quality would disappear.

Presumably the protagonist of the Idealist dialectic will
reply that his position is not a matter of assumption. His
dialectic elucidates what is the truth. The Law of Non-
175

Bradley's Dialectic

Contradiction is false because it would affirm mere or pure
negation. Any such notion is intolerable because it entails
the infinite judgement. "Wisdom is not blue" is a case in
point. Now, plainly, that judgement is silly; but that it is so on
the grounds advanced by Bradley is doubtful. "A something
that is only not something else, is a relation that terminates
in an impalpable void, a reflection thrown upon empty space.
It is a mere nonentity which cannot be real." Some of the
confusions, at least, that are basic to the notion that the Law
of Non-Contradiction entails the infinite judgement have
been exposed by W. E. Johnson. Moreover, Bradley's own
statements about the matter would seem to beg the question;
as is done, for example, in the passage quoted above.
Therein "a something that is only not something else" is
identified with "a relation ..." Now anything that were
"only not something else" a mere not would be something
outlandish indeed. But it is not in the least to defend the
reality of a something that were merely not something else,
to go on to point out that, by assuming this negative some-
thing to be a relation, Bradley begs the question whether
negation is relational or not.

Just as any negation that does not entail a common ground,
a third term, between its contents is rejected by Bradley
as being empty, so he condemns any statement of absolute
self-identity as being inane. We have seen that Bradley
writes in this connection that "The principle of Identity is
often stated in the form of a tautology, 'A is A'. If this really
means that no difference exists on the two sides of the
judgement, we may dismiss it at once. It is no judgement at
all. As Hegel tells us, it sins against the very form of judge-
ment; for while professing to say something, it really says
nothing. It does not even assert identity. For identity without
difference is nothing at all." Now this too begs the question
at issue. If and only if this conception of identity be true are
we constrained to dismiss the Law of Identity. In the passage

Relational and Absolute Identity

quoted above, Bradley makes plain his belief that what
Hegel tells him on this score is indeed the truth. Yet
that is the question at issue. And that question is merely
begged by the assumption that all identity is identity in
difference.

Moreover, in order that he might show that all identity
is identity in difference, Bradley would have to show that
some cases of identity are not properly represented by A is A.
This could not be established by any appeal to experience,
for no one could so canvass experience as to be sure that
no case of identity (such as that of two cases of the same hue)
is properly represented by A is A. Consequently, Bradley is
obliged to arrive at the truth of his principle by assuming it.
And that is what in fact he does. His theory of negation, his
dialectic of quality and relation, his theory of predication,
and the coherence theory of truth are not proofs of the
principle of identity in difference; rather, they are elucida-
tions of it.

Something of the bearing of the disjunction between
absolute and relational identity on the opposition between
Hume and Bradley may be indicated in the following way:
It has been pointed out elsewhere that "The five major
assumptions of Hume's epistemology would seern to be (i)
that experience may be exhaustively analyzed into elements;
(2) that every simple idea is the copy of a simple impres-
sion; (3) that resemblance and difference (taken ' 'philo-
sophically") are neither qualifying predicates nor relations;
(4) that what is distinguishable is separable; and (5) the
attraction of association."*

In the same connection, it was pointed out that the
second and fifth of these assumptions state the main content
of the first, and also that the third and the fourth assump-
tions express Hume's own view of his so-called atomism in
philosophy. For Hume held that the elements of his

* Hume's Theory of the Understanding, London, 1935, p. 218.

177 M

Bradley's Dialectic

philosophy are not connected by their being the same, or
by their being different.

This is to say that the resemblance of any simple ideas
to each other is not "a point or circumstance" distinct from
the respective ideas themselves. Their resemblance is not
"a common circumstance"; for it is in no wise distinct from
those very ideas. To find that two simple ideas pi and ps
are resembling is to find that pi and ps are the same in
quality or character. In Hume's view, this is to find that
they are the same not in point of "a common circumstance' '
that would be distinct from those simple ideas, but rather in
virtue of their being qualitatively the same in and of them-
selves alone.

Thus we may see that for Hume "resemblance" means
what is meant by the qualitative identity of simple ideas.
A resemblance is any case of a qualitative identity (e.g., a
perceived middle C) that exists in at least two cases of itself;
and any case of a qualitative identity that exists in two or
more cases of itself is properly called a "resemblance".
This holds likewise of complex experiences. Experience
MNOPj and experience P 2 QRST are respmbling or the
same in respect of their constituent P, for P is the same in
P l and P 2 . Thus it may be noticed that Hume assumed that
"resemblance" designates a qualitative identity that is
distributed in at least two cases of itself. And this is to assume
that a single quality or character (e.g., perceived azurite)
may be repeated in two or more cases of itself.

Thus, on Hume's theory of resemblance, the elements of
experience are intrinsically self-identical, and not what they
respectively are by virtue of any context whatever. This is
to say that those elements (as distinguished from the
"perceptions of the mind" which they contribute to con-
stitute) are self-contained, each one intrinsically and in its
own right. That is why, on this view, any difference is the
mere "negation" of a resemblance. Consequently any com-

Relational and Absolute Identity

plex may be analyzed into its elements without remainder,
and without thereby altering those elements. "Whatever is
distinguishable is separable," on this view, because wherever
there is a distinction a separation may occur. For whatever
is distinct from anything else is a distinct being whose
identity is intrinsic, not relational. Since this self-identity is
intrinsic or absolute, it is not alterable. No more is it con-
tingent, in any sense of the term. This is why, on Hume's
view, any distinct and therefore distinguishable element of
experience is separable from any other without the effect
of any alteration whatever in either one. In virtue of the
intrinsic self-identity of it, no element can be altered. And
since resembling elements are no more than the qualitatively
identical elements themselves, no analysis of a complex can
either alter, or find anything above, the elements of the
complex altered.

We saw that Bradley denounces as quite fatuous any
notion of relations as "merely internal". In that connection
it was remarked in passing that Hume's analysis of the basic
philosophical relations of resemblance affords a good
example of a "merely internal" relation. In this connection,
we saw, Bradley writes: "What (I ask next) should, on the
other hand, be meant by a relation viewed as absolutely
and merely internal? You, I presume, still in this case
would continue to take the terms each one as, so far,
in and by itself real, and as independent absolutely
of any whole that could be said to contain them."* A
quality as "ultimately and absolutely real" would be a
quality whose character or nature were quite self-con-
tained.

Any such entity would be what it is in virtue of itself

alone. This is to say that it would be itself absolutely, not

by virtue of any relations to anything ulterior to itself. If

we take two such entities, and notice that their being

* Collected Essays , p. 642.

179

Bradley 's Dialectic

"related" derives from and is exhausted by the respective
entities themselves, then we have merely internal relations.
"Relations would be merely internal if, the terms being
taken as real independently, each in itself, the relations
between them (as a class, or in this or that particular case)
in fact arose or were due merely to the character of the
terms as so taken."* This view of the matter is condemned
roundly by Bradley as a "ludicrous" if "natural" mistake.

The theory of relations as merely internal to or exhausted
in and by the terms that constitute a relation could only be'
mistaken, in Bradley's view, because it fails to see that a
relation, to be at all, must be at once internal and external
to its terms. A relation must be partially internal to its
terms in order that it may relate them: it must be partially
external to them in order that it may be anything at all in
its own right.

Consequently, any theory of relations as merely internal
is in stark opposition to the doctrine that is elucidated in
the chapter Relation and Quality. To be a quality at all is to
be distinct, and to be distinct is to be differentiated. This
differentiation or relation contributes to constitute what it
differentiates. So far, then, a relation is internal to its
qualities. But no relation is internal merely. Any relation
will to some extent fall between the qualities it differentiates.
In this respect a relation is the third moment in the un-
ending fission that is process. The third moment is the
middle term by which the fission is at once differentiated
and mediated. We have seen that the necessary (though
of course not sufficient) principle of this dialectic of quality
and relation is that the contradictory be identical with the
contrary. That principle yields negation and identity as
relational, or mediated by a third term.

We have noticed that for Hume the identity of simple
elements of experience is absolute. Any element of experience

* Ibid., p. 665.
180

Relational and Absolute Identity

either resembles or is different from any other, and a
difference is merely the "negation" of a resemblance. For
Hume a resemblance is any qualitative identity that is
distributed in at least two cases of itself. Thus a resemblance
is merely internal to or wholly exhausted in its terms: it is an
example of what Bradley calls a "merely internal" relation.
Thus we may see how it is that Hume and Bradley stand on
the opposite sides of the disjunction between absolute and
relational identity.

Since this point would seem to be easily accessible, it is
the more surprising that it is so frequently slurred over, or
missed altogether. As a recent example of this, consider the
following statements made by Professor H. H. Price. "I now
turn to the contention, that Hume errs by being an Atomist.
This is an even more extraordinary muddle, and I shall not
attempt to unravel it in detail. I will simply ask, what is
the positive alternative to Atomism? What do non-atomistic
philosophers assert? I suppose they assert that what we are
aware of is always a continuum of some sort, a continuous
stream of events, or of presentations, or what not."* Now
this is indeed an extraordinary muddle. One would have
thought that in Oxford it would be remembered that
"what (some) non-atomistic philsophers assert" is monism,
and a monism that derives from the internality of relations.
For those who identify the contradictory and the contrary
Hume's atomism is perforce an error, as is the logical
atomism of every other thinker for whom A is not both A
and not-A. The Bradleian dialectic condemns Hume's
atomism and all logically cognate doctrine as a sin against
the very principle of thinking. This condemnation may be
mistaken; the principle it would defend may entail difficul-
ties of a very discouraging nature; but it is anything but
muddled. Bradley knew what he was doing. And he carried
it through to the bitter end of Ultimate Doubts.
* Philosophy, Vol. XV, No. 57, p. 27.

181

Bradley's Dialectic

To see that absolute identity and relational identity are
wholly opposed is to see that neither Bradley nor Hume
is or could be a refutation of the other. For neither one of the
two terms of a disjunction is a refutation of the other one.
It is open to Bradley to demonstrate his first principle of
identity in difference if and only if the contradictory be
identical with the contrary. For then and only then will
differences be mediated by a third term.

The asserted identity of the contradictory with the
contrary can be known to be true if and only if the Law of
Non-Contradiction is known to be invalid. How is this
knowledge to be arrived at? Not by induction, surely. And
any attempt to arrive at it by deduction could only beg the
question. Since, according to the proponents of it, identity
in difference is the principle of all thought, perforce it would
be the principle of the extirpation of the Law of Non-
Contradiction. And it is so employed as the two-edged
discursus of a dialectic which seeks and claims to be self-
justifying. The Law of Non-Contradiction is held to be
invalid not because it is inconsistent, but because it stands
as an inane and sterile obstacle in the way of the fertility of
dialectic.

To this it must be replied that a method of elucidation
may be as fertile as anyone likes to deem it, or cares to make
it. This would recommend it to a man interested in elucida-
tion for its own sake. But it would not even tend to show the
Law of Non-Contradiction to be invalid. The principle of
identity in difference can be instituted only by assuming that
the laws of thought are invalid, not by an appeal to the
dialectic that denies them.

It has been pointed out above that absolute and relational
identity stand opposed as the terms of a disjunction. We may
be free to elect the one or the other; we may not adopt both
without confusion. There are those who suggest that any
choice between these two disjuncts will be more or less

182

Relational and Absolute Identity

unconscious, and dictated almost entirely by matters of
temperament. As some men are born little Platonists and
others little Aristotelians, so in some men there is a predilec-
tion for the relational dialectic of imagination, while others
are bound by logic.

It would seem to be fairly clear that no rational grounds
for a choice between these disjuncts can be demonstrated
a priori. The hackneyed point that you cannot contradict
the Law of Non-Contradiction without thereby reinstating
it is not free from confusion, and is of no avail at all against
Bradley's position. Bradley does not first affirm the Law of
Non-Contradiction and then turn around and deny it,
thus to fall into contradiction. He denounces what is to him
no law but a delusion, and proclaims as basic to elucidation
the Law of Contrariety. This excludes the Law of Excluded
Middle and affirms the reality of a middle term between
any two beings. The oft-repeated point in question assumes
(among other things) that any denial of the Law of Non-
Contradiction will be a contradiction. Yet this is hardly the
case. A denial may be the rejection of a proposal; not the
unsaying of something previously affirmed by the person
who rejects a proposal.

Since the logic of contradictories, as distinguished from a
dialectic of contraries, could hardly exclude the Law of
Non-Contradiction, any attempt to find by or in logic
grounds for a choice between absolute and relational
identity could only beg the question. For any such attempt,
consistently carried out, perforce would involve the Law of
Identity.

No more is it open to a follower of Bradley to avail himself
of the dialectic of contraries to prove the primacy of rela-
tional identity. We have noticed that Bradley does not
make the attempt. Rather, he denounces the laws of thought
as tautologies; upon the assumption that to be a tautology
is to be inane; and proceeds to identify the contradictory

183

Bradley's Dialectic

with the contrary. Yet this is done without benefit of either
logic or dialectic. Clearly logic could not sanction that
union. No more could the dialectic. For if and only if the
contradictory be identical with the contrary is the dialectic
of contraries valid. To appeal to that dialectic for a demon-
stration of the assumption that identity is relational could
only beg the question.

So much, it may be objected, is even obvious. And in
some quarters it will be urged that in the matter before us
the touchstone of rational decision is supplied by experience*
In the sentience that is perception or imagination the
incipient workings of the dialectic may be discerned. Once
this process has passed over into the stage of elucidation that
is judgement, the inherent logic of experience is made
explicit to some extent. This logic is (as the coherence
theory of degrees of reality and truth makes plain) self-
fulfilling and self-fulfilled. Anyone who understands it at all
will realize its justification to be internal and not a matter
of "linear" demonstration. For the binding constraint of the
relational way of thought derives from the coherence of it
with itself. And that self-coherence derives from the intern-
ality of relations which is the essence of process or experience
everywhere and always.

The appeal to experience, considered so hard-headed and
sensible by its proponents whether they be Idealists, Positiv-
ists, Neo-Realists, Pragmatists, or members of other philo-
sophical sects, is a hardy and widely variegated perennial.
Men have been appealing to experience for a long time. In
her name they have castigated their opponents; men no
less sensible than themselves, perhaps, who then replied in
the name of their experience. Thus Idealists pour scorn on
the "atomism" of views of experience not their own.
And those who seek to elucidate the content of their
perceptions in terms of (say) sensa declare Idealists to be
dim, thoughtless, or merely unintelligible.

184

Relational and Absolute Identity

Presumably it is fairly plain that to appeal to experience
is to appeal to it as it is understood in one way or another.
There is more than a little evidence that some Idealists
emphasize this point in the course of their criticism of their
precritical opponents. Those thinkers construed experience
wrongly because they were imbued with an epistemology
that is false. Now Hume's conception of experience, it may
well be agreed, is crudely mistaken. This has been made out
on a score of grounds; grounds that are internal to Hume's
assertions in that regard, and have no connection with "the
relational way of thought". Such criticism is one thing, and
exigetical; but to argue that the notion of impressions and
ideas, or any other view, is false because contrary to Hegelian
tenets, is another, and doctrinaire. Yet this is a practice to
which some proponents of the Idealist dialectic are prone.
They make the assumptions upon which the dialectic of the
Phenomenology can be elucidated, and then proceed to
establish those presumptions by showing that contrary views
are not in accord with them. If and only if you adopt the
relational view of thought may you correctly construe
experience. For that is the only way in which you may
construe experience in accordance with the relational way
of thought.

This question-begging procedure is not improved upon,
of course, by those who use it on the opposite side of the
disjunction between relational and absolute identity. The
assumption that experience consists of self-identical "atomic"
constituents, rather than "relational situations", is shot
through with presumptions. The main point is simply that
to appeal to a doctrine about experience for grounds upon
which to decide between logic and dialectic is perforce
futile. The validity of the Law of Non-Contradiction could
hardly be made out by any appeal to the conception of
experience that is developed in the Phenomenology. No more
could the dialectic be justified by any one or more of the

185

Bradley 5 s Dialectic

conceptions of experience which the relational way of thought
denies. In short, either you adduce an empiricism that is in
accord with logic, or that is in accord with the dialectic, as
the case may be, and so beg the question; or else you adduce
a contrary empiricism, and so perforce miss the point of
your attempted justification.

Any appeal to the given can only beg the question for
the other fellow. For it will assume that what for you is
given (in the innocuous but quite wholesome sense made
out by Professor Price) is likewise given for soijieone else*.
But in philosophy no one is obligated to decide questions
for the other fellow. And it is presumptuous or worse to
make the attempt. If he is capable of following your argu-
ments, he is competent to make up his own mind in taking
basic decisions.

It would seem that there is no rigorous means by which
logic or Hegelian dialectic can justify as distinguished
from explain itself to the other fellow. Each student of
philosophy can only ask himself about the relevance of the
one or the other to what he himself finds for examination
and elucidation. To do this is indeed to appeal to experience.
But it is to appeal to his own experience as being his own,
not as something that imposes a philosophical obligation on
his fellow men.

If a man finds in his experience any two respects that
are strictly the same, then he finds something that is incom-
patible with the conclusion of the dialectic of relation and
quality that identity implies qualitative difference. We have
seen above (chapter VIII) that if identity implies qualitative
difference, then every difference, however slight, is unique.
Thus no two experiences could be strictly the same in any
respect; for every experience would be unmatched in every
respect. It might be suggested that the difference between
this case of perceived middle C and that one is infinitesimal
in some rather ambiguous sense or other of that term. This

1 86

Relational and Absolute Identity

would be a singularly inappropriate suggestion in this con-
nection. For the differences that are relations are constituent
differences. They are actual differentiations in sentience, not
ideal distinctions. If identity implies qualitative difference,
then no two experiences can be strictly the same in any
respect.

Protagonists of the relational way of imagination are
constrained by doctrine to deny that any two experiences,
whether perceptual, imaginative, or intellectual, are in any
icspect tb~ same. This is to say that for these protagonists
every experience, however comprehensive or merely tenuous,
is quite unique. To be sure, they sometimes deal with this
stark consequence by saying that every experience is in part
unique. But then the parts that would not be unique or
unmatched would be matched by something else. And this
would fly in the face of the conclusion that identity implies
qualitative difference. The doctrine of relational identity
is categorical; it admits of no exceptions. That is why it can
only be rejected by anyone in whose consciousness there are
ever two or more experiences that are in any respect strictly
the same.

INDEX

ABSOLUTE, the, and our imperfect

knowledge of it, 73
as beyond Appearance, 1 i g

passim

as allegedly not supra-rela-
tional, 127-129
Bradley's conclusion about,

152
And, the enumerative vs. the

conjunctive, 19-21
" Appearance," meaning of, for

Bradley, 25, 26

Appearance and succession, 71, 72
as relational throughout, 15-
42, 119-122, 167-169

Bradley, can he answer Hume ?,

172-181
Broad, G. D., 162-164

Consistency, or self-coherence, the
standard of Reality, 63, 64

Contradiction, Law of, 22, 23
true meaning of, 117

Contradictory, the, and the Con-
trary, 23-25, 117, I74"i79

Contrariety and identity, 109-118

Contrary, the, 109 passim

Degrees of individuality, 1 55
of truth and reality, 157-160

Differentiations as relations, 28-31
relations, and judgement, 83
passim

Error, 94-104

and irrelevance, 96-99

and truth, 97-99
Ewing, A. C., 42-55, 165, 1 66, 169

Hegel, 22
Hume, 172-181

Ideal, the, and the " what " and

the " that," 81-87
Ideality and self-transcendence,

85-87

of the finite, 81-91
Identity and difference/ 22

implies difference, 67, 68,

147-150, 161
in difference, 1 09- 1 1 8
relational vs. absolute, 150,

151, 175-181

Immediacy as quality, 26-29
Individuality, degrees of, 155
Internality of relations, 28-31,
35-42

Judgement, 23, 83-93, 112, 115,
1 16, 1 18

Knowledge of the " this " and the
" mine," 73-79

Laird, Professor John, on internal
relations, 51, 52

Monism, Bradley's, 133

Negation, self - coherence, and

Reality, 63-65

Negation and the Absolute, 1 3 1
NonrContradiction, Law of, 182-

183
Numerical difference, 144-146,

Predication, judgement, and the

Absolute, 92, 93
Price, Professor H. H., 181
Psychical fact, 82

Qualities, primary and secondary,
i5 16

1 88

Index

" Quality," meaning of; for

Bradley, 26, 27

Quality and relation, 22 passim
as immediacy, 26-29

Reality and the Absolute, 68-75
the standard or criterion of,

63-65

and Monism, 70-73
as against pluralism, 65-70
" Relation," meaning of, for

Bradley, 26, 27

Relational way of thought, the,
"62, 81% ,
Relations, as independently real,

17-22

as differentiation, 28-31
and Quality, 22-40, 73-78
and the " this " and the

" mine," 73-79 I

as " merely internal," 170- !

172 I

Space and time, 56-61

Tautology and judgement, 22, 23
The " this " and the " mine," 73-

79
as affording an intimation of

the Absolute, 122-126
Thought, the relational way of,

62, 81-93

as the moment of differentia-
tion, 83

Time and space, 56-61
Truth and the Absolute, 104-107
and coherence, 104-108
and judgement, 105, 106
and Reality, 104-108

Ultimate complexes, an erroneous

view, 112-114
Unity in diversity, 1 1 i-i 18

"What," the, and the "that,"
81-91

Printed in Great Britain by
Billing and Sons Ltd., Guildford and Esher